diff options
| author | Anton Samokhvalov <pg83@yandex.ru> | 2022-02-10 16:45:15 +0300 |
|---|---|---|
| committer | Daniil Cherednik <dcherednik@yandex-team.ru> | 2022-02-10 16:45:15 +0300 |
| commit | 72cb13b4aff9bc9cf22e49251bc8fd143f82538f (patch) | |
| tree | da2c34829458c7d4e74bdfbdf85dff449e9e7fb8 /contrib/libs/pcre | |
| parent | 778e51ba091dc39e7b7fcab2b9cf4dbedfb6f2b5 (diff) | |
| download | ydb-72cb13b4aff9bc9cf22e49251bc8fd143f82538f.tar.gz | |
Restoring authorship annotation for Anton Samokhvalov <pg83@yandex.ru>. Commit 1 of 2.
Diffstat (limited to 'contrib/libs/pcre')
25 files changed, 13016 insertions, 13016 deletions
diff --git a/contrib/libs/pcre/pcre.h b/contrib/libs/pcre/pcre.h index 86e3956c21..133d828216 100644 --- a/contrib/libs/pcre/pcre.h +++ b/contrib/libs/pcre/pcre.h @@ -1,51 +1,51 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* This is the public header file for the PCRE library, to be #included by -applications that call the PCRE functions. - +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* This is the public header file for the PCRE library, to be #included by +applications that call the PCRE functions. + Copyright (c) 1997-2014 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - -#ifndef _PCRE_H -#define _PCRE_H - -/* The current PCRE version information. */ - + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + +#ifndef _PCRE_H +#define _PCRE_H + +/* The current PCRE version information. */ + #define PCRE_MAJOR 8 #define PCRE_MINOR 44 #define PCRE_PRERELEASE #define PCRE_DATE 2020-02-12 - + /* When an application links to a PCRE DLL in Windows, the symbols that are imported have to be identified as such. When building PCRE, the appropriate export setting is defined in pcre_internal.h, which includes this file. So we @@ -65,16 +65,16 @@ don't change existing definitions of PCRE_EXP_DECL and PCRECPP_EXP_DECL. */ # endif #endif -/* By default, we use the standard "extern" declarations. */ - -#ifndef PCRE_EXP_DECL +/* By default, we use the standard "extern" declarations. */ + +#ifndef PCRE_EXP_DECL # ifdef __cplusplus # define PCRE_EXP_DECL extern "C" # else # define PCRE_EXP_DECL extern # endif -#endif - +#endif + #ifdef __cplusplus # ifndef PCRECPP_EXP_DECL # define PCRECPP_EXP_DECL extern @@ -82,19 +82,19 @@ don't change existing definitions of PCRE_EXP_DECL and PCRECPP_EXP_DECL. */ # ifndef PCRECPP_EXP_DEFN # define PCRECPP_EXP_DEFN # endif -#endif - -/* Have to include stdlib.h in order to ensure that size_t is defined; -it is needed here for malloc. */ - -#include <stdlib.h> - -/* Allow for C++ users */ - -#ifdef __cplusplus -extern "C" { -#endif - +#endif + +/* Have to include stdlib.h in order to ensure that size_t is defined; +it is needed here for malloc. */ + +#include <stdlib.h> + +/* Allow for C++ users */ + +#ifdef __cplusplus +extern "C" { +#endif + /* Public options. Some are compile-time only, some are run-time only, and some are both. Most of the compile-time options are saved with the compiled regex so that they can be inspected during studying (and therefore JIT compiling). Note @@ -104,12 +104,12 @@ are now used, so in order to conserve them, option bits that were previously only recognized at matching time (i.e. by pcre_exec() or pcre_dfa_exec()) may also be used for compile-time options that affect only compiling and are not relevant for studying or JIT compiling. - + Some options for pcre_compile() change its behaviour but do not affect the behaviour of the execution functions. Other options are passed through to the execution functions and affect their behaviour, with or without affecting the behaviour of pcre_compile(). - + Options that can be passed to pcre_compile() are tagged Cx below, with these variants: @@ -170,8 +170,8 @@ with J. */ #define PCRE_NOTEMPTY_ATSTART 0x10000000 /* E D J */ #define PCRE_UCP 0x20000000 /* C3 */ -/* Exec-time and get/set-time error codes */ - +/* Exec-time and get/set-time error codes */ + #define PCRE_ERROR_NOMATCH (-1) #define PCRE_ERROR_NULL (-2) #define PCRE_ERROR_BADOPTION (-3) @@ -210,7 +210,7 @@ with J. */ #define PCRE_ERROR_JIT_BADOPTION (-31) #define PCRE_ERROR_BADLENGTH (-32) #define PCRE_ERROR_UNSET (-33) - + /* Specific error codes for UTF-8 validity checks */ #define PCRE_UTF8_ERR0 0 @@ -252,24 +252,24 @@ with J. */ #define PCRE_UTF32_ERR2 2 /* Unused (was non-character) */ #define PCRE_UTF32_ERR3 3 -/* Request types for pcre_fullinfo() */ - -#define PCRE_INFO_OPTIONS 0 -#define PCRE_INFO_SIZE 1 -#define PCRE_INFO_CAPTURECOUNT 2 -#define PCRE_INFO_BACKREFMAX 3 -#define PCRE_INFO_FIRSTBYTE 4 -#define PCRE_INFO_FIRSTCHAR 4 /* For backwards compatibility */ -#define PCRE_INFO_FIRSTTABLE 5 -#define PCRE_INFO_LASTLITERAL 6 -#define PCRE_INFO_NAMEENTRYSIZE 7 -#define PCRE_INFO_NAMECOUNT 8 -#define PCRE_INFO_NAMETABLE 9 -#define PCRE_INFO_STUDYSIZE 10 -#define PCRE_INFO_DEFAULT_TABLES 11 -#define PCRE_INFO_OKPARTIAL 12 -#define PCRE_INFO_JCHANGED 13 -#define PCRE_INFO_HASCRORLF 14 +/* Request types for pcre_fullinfo() */ + +#define PCRE_INFO_OPTIONS 0 +#define PCRE_INFO_SIZE 1 +#define PCRE_INFO_CAPTURECOUNT 2 +#define PCRE_INFO_BACKREFMAX 3 +#define PCRE_INFO_FIRSTBYTE 4 +#define PCRE_INFO_FIRSTCHAR 4 /* For backwards compatibility */ +#define PCRE_INFO_FIRSTTABLE 5 +#define PCRE_INFO_LASTLITERAL 6 +#define PCRE_INFO_NAMEENTRYSIZE 7 +#define PCRE_INFO_NAMECOUNT 8 +#define PCRE_INFO_NAMETABLE 9 +#define PCRE_INFO_STUDYSIZE 10 +#define PCRE_INFO_DEFAULT_TABLES 11 +#define PCRE_INFO_OKPARTIAL 12 +#define PCRE_INFO_JCHANGED 13 +#define PCRE_INFO_HASCRORLF 14 #define PCRE_INFO_MINLENGTH 15 #define PCRE_INFO_JIT 16 #define PCRE_INFO_JITSIZE 17 @@ -281,25 +281,25 @@ with J. */ #define PCRE_INFO_MATCHLIMIT 23 #define PCRE_INFO_RECURSIONLIMIT 24 #define PCRE_INFO_MATCH_EMPTY 25 - -/* Request types for pcre_config(). Do not re-arrange, in order to remain -compatible. */ - -#define PCRE_CONFIG_UTF8 0 -#define PCRE_CONFIG_NEWLINE 1 -#define PCRE_CONFIG_LINK_SIZE 2 -#define PCRE_CONFIG_POSIX_MALLOC_THRESHOLD 3 -#define PCRE_CONFIG_MATCH_LIMIT 4 -#define PCRE_CONFIG_STACKRECURSE 5 -#define PCRE_CONFIG_UNICODE_PROPERTIES 6 -#define PCRE_CONFIG_MATCH_LIMIT_RECURSION 7 -#define PCRE_CONFIG_BSR 8 + +/* Request types for pcre_config(). Do not re-arrange, in order to remain +compatible. */ + +#define PCRE_CONFIG_UTF8 0 +#define PCRE_CONFIG_NEWLINE 1 +#define PCRE_CONFIG_LINK_SIZE 2 +#define PCRE_CONFIG_POSIX_MALLOC_THRESHOLD 3 +#define PCRE_CONFIG_MATCH_LIMIT 4 +#define PCRE_CONFIG_STACKRECURSE 5 +#define PCRE_CONFIG_UNICODE_PROPERTIES 6 +#define PCRE_CONFIG_MATCH_LIMIT_RECURSION 7 +#define PCRE_CONFIG_BSR 8 #define PCRE_CONFIG_JIT 9 #define PCRE_CONFIG_UTF16 10 #define PCRE_CONFIG_JITTARGET 11 #define PCRE_CONFIG_UTF32 12 #define PCRE_CONFIG_PARENS_LIMIT 13 - + /* Request types for pcre_study(). Do not re-arrange, in order to remain compatible. */ @@ -309,21 +309,21 @@ compatible. */ #define PCRE_STUDY_EXTRA_NEEDED 0x0008 /* Bit flags for the pcre[16|32]_extra structure. Do not re-arrange or redefine -these bits, just add new ones on the end, in order to remain compatible. */ - -#define PCRE_EXTRA_STUDY_DATA 0x0001 -#define PCRE_EXTRA_MATCH_LIMIT 0x0002 -#define PCRE_EXTRA_CALLOUT_DATA 0x0004 -#define PCRE_EXTRA_TABLES 0x0008 -#define PCRE_EXTRA_MATCH_LIMIT_RECURSION 0x0010 +these bits, just add new ones on the end, in order to remain compatible. */ + +#define PCRE_EXTRA_STUDY_DATA 0x0001 +#define PCRE_EXTRA_MATCH_LIMIT 0x0002 +#define PCRE_EXTRA_CALLOUT_DATA 0x0004 +#define PCRE_EXTRA_TABLES 0x0008 +#define PCRE_EXTRA_MATCH_LIMIT_RECURSION 0x0010 #define PCRE_EXTRA_MARK 0x0020 #define PCRE_EXTRA_EXECUTABLE_JIT 0x0040 - -/* Types */ - + +/* Types */ + struct real_pcre8_or_16; /* declaration; the definition is private */ typedef struct real_pcre8_or_16 pcre; - + struct real_pcre8_or_16; /* declaration; the definition is private */ typedef struct real_pcre8_or_16 pcre16; @@ -361,29 +361,29 @@ pcre32 functions are not implemented. There is a check for this in pcre_internal #define PCRE_SPTR32 const PCRE_UCHAR32 * #endif -/* When PCRE is compiled as a C++ library, the subject pointer type can be -replaced with a custom type. For conventional use, the public interface is a -const char *. */ - -#ifndef PCRE_SPTR -#define PCRE_SPTR const char * -#endif - -/* The structure for passing additional data to pcre_exec(). This is defined in -such as way as to be extensible. Always add new fields at the end, in order to -remain compatible. */ - -typedef struct pcre_extra { - unsigned long int flags; /* Bits for which fields are set */ - void *study_data; /* Opaque data from pcre_study() */ - unsigned long int match_limit; /* Maximum number of calls to match() */ - void *callout_data; /* Data passed back in callouts */ - const unsigned char *tables; /* Pointer to character tables */ - unsigned long int match_limit_recursion; /* Max recursive calls to match() */ +/* When PCRE is compiled as a C++ library, the subject pointer type can be +replaced with a custom type. For conventional use, the public interface is a +const char *. */ + +#ifndef PCRE_SPTR +#define PCRE_SPTR const char * +#endif + +/* The structure for passing additional data to pcre_exec(). This is defined in +such as way as to be extensible. Always add new fields at the end, in order to +remain compatible. */ + +typedef struct pcre_extra { + unsigned long int flags; /* Bits for which fields are set */ + void *study_data; /* Opaque data from pcre_study() */ + unsigned long int match_limit; /* Maximum number of calls to match() */ + void *callout_data; /* Data passed back in callouts */ + const unsigned char *tables; /* Pointer to character tables */ + unsigned long int match_limit_recursion; /* Max recursive calls to match() */ unsigned char **mark; /* For passing back a mark pointer */ void *executable_jit; /* Contains a pointer to a compiled jit code */ -} pcre_extra; - +} pcre_extra; + /* Same structure as above, but with 16 bit char pointers. */ typedef struct pcre16_extra { @@ -410,31 +410,31 @@ typedef struct pcre32_extra { void *executable_jit; /* Contains a pointer to a compiled jit code */ } pcre32_extra; -/* The structure for passing out data via the pcre_callout_function. We use a -structure so that new fields can be added on the end in future versions, -without changing the API of the function, thereby allowing old clients to work -without modification. */ - -typedef struct pcre_callout_block { - int version; /* Identifies version of block */ - /* ------------------------ Version 0 ------------------------------- */ - int callout_number; /* Number compiled into pattern */ - int *offset_vector; /* The offset vector */ - PCRE_SPTR subject; /* The subject being matched */ - int subject_length; /* The length of the subject */ - int start_match; /* Offset to start of this match attempt */ - int current_position; /* Where we currently are in the subject */ - int capture_top; /* Max current capture */ - int capture_last; /* Most recently closed capture */ - void *callout_data; /* Data passed in with the call */ - /* ------------------- Added for Version 1 -------------------------- */ - int pattern_position; /* Offset to next item in the pattern */ - int next_item_length; /* Length of next item in the pattern */ +/* The structure for passing out data via the pcre_callout_function. We use a +structure so that new fields can be added on the end in future versions, +without changing the API of the function, thereby allowing old clients to work +without modification. */ + +typedef struct pcre_callout_block { + int version; /* Identifies version of block */ + /* ------------------------ Version 0 ------------------------------- */ + int callout_number; /* Number compiled into pattern */ + int *offset_vector; /* The offset vector */ + PCRE_SPTR subject; /* The subject being matched */ + int subject_length; /* The length of the subject */ + int start_match; /* Offset to start of this match attempt */ + int current_position; /* Where we currently are in the subject */ + int capture_top; /* Max current capture */ + int capture_last; /* Most recently closed capture */ + void *callout_data; /* Data passed in with the call */ + /* ------------------- Added for Version 1 -------------------------- */ + int pattern_position; /* Offset to next item in the pattern */ + int next_item_length; /* Length of next item in the pattern */ /* ------------------- Added for Version 2 -------------------------- */ const unsigned char *mark; /* Pointer to current mark or NULL */ - /* ------------------------------------------------------------------ */ -} pcre_callout_block; - + /* ------------------------------------------------------------------ */ +} pcre_callout_block; + /* Same structure as above, but with 16 bit char pointers. */ typedef struct pcre16_callout_block { @@ -479,18 +479,18 @@ typedef struct pcre32_callout_block { /* ------------------------------------------------------------------ */ } pcre32_callout_block; -/* Indirection for store get and free functions. These can be set to -alternative malloc/free functions if required. Special ones are used in the -non-recursive case for "frames". There is also an optional callout function -that is triggered by the (?) regex item. For Virtual Pascal, these definitions -have to take another form. */ - -#ifndef VPCOMPAT -PCRE_EXP_DECL void *(*pcre_malloc)(size_t); -PCRE_EXP_DECL void (*pcre_free)(void *); -PCRE_EXP_DECL void *(*pcre_stack_malloc)(size_t); -PCRE_EXP_DECL void (*pcre_stack_free)(void *); -PCRE_EXP_DECL int (*pcre_callout)(pcre_callout_block *); +/* Indirection for store get and free functions. These can be set to +alternative malloc/free functions if required. Special ones are used in the +non-recursive case for "frames". There is also an optional callout function +that is triggered by the (?) regex item. For Virtual Pascal, these definitions +have to take another form. */ + +#ifndef VPCOMPAT +PCRE_EXP_DECL void *(*pcre_malloc)(size_t); +PCRE_EXP_DECL void (*pcre_free)(void *); +PCRE_EXP_DECL void *(*pcre_stack_malloc)(size_t); +PCRE_EXP_DECL void (*pcre_stack_free)(void *); +PCRE_EXP_DECL int (*pcre_callout)(pcre_callout_block *); PCRE_EXP_DECL int (*pcre_stack_guard)(void); PCRE_EXP_DECL void *(*pcre16_malloc)(size_t); @@ -506,12 +506,12 @@ PCRE_EXP_DECL void *(*pcre32_stack_malloc)(size_t); PCRE_EXP_DECL void (*pcre32_stack_free)(void *); PCRE_EXP_DECL int (*pcre32_callout)(pcre32_callout_block *); PCRE_EXP_DECL int (*pcre32_stack_guard)(void); -#else /* VPCOMPAT */ -PCRE_EXP_DECL void *pcre_malloc(size_t); -PCRE_EXP_DECL void pcre_free(void *); -PCRE_EXP_DECL void *pcre_stack_malloc(size_t); -PCRE_EXP_DECL void pcre_stack_free(void *); -PCRE_EXP_DECL int pcre_callout(pcre_callout_block *); +#else /* VPCOMPAT */ +PCRE_EXP_DECL void *pcre_malloc(size_t); +PCRE_EXP_DECL void pcre_free(void *); +PCRE_EXP_DECL void *pcre_stack_malloc(size_t); +PCRE_EXP_DECL void pcre_stack_free(void *); +PCRE_EXP_DECL int pcre_callout(pcre_callout_block *); PCRE_EXP_DECL int pcre_stack_guard(void); PCRE_EXP_DECL void *pcre16_malloc(size_t); @@ -527,33 +527,33 @@ PCRE_EXP_DECL void *pcre32_stack_malloc(size_t); PCRE_EXP_DECL void pcre32_stack_free(void *); PCRE_EXP_DECL int pcre32_callout(pcre32_callout_block *); PCRE_EXP_DECL int pcre32_stack_guard(void); -#endif /* VPCOMPAT */ - +#endif /* VPCOMPAT */ + /* User defined callback which provides a stack just before the match starts. */ typedef pcre_jit_stack *(*pcre_jit_callback)(void *); typedef pcre16_jit_stack *(*pcre16_jit_callback)(void *); typedef pcre32_jit_stack *(*pcre32_jit_callback)(void *); -/* Exported PCRE functions */ - -PCRE_EXP_DECL pcre *pcre_compile(const char *, int, const char **, int *, - const unsigned char *); +/* Exported PCRE functions */ + +PCRE_EXP_DECL pcre *pcre_compile(const char *, int, const char **, int *, + const unsigned char *); PCRE_EXP_DECL pcre16 *pcre16_compile(PCRE_SPTR16, int, const char **, int *, const unsigned char *); PCRE_EXP_DECL pcre32 *pcre32_compile(PCRE_SPTR32, int, const char **, int *, const unsigned char *); -PCRE_EXP_DECL pcre *pcre_compile2(const char *, int, int *, const char **, - int *, const unsigned char *); +PCRE_EXP_DECL pcre *pcre_compile2(const char *, int, int *, const char **, + int *, const unsigned char *); PCRE_EXP_DECL pcre16 *pcre16_compile2(PCRE_SPTR16, int, int *, const char **, int *, const unsigned char *); PCRE_EXP_DECL pcre32 *pcre32_compile2(PCRE_SPTR32, int, int *, const char **, int *, const unsigned char *); -PCRE_EXP_DECL int pcre_config(int, void *); +PCRE_EXP_DECL int pcre_config(int, void *); PCRE_EXP_DECL int pcre16_config(int, void *); PCRE_EXP_DECL int pcre32_config(int, void *); -PCRE_EXP_DECL int pcre_copy_named_substring(const pcre *, const char *, - int *, int, const char *, char *, int); +PCRE_EXP_DECL int pcre_copy_named_substring(const pcre *, const char *, + int *, int, const char *, char *, int); PCRE_EXP_DECL int pcre16_copy_named_substring(const pcre16 *, PCRE_SPTR16, int *, int, PCRE_SPTR16, PCRE_UCHAR16 *, int); PCRE_EXP_DECL int pcre32_copy_named_substring(const pcre32 *, PCRE_SPTR32, @@ -564,14 +564,14 @@ PCRE_EXP_DECL int pcre16_copy_substring(PCRE_SPTR16, int *, int, int, PCRE_UCHAR16 *, int); PCRE_EXP_DECL int pcre32_copy_substring(PCRE_SPTR32, int *, int, int, PCRE_UCHAR32 *, int); -PCRE_EXP_DECL int pcre_dfa_exec(const pcre *, const pcre_extra *, - const char *, int, int, int, int *, int , int *, int); +PCRE_EXP_DECL int pcre_dfa_exec(const pcre *, const pcre_extra *, + const char *, int, int, int, int *, int , int *, int); PCRE_EXP_DECL int pcre16_dfa_exec(const pcre16 *, const pcre16_extra *, PCRE_SPTR16, int, int, int, int *, int , int *, int); PCRE_EXP_DECL int pcre32_dfa_exec(const pcre32 *, const pcre32_extra *, PCRE_SPTR32, int, int, int, int *, int , int *, int); -PCRE_EXP_DECL int pcre_exec(const pcre *, const pcre_extra *, PCRE_SPTR, - int, int, int, int *, int); +PCRE_EXP_DECL int pcre_exec(const pcre *, const pcre_extra *, PCRE_SPTR, + int, int, int, int *, int); PCRE_EXP_DECL int pcre16_exec(const pcre16 *, const pcre16_extra *, PCRE_SPTR16, int, int, int, int *, int); PCRE_EXP_DECL int pcre32_exec(const pcre32 *, const pcre32_extra *, @@ -585,61 +585,61 @@ PCRE_EXP_DECL int pcre16_jit_exec(const pcre16 *, const pcre16_extra *, PCRE_EXP_DECL int pcre32_jit_exec(const pcre32 *, const pcre32_extra *, PCRE_SPTR32, int, int, int, int *, int, pcre32_jit_stack *); -PCRE_EXP_DECL void pcre_free_substring(const char *); +PCRE_EXP_DECL void pcre_free_substring(const char *); PCRE_EXP_DECL void pcre16_free_substring(PCRE_SPTR16); PCRE_EXP_DECL void pcre32_free_substring(PCRE_SPTR32); -PCRE_EXP_DECL void pcre_free_substring_list(const char **); +PCRE_EXP_DECL void pcre_free_substring_list(const char **); PCRE_EXP_DECL void pcre16_free_substring_list(PCRE_SPTR16 *); PCRE_EXP_DECL void pcre32_free_substring_list(PCRE_SPTR32 *); -PCRE_EXP_DECL int pcre_fullinfo(const pcre *, const pcre_extra *, int, - void *); +PCRE_EXP_DECL int pcre_fullinfo(const pcre *, const pcre_extra *, int, + void *); PCRE_EXP_DECL int pcre16_fullinfo(const pcre16 *, const pcre16_extra *, int, void *); PCRE_EXP_DECL int pcre32_fullinfo(const pcre32 *, const pcre32_extra *, int, void *); -PCRE_EXP_DECL int pcre_get_named_substring(const pcre *, const char *, - int *, int, const char *, const char **); +PCRE_EXP_DECL int pcre_get_named_substring(const pcre *, const char *, + int *, int, const char *, const char **); PCRE_EXP_DECL int pcre16_get_named_substring(const pcre16 *, PCRE_SPTR16, int *, int, PCRE_SPTR16, PCRE_SPTR16 *); PCRE_EXP_DECL int pcre32_get_named_substring(const pcre32 *, PCRE_SPTR32, int *, int, PCRE_SPTR32, PCRE_SPTR32 *); -PCRE_EXP_DECL int pcre_get_stringnumber(const pcre *, const char *); +PCRE_EXP_DECL int pcre_get_stringnumber(const pcre *, const char *); PCRE_EXP_DECL int pcre16_get_stringnumber(const pcre16 *, PCRE_SPTR16); PCRE_EXP_DECL int pcre32_get_stringnumber(const pcre32 *, PCRE_SPTR32); -PCRE_EXP_DECL int pcre_get_stringtable_entries(const pcre *, const char *, - char **, char **); +PCRE_EXP_DECL int pcre_get_stringtable_entries(const pcre *, const char *, + char **, char **); PCRE_EXP_DECL int pcre16_get_stringtable_entries(const pcre16 *, PCRE_SPTR16, PCRE_UCHAR16 **, PCRE_UCHAR16 **); PCRE_EXP_DECL int pcre32_get_stringtable_entries(const pcre32 *, PCRE_SPTR32, PCRE_UCHAR32 **, PCRE_UCHAR32 **); -PCRE_EXP_DECL int pcre_get_substring(const char *, int *, int, int, - const char **); +PCRE_EXP_DECL int pcre_get_substring(const char *, int *, int, int, + const char **); PCRE_EXP_DECL int pcre16_get_substring(PCRE_SPTR16, int *, int, int, PCRE_SPTR16 *); PCRE_EXP_DECL int pcre32_get_substring(PCRE_SPTR32, int *, int, int, PCRE_SPTR32 *); -PCRE_EXP_DECL int pcre_get_substring_list(const char *, int *, int, - const char ***); +PCRE_EXP_DECL int pcre_get_substring_list(const char *, int *, int, + const char ***); PCRE_EXP_DECL int pcre16_get_substring_list(PCRE_SPTR16, int *, int, PCRE_SPTR16 **); PCRE_EXP_DECL int pcre32_get_substring_list(PCRE_SPTR32, int *, int, PCRE_SPTR32 **); -PCRE_EXP_DECL const unsigned char *pcre_maketables(void); +PCRE_EXP_DECL const unsigned char *pcre_maketables(void); PCRE_EXP_DECL const unsigned char *pcre16_maketables(void); PCRE_EXP_DECL const unsigned char *pcre32_maketables(void); -PCRE_EXP_DECL int pcre_refcount(pcre *, int); +PCRE_EXP_DECL int pcre_refcount(pcre *, int); PCRE_EXP_DECL int pcre16_refcount(pcre16 *, int); PCRE_EXP_DECL int pcre32_refcount(pcre32 *, int); -PCRE_EXP_DECL pcre_extra *pcre_study(const pcre *, int, const char **); +PCRE_EXP_DECL pcre_extra *pcre_study(const pcre *, int, const char **); PCRE_EXP_DECL pcre16_extra *pcre16_study(const pcre16 *, int, const char **); PCRE_EXP_DECL pcre32_extra *pcre32_study(const pcre32 *, int, const char **); PCRE_EXP_DECL void pcre_free_study(pcre_extra *); PCRE_EXP_DECL void pcre16_free_study(pcre16_extra *); PCRE_EXP_DECL void pcre32_free_study(pcre32_extra *); -PCRE_EXP_DECL const char *pcre_version(void); +PCRE_EXP_DECL const char *pcre_version(void); PCRE_EXP_DECL const char *pcre16_version(void); PCRE_EXP_DECL const char *pcre32_version(void); - + /* Utility functions for byte order swaps. */ PCRE_EXP_DECL int pcre_pattern_to_host_byte_order(pcre *, pcre_extra *, const unsigned char *); @@ -670,8 +670,8 @@ PCRE_EXP_DECL void pcre_jit_free_unused_memory(void); PCRE_EXP_DECL void pcre16_jit_free_unused_memory(void); PCRE_EXP_DECL void pcre32_jit_free_unused_memory(void); -#ifdef __cplusplus -} /* extern "C" */ -#endif - -#endif /* End of pcre.h */ +#ifdef __cplusplus +} /* extern "C" */ +#endif + +#endif /* End of pcre.h */ diff --git a/contrib/libs/pcre/pcre_chartables.c b/contrib/libs/pcre/pcre_chartables.c index f22172b835..87ecc05e91 100644 --- a/contrib/libs/pcre/pcre_chartables.c +++ b/contrib/libs/pcre/pcre_chartables.c @@ -1,198 +1,198 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* This file contains character tables that are used when no external tables -are passed to PCRE by the application that calls it. The tables are used only -for characters whose code values are less than 256. - -This is a default version of the tables that assumes ASCII encoding. A program -called dftables (which is distributed with PCRE) can be used to build -alternative versions of this file. This is necessary if you are running in an -EBCDIC environment, or if you want to default to a different encoding, for -example ISO-8859-1. When dftables is run, it creates these tables in the -current locale. If PCRE is configured with --enable-rebuild-chartables, this -happens automatically. - +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* This file contains character tables that are used when no external tables +are passed to PCRE by the application that calls it. The tables are used only +for characters whose code values are less than 256. + +This is a default version of the tables that assumes ASCII encoding. A program +called dftables (which is distributed with PCRE) can be used to build +alternative versions of this file. This is necessary if you are running in an +EBCDIC environment, or if you want to default to a different encoding, for +example ISO-8859-1. When dftables is run, it creates these tables in the +current locale. If PCRE is configured with --enable-rebuild-chartables, this +happens automatically. + The following #includes are present because without them gcc 4.x may remove the -array definition from the final binary if PCRE is built into a static library -and dead code stripping is activated. This leads to link errors. Pulling in the -header ensures that the array gets flagged as "someone outside this compilation -unit might reference this" and so it will always be supplied to the linker. */ - -#ifdef HAVE_CONFIG_H +array definition from the final binary if PCRE is built into a static library +and dead code stripping is activated. This leads to link errors. Pulling in the +header ensures that the array gets flagged as "someone outside this compilation +unit might reference this" and so it will always be supplied to the linker. */ + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - +#endif + +#include "pcre_internal.h" + const pcre_uint8 PRIV(default_tables)[] = { - -/* This table is a lower casing table. */ - - 0, 1, 2, 3, 4, 5, 6, 7, - 8, 9, 10, 11, 12, 13, 14, 15, - 16, 17, 18, 19, 20, 21, 22, 23, - 24, 25, 26, 27, 28, 29, 30, 31, - 32, 33, 34, 35, 36, 37, 38, 39, - 40, 41, 42, 43, 44, 45, 46, 47, - 48, 49, 50, 51, 52, 53, 54, 55, - 56, 57, 58, 59, 60, 61, 62, 63, - 64, 97, 98, 99,100,101,102,103, - 104,105,106,107,108,109,110,111, - 112,113,114,115,116,117,118,119, - 120,121,122, 91, 92, 93, 94, 95, - 96, 97, 98, 99,100,101,102,103, - 104,105,106,107,108,109,110,111, - 112,113,114,115,116,117,118,119, - 120,121,122,123,124,125,126,127, - 128,129,130,131,132,133,134,135, - 136,137,138,139,140,141,142,143, - 144,145,146,147,148,149,150,151, - 152,153,154,155,156,157,158,159, - 160,161,162,163,164,165,166,167, - 168,169,170,171,172,173,174,175, - 176,177,178,179,180,181,182,183, - 184,185,186,187,188,189,190,191, - 192,193,194,195,196,197,198,199, - 200,201,202,203,204,205,206,207, - 208,209,210,211,212,213,214,215, - 216,217,218,219,220,221,222,223, - 224,225,226,227,228,229,230,231, - 232,233,234,235,236,237,238,239, - 240,241,242,243,244,245,246,247, - 248,249,250,251,252,253,254,255, - -/* This table is a case flipping table. */ - - 0, 1, 2, 3, 4, 5, 6, 7, - 8, 9, 10, 11, 12, 13, 14, 15, - 16, 17, 18, 19, 20, 21, 22, 23, - 24, 25, 26, 27, 28, 29, 30, 31, - 32, 33, 34, 35, 36, 37, 38, 39, - 40, 41, 42, 43, 44, 45, 46, 47, - 48, 49, 50, 51, 52, 53, 54, 55, - 56, 57, 58, 59, 60, 61, 62, 63, - 64, 97, 98, 99,100,101,102,103, - 104,105,106,107,108,109,110,111, - 112,113,114,115,116,117,118,119, - 120,121,122, 91, 92, 93, 94, 95, - 96, 65, 66, 67, 68, 69, 70, 71, - 72, 73, 74, 75, 76, 77, 78, 79, - 80, 81, 82, 83, 84, 85, 86, 87, - 88, 89, 90,123,124,125,126,127, - 128,129,130,131,132,133,134,135, - 136,137,138,139,140,141,142,143, - 144,145,146,147,148,149,150,151, - 152,153,154,155,156,157,158,159, - 160,161,162,163,164,165,166,167, - 168,169,170,171,172,173,174,175, - 176,177,178,179,180,181,182,183, - 184,185,186,187,188,189,190,191, - 192,193,194,195,196,197,198,199, - 200,201,202,203,204,205,206,207, - 208,209,210,211,212,213,214,215, - 216,217,218,219,220,221,222,223, - 224,225,226,227,228,229,230,231, - 232,233,234,235,236,237,238,239, - 240,241,242,243,244,245,246,247, - 248,249,250,251,252,253,254,255, - -/* This table contains bit maps for various character classes. Each map is 32 -bytes long and the bits run from the least significant end of each byte. The -classes that have their own maps are: space, xdigit, digit, upper, lower, word, -graph, print, punct, and cntrl. Other classes are built from combinations. */ - - 0x00,0x3e,0x00,0x00,0x01,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - - 0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03, - 0x7e,0x00,0x00,0x00,0x7e,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - - 0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0xfe,0xff,0xff,0x07,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0x07, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - - 0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03, - 0xfe,0xff,0xff,0x87,0xfe,0xff,0xff,0x07, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - - 0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0xff, - 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - - 0x00,0x00,0x00,0x00,0xff,0xff,0xff,0xff, - 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - - 0x00,0x00,0x00,0x00,0xfe,0xff,0x00,0xfc, - 0x01,0x00,0x00,0xf8,0x01,0x00,0x00,0x78, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - - 0xff,0xff,0xff,0xff,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x80, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, - -/* This table identifies various classes of character by individual bits: - 0x01 white space character - 0x02 letter - 0x04 decimal digit - 0x08 hexadecimal digit - 0x10 alphanumeric or '_' - 0x80 regular expression metacharacter or binary zero -*/ - - 0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 */ + +/* This table is a lower casing table. */ + + 0, 1, 2, 3, 4, 5, 6, 7, + 8, 9, 10, 11, 12, 13, 14, 15, + 16, 17, 18, 19, 20, 21, 22, 23, + 24, 25, 26, 27, 28, 29, 30, 31, + 32, 33, 34, 35, 36, 37, 38, 39, + 40, 41, 42, 43, 44, 45, 46, 47, + 48, 49, 50, 51, 52, 53, 54, 55, + 56, 57, 58, 59, 60, 61, 62, 63, + 64, 97, 98, 99,100,101,102,103, + 104,105,106,107,108,109,110,111, + 112,113,114,115,116,117,118,119, + 120,121,122, 91, 92, 93, 94, 95, + 96, 97, 98, 99,100,101,102,103, + 104,105,106,107,108,109,110,111, + 112,113,114,115,116,117,118,119, + 120,121,122,123,124,125,126,127, + 128,129,130,131,132,133,134,135, + 136,137,138,139,140,141,142,143, + 144,145,146,147,148,149,150,151, + 152,153,154,155,156,157,158,159, + 160,161,162,163,164,165,166,167, + 168,169,170,171,172,173,174,175, + 176,177,178,179,180,181,182,183, + 184,185,186,187,188,189,190,191, + 192,193,194,195,196,197,198,199, + 200,201,202,203,204,205,206,207, + 208,209,210,211,212,213,214,215, + 216,217,218,219,220,221,222,223, + 224,225,226,227,228,229,230,231, + 232,233,234,235,236,237,238,239, + 240,241,242,243,244,245,246,247, + 248,249,250,251,252,253,254,255, + +/* This table is a case flipping table. */ + + 0, 1, 2, 3, 4, 5, 6, 7, + 8, 9, 10, 11, 12, 13, 14, 15, + 16, 17, 18, 19, 20, 21, 22, 23, + 24, 25, 26, 27, 28, 29, 30, 31, + 32, 33, 34, 35, 36, 37, 38, 39, + 40, 41, 42, 43, 44, 45, 46, 47, + 48, 49, 50, 51, 52, 53, 54, 55, + 56, 57, 58, 59, 60, 61, 62, 63, + 64, 97, 98, 99,100,101,102,103, + 104,105,106,107,108,109,110,111, + 112,113,114,115,116,117,118,119, + 120,121,122, 91, 92, 93, 94, 95, + 96, 65, 66, 67, 68, 69, 70, 71, + 72, 73, 74, 75, 76, 77, 78, 79, + 80, 81, 82, 83, 84, 85, 86, 87, + 88, 89, 90,123,124,125,126,127, + 128,129,130,131,132,133,134,135, + 136,137,138,139,140,141,142,143, + 144,145,146,147,148,149,150,151, + 152,153,154,155,156,157,158,159, + 160,161,162,163,164,165,166,167, + 168,169,170,171,172,173,174,175, + 176,177,178,179,180,181,182,183, + 184,185,186,187,188,189,190,191, + 192,193,194,195,196,197,198,199, + 200,201,202,203,204,205,206,207, + 208,209,210,211,212,213,214,215, + 216,217,218,219,220,221,222,223, + 224,225,226,227,228,229,230,231, + 232,233,234,235,236,237,238,239, + 240,241,242,243,244,245,246,247, + 248,249,250,251,252,253,254,255, + +/* This table contains bit maps for various character classes. Each map is 32 +bytes long and the bits run from the least significant end of each byte. The +classes that have their own maps are: space, xdigit, digit, upper, lower, word, +graph, print, punct, and cntrl. Other classes are built from combinations. */ + + 0x00,0x3e,0x00,0x00,0x01,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + + 0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03, + 0x7e,0x00,0x00,0x00,0x7e,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + + 0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0xfe,0xff,0xff,0x07,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0x07, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + + 0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03, + 0xfe,0xff,0xff,0x87,0xfe,0xff,0xff,0x07, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + + 0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0xff, + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + + 0x00,0x00,0x00,0x00,0xff,0xff,0xff,0xff, + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + + 0x00,0x00,0x00,0x00,0xfe,0xff,0x00,0xfc, + 0x01,0x00,0x00,0xf8,0x01,0x00,0x00,0x78, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + + 0xff,0xff,0xff,0xff,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x80, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, + +/* This table identifies various classes of character by individual bits: + 0x01 white space character + 0x02 letter + 0x04 decimal digit + 0x08 hexadecimal digit + 0x10 alphanumeric or '_' + 0x80 regular expression metacharacter or binary zero +*/ + + 0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 */ 0x00,0x01,0x01,0x01,0x01,0x01,0x00,0x00, /* 8- 15 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */ - 0x01,0x00,0x00,0x00,0x80,0x00,0x00,0x00, /* - ' */ - 0x80,0x80,0x80,0x80,0x00,0x00,0x80,0x00, /* ( - / */ - 0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /* 0 - 7 */ - 0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x80, /* 8 - ? */ - 0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* @ - G */ - 0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* H - O */ - 0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* P - W */ - 0x12,0x12,0x12,0x80,0x80,0x00,0x80,0x10, /* X - _ */ - 0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* ` - g */ - 0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* h - o */ - 0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* p - w */ - 0x12,0x12,0x12,0x80,0x80,0x00,0x00,0x00, /* x -127 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */ - -/* End of pcre_chartables.c */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */ + 0x01,0x00,0x00,0x00,0x80,0x00,0x00,0x00, /* - ' */ + 0x80,0x80,0x80,0x80,0x00,0x00,0x80,0x00, /* ( - / */ + 0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /* 0 - 7 */ + 0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x80, /* 8 - ? */ + 0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* @ - G */ + 0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* H - O */ + 0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* P - W */ + 0x12,0x12,0x12,0x80,0x80,0x00,0x80,0x10, /* X - _ */ + 0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* ` - g */ + 0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* h - o */ + 0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* p - w */ + 0x12,0x12,0x12,0x80,0x80,0x00,0x00,0x00, /* x -127 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */ + +/* End of pcre_chartables.c */ diff --git a/contrib/libs/pcre/pcre_compile.c b/contrib/libs/pcre/pcre_compile.c index 8051988093..4501a7e47b 100644 --- a/contrib/libs/pcre/pcre_compile.c +++ b/contrib/libs/pcre/pcre_compile.c @@ -1,84 +1,84 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2020 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains the external function pcre_compile(), along with -supporting internal functions that are not used by other modules. */ - - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains the external function pcre_compile(), along with +supporting internal functions that are not used by other modules. */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#define NLBLOCK cd /* Block containing newline information */ +#endif + +#define NLBLOCK cd /* Block containing newline information */ #define PSSTART start_pattern /* Field containing pattern start */ #define PSEND end_pattern /* Field containing pattern end */ - -#include "pcre_internal.h" - - + +#include "pcre_internal.h" + + /* When PCRE_DEBUG is defined, we need the pcre(16|32)_printint() function, which is also used by pcretest. PCRE_DEBUG is not defined when building a production library. We do not need to select pcre16_printint.c specially, because the COMPILE_PCREx macro will already be appropriately set. */ - + #ifdef PCRE_DEBUG /* pcre_printint.c should not include any headers */ #define PCRE_INCLUDED #include "pcre_printint.c" #undef PCRE_INCLUDED -#endif - - -/* Macro for setting individual bits in class bitmaps. */ - +#endif + + +/* Macro for setting individual bits in class bitmaps. */ + #define SETBIT(a,b) a[(b)/8] |= (1U << ((b)&7)) - -/* Maximum length value to check against when making sure that the integer that -holds the compiled pattern length does not overflow. We make it a bit less than -INT_MAX to allow for adding in group terminating bytes, so that we don't have -to check them every time. */ - -#define OFLOW_MAX (INT_MAX - 20) - + +/* Maximum length value to check against when making sure that the integer that +holds the compiled pattern length does not overflow. We make it a bit less than +INT_MAX to allow for adding in group terminating bytes, so that we don't have +to check them every time. */ + +#define OFLOW_MAX (INT_MAX - 20) + /* Definitions to allow mutual recursion */ - + static int add_list_to_class(pcre_uint8 *, pcre_uchar **, int, compile_data *, const pcre_uint32 *, unsigned int); @@ -90,20 +90,20 @@ static BOOL -/************************************************* -* Code parameters and static tables * -*************************************************/ - -/* This value specifies the size of stack workspace that is used during the -first pre-compile phase that determines how much memory is required. The regex -is partly compiled into this space, but the compiled parts are discarded as -soon as they can be, so that hopefully there will never be an overrun. The code -does, however, check for an overrun. The largest amount I've seen used is 218, -so this number is very generous. - -The same workspace is used during the second, actual compile phase for -remembering forward references to groups so that they can be filled in at the -end. Each entry in this list occupies LINK_SIZE bytes, so even when LINK_SIZE +/************************************************* +* Code parameters and static tables * +*************************************************/ + +/* This value specifies the size of stack workspace that is used during the +first pre-compile phase that determines how much memory is required. The regex +is partly compiled into this space, but the compiled parts are discarded as +soon as they can be, so that hopefully there will never be an overrun. The code +does, however, check for an overrun. The largest amount I've seen used is 218, +so this number is very generous. + +The same workspace is used during the second, actual compile phase for +remembering forward references to groups so that they can be filled in at the +end. Each entry in this list occupies LINK_SIZE bytes, so even when LINK_SIZE is 4 there is plenty of room for most patterns. However, the memory can get filled up by repetitions of forward references, for example patterns like /(?1){0,1999}(b)/, and one user did hit the limit. The code has been changed so @@ -111,15 +111,15 @@ that the workspace is expanded using malloc() in this situation. The value below is therefore a minimum, and we put a maximum on it for safety. The minimum is now also defined in terms of LINK_SIZE so that the use of malloc() kicks in at the same number of forward references in all cases. */ - + #define COMPILE_WORK_SIZE (2048*LINK_SIZE) #define COMPILE_WORK_SIZE_MAX (100*COMPILE_WORK_SIZE) - + /* This value determines the size of the initial vector that is used for remembering named groups during the pre-compile. It is allocated on the stack, but if it is too small, it is expanded using malloc(), in a similar way to the workspace. The value is the number of slots in the list. */ - + #define NAMED_GROUP_LIST_SIZE 20 /* The overrun tests check for a slightly smaller size so that they detect the @@ -139,17 +139,17 @@ overrun before it actually does run off the end of the data block. */ #define UTF_LENGTH 0x10000000l /* The char contains its length. */ -/* Table for handling escaped characters in the range '0'-'z'. Positive returns -are simple data values; negative values are for special things like \d and so -on. Zero means further processing is needed (for things like \x), or the escape -is invalid. */ - +/* Table for handling escaped characters in the range '0'-'z'. Positive returns +are simple data values; negative values are for special things like \d and so +on. Zero means further processing is needed (for things like \x), or the escape +is invalid. */ + #ifndef EBCDIC /* This is the "normal" table for ASCII systems or for EBCDIC systems running in UTF-8 mode. */ -static const short int escapes[] = { +static const short int escapes[] = { 0, 0, 0, 0, 0, 0, @@ -188,59 +188,59 @@ static const short int escapes[] = { -ESC_v, -ESC_w, 0, 0, -ESC_z -}; - +}; + #else /* This is the "abnormal" table for EBCDIC systems without UTF-8 support. */ -static const short int escapes[] = { -/* 48 */ 0, 0, 0, '.', '<', '(', '+', '|', -/* 50 */ '&', 0, 0, 0, 0, 0, 0, 0, -/* 58 */ 0, 0, '!', '$', '*', ')', ';', '~', -/* 60 */ '-', '/', 0, 0, 0, 0, 0, 0, -/* 68 */ 0, 0, '|', ',', '%', '_', '>', '?', -/* 70 */ 0, 0, 0, 0, 0, 0, 0, 0, -/* 78 */ 0, '`', ':', '#', '@', '\'', '=', '"', +static const short int escapes[] = { +/* 48 */ 0, 0, 0, '.', '<', '(', '+', '|', +/* 50 */ '&', 0, 0, 0, 0, 0, 0, 0, +/* 58 */ 0, 0, '!', '$', '*', ')', ';', '~', +/* 60 */ '-', '/', 0, 0, 0, 0, 0, 0, +/* 68 */ 0, 0, '|', ',', '%', '_', '>', '?', +/* 70 */ 0, 0, 0, 0, 0, 0, 0, 0, +/* 78 */ 0, '`', ':', '#', '@', '\'', '=', '"', /* 80 */ 0, ESC_a, -ESC_b, 0, -ESC_d, ESC_e, ESC_f, 0, -/* 88 */-ESC_h, 0, 0, '{', 0, 0, 0, 0, +/* 88 */-ESC_h, 0, 0, '{', 0, 0, 0, 0, /* 90 */ 0, 0, -ESC_k, 0, 0, ESC_n, 0, -ESC_p, -/* 98 */ 0, ESC_r, 0, '}', 0, 0, 0, 0, -/* A0 */ 0, '~', -ESC_s, ESC_tee, 0,-ESC_v, -ESC_w, 0, -/* A8 */ 0,-ESC_z, 0, 0, 0, '[', 0, 0, -/* B0 */ 0, 0, 0, 0, 0, 0, 0, 0, -/* B8 */ 0, 0, 0, 0, 0, ']', '=', '-', -/* C0 */ '{',-ESC_A, -ESC_B, -ESC_C, -ESC_D,-ESC_E, 0, -ESC_G, -/* C8 */-ESC_H, 0, 0, 0, 0, 0, 0, 0, +/* 98 */ 0, ESC_r, 0, '}', 0, 0, 0, 0, +/* A0 */ 0, '~', -ESC_s, ESC_tee, 0,-ESC_v, -ESC_w, 0, +/* A8 */ 0,-ESC_z, 0, 0, 0, '[', 0, 0, +/* B0 */ 0, 0, 0, 0, 0, 0, 0, 0, +/* B8 */ 0, 0, 0, 0, 0, ']', '=', '-', +/* C0 */ '{',-ESC_A, -ESC_B, -ESC_C, -ESC_D,-ESC_E, 0, -ESC_G, +/* C8 */-ESC_H, 0, 0, 0, 0, 0, 0, 0, /* D0 */ '}', 0, -ESC_K, 0, 0,-ESC_N, 0, -ESC_P, -/* D8 */-ESC_Q,-ESC_R, 0, 0, 0, 0, 0, 0, -/* E0 */ '\\', 0, -ESC_S, 0, 0,-ESC_V, -ESC_W, -ESC_X, -/* E8 */ 0,-ESC_Z, 0, 0, 0, 0, 0, 0, -/* F0 */ 0, 0, 0, 0, 0, 0, 0, 0, -/* F8 */ 0, 0, 0, 0, 0, 0, 0, 0 -}; +/* D8 */-ESC_Q,-ESC_R, 0, 0, 0, 0, 0, 0, +/* E0 */ '\\', 0, -ESC_S, 0, 0,-ESC_V, -ESC_W, -ESC_X, +/* E8 */ 0,-ESC_Z, 0, 0, 0, 0, 0, 0, +/* F0 */ 0, 0, 0, 0, 0, 0, 0, 0, +/* F8 */ 0, 0, 0, 0, 0, 0, 0, 0 +}; /* We also need a table of characters that may follow \c in an EBCDIC environment for characters 0-31. */ static unsigned char ebcdic_escape_c[] = "@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_"; -#endif - - -/* Table of special "verbs" like (*PRUNE). This is a short table, so it is -searched linearly. Put all the names into a single string, in order to reduce +#endif + + +/* Table of special "verbs" like (*PRUNE). This is a short table, so it is +searched linearly. Put all the names into a single string, in order to reduce the number of relocations when a shared library is dynamically linked. The string is built from string macros so that it works in UTF-8 mode on EBCDIC platforms. */ - -typedef struct verbitem { + +typedef struct verbitem { int len; /* Length of verb name */ int op; /* Op when no arg, or -1 if arg mandatory */ int op_arg; /* Op when arg present, or -1 if not allowed */ -} verbitem; - -static const char verbnames[] = +} verbitem; + +static const char verbnames[] = "\0" /* Empty name is a shorthand for MARK */ STRING_MARK0 STRING_ACCEPT0 @@ -250,7 +250,7 @@ static const char verbnames[] = STRING_PRUNE0 STRING_SKIP0 STRING_THEN; - + static const verbitem verbs[] = { { 0, -1, OP_MARK }, { 4, -1, OP_MARK }, @@ -261,11 +261,11 @@ static const verbitem verbs[] = { { 5, OP_PRUNE, OP_PRUNE_ARG }, { 4, OP_SKIP, OP_SKIP_ARG }, { 4, OP_THEN, OP_THEN_ARG } -}; - +}; + static const int verbcount = sizeof(verbs)/sizeof(verbitem); - - + + /* Substitutes for [[:<:]] and [[:>:]], which mean start and end of word in another regex library. */ @@ -279,57 +279,57 @@ static const pcre_uchar sub_end_of_word[] = { CHAR_RIGHT_PARENTHESIS, '\0' }; -/* Tables of names of POSIX character classes and their lengths. The names are -now all in a single string, to reduce the number of relocations when a shared -library is dynamically loaded. The list of lengths is terminated by a zero -length entry. The first three must be alpha, lower, upper, as this is assumed +/* Tables of names of POSIX character classes and their lengths. The names are +now all in a single string, to reduce the number of relocations when a shared +library is dynamically loaded. The list of lengths is terminated by a zero +length entry. The first three must be alpha, lower, upper, as this is assumed for handling case independence. The indices for graph, print, and punct are needed, so identify them. */ - -static const char posix_names[] = + +static const char posix_names[] = STRING_alpha0 STRING_lower0 STRING_upper0 STRING_alnum0 STRING_ascii0 STRING_blank0 STRING_cntrl0 STRING_digit0 STRING_graph0 STRING_print0 STRING_punct0 STRING_space0 STRING_word0 STRING_xdigit; - + static const pcre_uint8 posix_name_lengths[] = { - 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 4, 6, 0 }; - + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 4, 6, 0 }; + #define PC_GRAPH 8 #define PC_PRINT 9 #define PC_PUNCT 10 -/* Table of class bit maps for each POSIX class. Each class is formed from a -base map, with an optional addition or removal of another map. Then, for some -classes, there is some additional tweaking: for [:blank:] the vertical space -characters are removed, and for [:alpha:] and [:alnum:] the underscore -character is removed. The triples in the table consist of the base map offset, -second map offset or -1 if no second map, and a non-negative value for map -addition or a negative value for map subtraction (if there are two maps). The -absolute value of the third field has these meanings: 0 => no tweaking, 1 => -remove vertical space characters, 2 => remove underscore. */ - -static const int posix_class_maps[] = { - cbit_word, cbit_digit, -2, /* alpha */ - cbit_lower, -1, 0, /* lower */ - cbit_upper, -1, 0, /* upper */ - cbit_word, -1, 2, /* alnum - word without underscore */ - cbit_print, cbit_cntrl, 0, /* ascii */ - cbit_space, -1, 1, /* blank - a GNU extension */ - cbit_cntrl, -1, 0, /* cntrl */ - cbit_digit, -1, 0, /* digit */ - cbit_graph, -1, 0, /* graph */ - cbit_print, -1, 0, /* print */ - cbit_punct, -1, 0, /* punct */ - cbit_space, -1, 0, /* space */ - cbit_word, -1, 0, /* word - a Perl extension */ - cbit_xdigit,-1, 0 /* xdigit */ -}; - +/* Table of class bit maps for each POSIX class. Each class is formed from a +base map, with an optional addition or removal of another map. Then, for some +classes, there is some additional tweaking: for [:blank:] the vertical space +characters are removed, and for [:alpha:] and [:alnum:] the underscore +character is removed. The triples in the table consist of the base map offset, +second map offset or -1 if no second map, and a non-negative value for map +addition or a negative value for map subtraction (if there are two maps). The +absolute value of the third field has these meanings: 0 => no tweaking, 1 => +remove vertical space characters, 2 => remove underscore. */ + +static const int posix_class_maps[] = { + cbit_word, cbit_digit, -2, /* alpha */ + cbit_lower, -1, 0, /* lower */ + cbit_upper, -1, 0, /* upper */ + cbit_word, -1, 2, /* alnum - word without underscore */ + cbit_print, cbit_cntrl, 0, /* ascii */ + cbit_space, -1, 1, /* blank - a GNU extension */ + cbit_cntrl, -1, 0, /* cntrl */ + cbit_digit, -1, 0, /* digit */ + cbit_graph, -1, 0, /* graph */ + cbit_print, -1, 0, /* print */ + cbit_punct, -1, 0, /* punct */ + cbit_space, -1, 0, /* space */ + cbit_word, -1, 0, /* word - a Perl extension */ + cbit_xdigit,-1, 0 /* xdigit */ +}; + /* Table of substitutes for \d etc when PCRE_UCP is set. They are replaced by Unicode property escapes. */ - + #ifdef SUPPORT_UCP static const pcre_uchar string_PNd[] = { CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET, @@ -434,99 +434,99 @@ static const pcre_uchar *posix_substitutes[] = { #define POSIX_SUBSIZE (sizeof(posix_substitutes) / sizeof(pcre_uchar *)) #endif -#define STRING(a) # a -#define XSTRING(s) STRING(s) - -/* The texts of compile-time error messages. These are "char *" because they -are passed to the outside world. Do not ever re-use any error number, because -they are documented. Always add a new error instead. Messages marked DEAD below -are no longer used. This used to be a table of strings, but in order to reduce -the number of relocations needed when a shared library is loaded dynamically, -it is now one long string. We cannot use a table of offsets, because the -lengths of inserts such as XSTRING(MAX_NAME_SIZE) are not known. Instead, we -simply count through to the one we want - this isn't a performance issue +#define STRING(a) # a +#define XSTRING(s) STRING(s) + +/* The texts of compile-time error messages. These are "char *" because they +are passed to the outside world. Do not ever re-use any error number, because +they are documented. Always add a new error instead. Messages marked DEAD below +are no longer used. This used to be a table of strings, but in order to reduce +the number of relocations needed when a shared library is loaded dynamically, +it is now one long string. We cannot use a table of offsets, because the +lengths of inserts such as XSTRING(MAX_NAME_SIZE) are not known. Instead, we +simply count through to the one we want - this isn't a performance issue because these strings are used only when there is a compilation error. - + Each substring ends with \0 to insert a null character. This includes the final substring, so that the whole string ends with \0\0, which can be detected when counting through. */ -static const char error_texts[] = - "no error\0" - "\\ at end of pattern\0" - "\\c at end of pattern\0" - "unrecognized character follows \\\0" - "numbers out of order in {} quantifier\0" - /* 5 */ - "number too big in {} quantifier\0" - "missing terminating ] for character class\0" - "invalid escape sequence in character class\0" - "range out of order in character class\0" - "nothing to repeat\0" - /* 10 */ +static const char error_texts[] = + "no error\0" + "\\ at end of pattern\0" + "\\c at end of pattern\0" + "unrecognized character follows \\\0" + "numbers out of order in {} quantifier\0" + /* 5 */ + "number too big in {} quantifier\0" + "missing terminating ] for character class\0" + "invalid escape sequence in character class\0" + "range out of order in character class\0" + "nothing to repeat\0" + /* 10 */ "internal error: invalid forward reference offset\0" - "internal error: unexpected repeat\0" - "unrecognized character after (? or (?-\0" - "POSIX named classes are supported only within a class\0" - "missing )\0" - /* 15 */ - "reference to non-existent subpattern\0" - "erroffset passed as NULL\0" - "unknown option bit(s) set\0" - "missing ) after comment\0" - "parentheses nested too deeply\0" /** DEAD **/ - /* 20 */ - "regular expression is too large\0" - "failed to get memory\0" - "unmatched parentheses\0" - "internal error: code overflow\0" - "unrecognized character after (?<\0" - /* 25 */ - "lookbehind assertion is not fixed length\0" - "malformed number or name after (?(\0" - "conditional group contains more than two branches\0" + "internal error: unexpected repeat\0" + "unrecognized character after (? or (?-\0" + "POSIX named classes are supported only within a class\0" + "missing )\0" + /* 15 */ + "reference to non-existent subpattern\0" + "erroffset passed as NULL\0" + "unknown option bit(s) set\0" + "missing ) after comment\0" + "parentheses nested too deeply\0" /** DEAD **/ + /* 20 */ + "regular expression is too large\0" + "failed to get memory\0" + "unmatched parentheses\0" + "internal error: code overflow\0" + "unrecognized character after (?<\0" + /* 25 */ + "lookbehind assertion is not fixed length\0" + "malformed number or name after (?(\0" + "conditional group contains more than two branches\0" "assertion expected after (?( or (?(?C)\0" - "(?R or (?[+-]digits must be followed by )\0" - /* 30 */ - "unknown POSIX class name\0" - "POSIX collating elements are not supported\0" + "(?R or (?[+-]digits must be followed by )\0" + /* 30 */ + "unknown POSIX class name\0" + "POSIX collating elements are not supported\0" "this version of PCRE is compiled without UTF support\0" - "spare error\0" /** DEAD **/ + "spare error\0" /** DEAD **/ "character value in \\x{} or \\o{} is too large\0" - /* 35 */ - "invalid condition (?(0)\0" - "\\C not allowed in lookbehind assertion\0" + /* 35 */ + "invalid condition (?(0)\0" + "\\C not allowed in lookbehind assertion\0" "PCRE does not support \\L, \\l, \\N{name}, \\U, or \\u\0" - "number after (?C is > 255\0" - "closing ) for (?C expected\0" - /* 40 */ - "recursive call could loop indefinitely\0" - "unrecognized character after (?P\0" - "syntax error in subpattern name (missing terminator)\0" - "two named subpatterns have the same name\0" - "invalid UTF-8 string\0" - /* 45 */ - "support for \\P, \\p, and \\X has not been compiled\0" - "malformed \\P or \\p sequence\0" - "unknown property name after \\P or \\p\0" - "subpattern name is too long (maximum " XSTRING(MAX_NAME_SIZE) " characters)\0" - "too many named subpatterns (maximum " XSTRING(MAX_NAME_COUNT) ")\0" - /* 50 */ - "repeated subpattern is too long\0" /** DEAD **/ + "number after (?C is > 255\0" + "closing ) for (?C expected\0" + /* 40 */ + "recursive call could loop indefinitely\0" + "unrecognized character after (?P\0" + "syntax error in subpattern name (missing terminator)\0" + "two named subpatterns have the same name\0" + "invalid UTF-8 string\0" + /* 45 */ + "support for \\P, \\p, and \\X has not been compiled\0" + "malformed \\P or \\p sequence\0" + "unknown property name after \\P or \\p\0" + "subpattern name is too long (maximum " XSTRING(MAX_NAME_SIZE) " characters)\0" + "too many named subpatterns (maximum " XSTRING(MAX_NAME_COUNT) ")\0" + /* 50 */ + "repeated subpattern is too long\0" /** DEAD **/ "octal value is greater than \\377 in 8-bit non-UTF-8 mode\0" - "internal error: overran compiling workspace\0" - "internal error: previously-checked referenced subpattern not found\0" - "DEFINE group contains more than one branch\0" - /* 55 */ + "internal error: overran compiling workspace\0" + "internal error: previously-checked referenced subpattern not found\0" + "DEFINE group contains more than one branch\0" + /* 55 */ "repeating a DEFINE group is not allowed\0" /** DEAD **/ - "inconsistent NEWLINE options\0" + "inconsistent NEWLINE options\0" "\\g is not followed by a braced, angle-bracketed, or quoted name/number or by a plain number\0" "a numbered reference must not be zero\0" "an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)\0" - /* 60 */ + /* 60 */ "(*VERB) not recognized or malformed\0" - "number is too big\0" - "subpattern name expected\0" + "number is too big\0" + "subpattern name expected\0" "digit expected after (?+\0" "] is an invalid data character in JavaScript compatibility mode\0" /* 65 */ @@ -562,23 +562,23 @@ static const char error_texts[] = "digits missing in \\x{} or \\o{}\0" "regular expression is too complicated\0" ; - -/* Table to identify digits and hex digits. This is used when compiling -patterns. Note that the tables in chartables are dependent on the locale, and -may mark arbitrary characters as digits - but the PCRE compiling code expects -to handle only 0-9, a-z, and A-Z as digits when compiling. That is why we have -a private table here. It costs 256 bytes, but it is a lot faster than doing -character value tests (at least in some simple cases I timed), and in some -applications one wants PCRE to compile efficiently as well as match -efficiently. - -For convenience, we use the same bit definitions as in chartables: - - 0x04 decimal digit - 0x08 hexadecimal digit - -Then we can use ctype_digit and ctype_xdigit in the code. */ - + +/* Table to identify digits and hex digits. This is used when compiling +patterns. Note that the tables in chartables are dependent on the locale, and +may mark arbitrary characters as digits - but the PCRE compiling code expects +to handle only 0-9, a-z, and A-Z as digits when compiling. That is why we have +a private table here. It costs 256 bytes, but it is a lot faster than doing +character value tests (at least in some simple cases I timed), and in some +applications one wants PCRE to compile efficiently as well as match +efficiently. + +For convenience, we use the same bit definitions as in chartables: + + 0x04 decimal digit + 0x08 hexadecimal digit + +Then we can use ctype_digit and ctype_xdigit in the code. */ + /* Using a simple comparison for decimal numbers rather than a memory read is much faster, and the resulting code is simpler (the compiler turns it into a subtraction and unsigned comparison). */ @@ -591,131 +591,131 @@ into a subtraction and unsigned comparison). */ UTF-8 mode. */ static const pcre_uint8 digitab[] = - { - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 8- 15 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - ' */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ( - / */ - 0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c, /* 0 - 7 */ - 0x0c,0x0c,0x00,0x00,0x00,0x00,0x00,0x00, /* 8 - ? */ - 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* @ - G */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* H - O */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* P - W */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* X - _ */ - 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* ` - g */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* h - o */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* p - w */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* x -127 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */ - + { + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 8- 15 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - ' */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ( - / */ + 0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c, /* 0 - 7 */ + 0x0c,0x0c,0x00,0x00,0x00,0x00,0x00,0x00, /* 8 - ? */ + 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* @ - G */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* H - O */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* P - W */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* X - _ */ + 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* ` - g */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* h - o */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* p - w */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* x -127 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */ + #else /* This is the "abnormal" case, for EBCDIC systems not running in UTF-8 mode. */ static const pcre_uint8 digitab[] = - { - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 0 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 8- 15 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 10 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 32- 39 20 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 40- 47 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 48- 55 30 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 56- 63 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - 71 40 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 72- | */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* & - 87 50 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 88- 95 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - -103 60 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 104- ? */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 112-119 70 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 120- " */ - 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* 128- g 80 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* h -143 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144- p 90 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* q -159 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160- x A0 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* y -175 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ^ -183 B0 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */ - 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* { - G C0 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* H -207 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* } - P D0 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* Q -223 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* \ - X E0 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* Y -239 */ - 0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c, /* 0 - 7 F0 */ - 0x0c,0x0c,0x00,0x00,0x00,0x00,0x00,0x00};/* 8 -255 */ - + { + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 0 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 8- 15 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 10 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 32- 39 20 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 40- 47 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 48- 55 30 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 56- 63 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - 71 40 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 72- | */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* & - 87 50 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 88- 95 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - -103 60 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 104- ? */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 112-119 70 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 120- " */ + 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* 128- g 80 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* h -143 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144- p 90 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* q -159 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160- x A0 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* y -175 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ^ -183 B0 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */ + 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* { - G C0 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* H -207 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* } - P D0 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* Q -223 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* \ - X E0 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* Y -239 */ + 0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c, /* 0 - 7 F0 */ + 0x0c,0x0c,0x00,0x00,0x00,0x00,0x00,0x00};/* 8 -255 */ + static const pcre_uint8 ebcdic_chartab[] = { /* chartable partial dup */ - 0x80,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 0- 7 */ - 0x00,0x00,0x00,0x00,0x01,0x01,0x00,0x00, /* 8- 15 */ - 0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 16- 23 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */ - 0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 32- 39 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 40- 47 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 48- 55 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 56- 63 */ - 0x01,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - 71 */ - 0x00,0x00,0x00,0x80,0x00,0x80,0x80,0x80, /* 72- | */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* & - 87 */ - 0x00,0x00,0x00,0x80,0x80,0x80,0x00,0x00, /* 88- 95 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - -103 */ - 0x00,0x00,0x00,0x00,0x00,0x10,0x00,0x80, /* 104- ? */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 112-119 */ - 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 120- " */ - 0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* 128- g */ - 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* h -143 */ - 0x00,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* 144- p */ - 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* q -159 */ - 0x00,0x00,0x12,0x12,0x12,0x12,0x12,0x12, /* 160- x */ - 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* y -175 */ - 0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ^ -183 */ - 0x00,0x00,0x80,0x00,0x00,0x00,0x00,0x00, /* 184-191 */ - 0x80,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* { - G */ - 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* H -207 */ - 0x00,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* } - P */ - 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* Q -223 */ - 0x00,0x00,0x12,0x12,0x12,0x12,0x12,0x12, /* \ - X */ - 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* Y -239 */ - 0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /* 0 - 7 */ - 0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x00};/* 8 -255 */ -#endif - - + 0x80,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 0- 7 */ + 0x00,0x00,0x00,0x00,0x01,0x01,0x00,0x00, /* 8- 15 */ + 0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 16- 23 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */ + 0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 32- 39 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 40- 47 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 48- 55 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 56- 63 */ + 0x01,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - 71 */ + 0x00,0x00,0x00,0x80,0x00,0x80,0x80,0x80, /* 72- | */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* & - 87 */ + 0x00,0x00,0x00,0x80,0x80,0x80,0x00,0x00, /* 88- 95 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - -103 */ + 0x00,0x00,0x00,0x00,0x00,0x10,0x00,0x80, /* 104- ? */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 112-119 */ + 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 120- " */ + 0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* 128- g */ + 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* h -143 */ + 0x00,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* 144- p */ + 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* q -159 */ + 0x00,0x00,0x12,0x12,0x12,0x12,0x12,0x12, /* 160- x */ + 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* y -175 */ + 0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ^ -183 */ + 0x00,0x00,0x80,0x00,0x00,0x00,0x00,0x00, /* 184-191 */ + 0x80,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* { - G */ + 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* H -207 */ + 0x00,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* } - P */ + 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* Q -223 */ + 0x00,0x00,0x12,0x12,0x12,0x12,0x12,0x12, /* \ - X */ + 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* Y -239 */ + 0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /* 0 - 7 */ + 0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x00};/* 8 -255 */ +#endif + + /* This table is used to check whether auto-possessification is possible between adjacent character-type opcodes. The left-hand (repeated) opcode is used to select the row, and the right-hand opcode is use to select the column. A value of 1 means that auto-possessification is OK. For example, the second value in the first row means that \D+\d can be turned into \D++\d. - + The Unicode property types (\P and \p) have to be present to fill out the table because of what their opcode values are, but the table values should always be zero because property types are handled separately in the code. The last four columns apply to items that cannot be repeated, so there is no need to have rows for them. Note that OP_DIGIT etc. are generated only when PCRE_UCP is *not* set. When it is set, \d etc. are converted into OP_(NOT_)PROP codes. */ - + #define APTROWS (LAST_AUTOTAB_LEFT_OP - FIRST_AUTOTAB_OP + 1) #define APTCOLS (LAST_AUTOTAB_RIGHT_OP - FIRST_AUTOTAB_OP + 1) - + static const pcre_uint8 autoposstab[APTROWS][APTCOLS] = { /* \D \d \S \s \W \w . .+ \C \P \p \R \H \h \V \v \X \Z \z $ $M */ { 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0 }, /* \D */ @@ -736,7 +736,7 @@ static const pcre_uint8 autoposstab[APTROWS][APTCOLS] = { { 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 1, 0, 0 }, /* \v */ { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0 } /* \X */ }; - + /* This table is used to check whether auto-possessification is possible between adjacent Unicode property opcodes (OP_PROP and OP_NOTPROP). The @@ -878,34 +878,34 @@ static const pcre_uint8 opcode_possessify[] = { -/************************************************* -* Find an error text * -*************************************************/ - -/* The error texts are now all in one long string, to save on relocations. As -some of the text is of unknown length, we can't use a table of offsets. -Instead, just count through the strings. This is not a performance issue -because it happens only when there has been a compilation error. - -Argument: the error number -Returns: pointer to the error string -*/ - -static const char * -find_error_text(int n) -{ -const char *s = error_texts; +/************************************************* +* Find an error text * +*************************************************/ + +/* The error texts are now all in one long string, to save on relocations. As +some of the text is of unknown length, we can't use a table of offsets. +Instead, just count through the strings. This is not a performance issue +because it happens only when there has been a compilation error. + +Argument: the error number +Returns: pointer to the error string +*/ + +static const char * +find_error_text(int n) +{ +const char *s = error_texts; for (; n > 0; n--) { while (*s++ != CHAR_NULL) {}; if (*s == CHAR_NULL) return "Error text not found (please report)"; } -return s; -} - - +return s; +} + + -/************************************************* +/************************************************* * Expand the workspace * *************************************************/ @@ -978,82 +978,82 @@ return (*p == CHAR_RIGHT_CURLY_BRACKET); /************************************************* -* Handle escapes * -*************************************************/ - -/* This function is called when a \ has been encountered. It either returns a +* Handle escapes * +*************************************************/ + +/* This function is called when a \ has been encountered. It either returns a positive value for a simple escape such as \n, or 0 for a data character which will be placed in chptr. A backreference to group n is returned as negative n. When UTF-8 is enabled, a positive value greater than 255 may be returned in chptr. On entry, ptr is pointing at the \. On exit, it is on the final character of the escape sequence. - -Arguments: - ptrptr points to the pattern position pointer + +Arguments: + ptrptr points to the pattern position pointer chptr points to a returned data character - errorcodeptr points to the errorcode variable - bracount number of previous extracting brackets - options the options bits - isclass TRUE if inside a character class - + errorcodeptr points to the errorcode variable + bracount number of previous extracting brackets + options the options bits + isclass TRUE if inside a character class + Returns: zero => a data character positive => a special escape sequence negative => a back reference - on error, errorcodeptr is set -*/ - -static int + on error, errorcodeptr is set +*/ + +static int check_escape(const pcre_uchar **ptrptr, pcre_uint32 *chptr, int *errorcodeptr, int bracount, int options, BOOL isclass) -{ +{ /* PCRE_UTF16 has the same value as PCRE_UTF8. */ BOOL utf = (options & PCRE_UTF8) != 0; const pcre_uchar *ptr = *ptrptr + 1; pcre_uint32 c; int escape = 0; int i; - -GETCHARINCTEST(c, ptr); /* Get character value, increment pointer */ -ptr--; /* Set pointer back to the last byte */ - -/* If backslash is at the end of the pattern, it's an error. */ - + +GETCHARINCTEST(c, ptr); /* Get character value, increment pointer */ +ptr--; /* Set pointer back to the last byte */ + +/* If backslash is at the end of the pattern, it's an error. */ + if (c == CHAR_NULL) *errorcodeptr = ERR1; - -/* Non-alphanumerics are literals. For digits or letters, do an initial lookup -in a table. A non-zero result is something that can be returned immediately. -Otherwise further processing may be required. */ - + +/* Non-alphanumerics are literals. For digits or letters, do an initial lookup +in a table. A non-zero result is something that can be returned immediately. +Otherwise further processing may be required. */ + #ifndef EBCDIC /* ASCII/UTF-8 coding */ /* Not alphanumeric */ else if (c < CHAR_0 || c > CHAR_z) {} else if ((i = escapes[c - CHAR_0]) != 0) { if (i > 0) c = (pcre_uint32)i; else escape = -i; } - -#else /* EBCDIC coding */ + +#else /* EBCDIC coding */ /* Not alphanumeric */ else if (c < CHAR_a || (!MAX_255(c) || (ebcdic_chartab[c] & 0x0E) == 0)) {} else if ((i = escapes[c - 0x48]) != 0) { if (i > 0) c = (pcre_uint32)i; else escape = -i; } -#endif - -/* Escapes that need further processing, or are illegal. */ - -else - { +#endif + +/* Escapes that need further processing, or are illegal. */ + +else + { const pcre_uchar *oldptr; BOOL braced, negated, overflow; int s; - - switch (c) - { - /* A number of Perl escapes are not handled by PCRE. We give an explicit - error. */ - + + switch (c) + { + /* A number of Perl escapes are not handled by PCRE. We give an explicit + error. */ + case CHAR_l: case CHAR_L: - *errorcodeptr = ERR37; - break; - + *errorcodeptr = ERR37; + break; + case CHAR_u: if ((options & PCRE_JAVASCRIPT_COMPAT) != 0) { @@ -1076,7 +1076,7 @@ else c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10)); #endif } - + #if defined COMPILE_PCRE8 if (c > (utf ? 0x10ffffU : 0xffU)) #elif defined COMPILE_PCRE16 @@ -1119,7 +1119,7 @@ else case CHAR_g: if (isclass) break; if (ptr[1] == CHAR_LESS_THAN_SIGN || ptr[1] == CHAR_APOSTROPHE) - { + { escape = ESC_g; break; } @@ -1132,27 +1132,27 @@ else for (p = ptr+2; *p != CHAR_NULL && *p != CHAR_RIGHT_CURLY_BRACKET; p++) if (*p != CHAR_MINUS && !IS_DIGIT(*p)) break; if (*p != CHAR_NULL && *p != CHAR_RIGHT_CURLY_BRACKET) - { + { escape = ESC_k; - break; - } - braced = TRUE; - ptr++; - } - else braced = FALSE; - + break; + } + braced = TRUE; + ptr++; + } + else braced = FALSE; + if (ptr[1] == CHAR_MINUS) - { - negated = TRUE; - ptr++; - } - else negated = FALSE; - + { + negated = TRUE; + ptr++; + } + else negated = FALSE; + /* The integer range is limited by the machine's int representation. */ s = 0; overflow = FALSE; while (IS_DIGIT(ptr[1])) - { + { if (s > INT_MAX / 10 - 1) /* Integer overflow */ { overflow = TRUE; @@ -1164,62 +1164,62 @@ else { while (IS_DIGIT(ptr[1])) ptr++; - *errorcodeptr = ERR61; - break; - } - + *errorcodeptr = ERR61; + break; + } + if (braced && *(++ptr) != CHAR_RIGHT_CURLY_BRACKET) - { - *errorcodeptr = ERR57; - break; - } - + { + *errorcodeptr = ERR57; + break; + } + if (s == 0) { *errorcodeptr = ERR58; break; } - if (negated) - { + if (negated) + { if (s > bracount) - { - *errorcodeptr = ERR15; - break; - } + { + *errorcodeptr = ERR15; + break; + } s = bracount - (s - 1); - } - + } + escape = -s; - break; - - /* The handling of escape sequences consisting of a string of digits + break; + + /* The handling of escape sequences consisting of a string of digits starting with one that is not zero is not straightforward. Perl has changed over the years. Nowadays \g{} for backreferences and \o{} for octal are recommended to avoid the ambiguities in the old syntax. - - Outside a character class, the digits are read as a decimal number. If the + + Outside a character class, the digits are read as a decimal number. If the number is less than 8 (used to be 10), or if there are that many previous extracting left brackets, then it is a back reference. Otherwise, up to three octal digits are read to form an escaped byte. Thus \123 is likely to be octal 123 (cf \0123, which is octal 012 followed by the literal 3). If the octal value is greater than 377, the least significant 8 bits are taken. \8 and \9 are treated as the literal characters 8 and 9. - + Inside a character class, \ followed by a digit is always either a literal 8 or 9 or an octal number. */ - + case CHAR_1: case CHAR_2: case CHAR_3: case CHAR_4: case CHAR_5: case CHAR_6: case CHAR_7: case CHAR_8: case CHAR_9: - if (!isclass) - { - oldptr = ptr; + if (!isclass) + { + oldptr = ptr; /* The integer range is limited by the machine's int representation. */ s = (int)(c -CHAR_0); overflow = FALSE; while (IS_DIGIT(ptr[1])) - { + { if (s > INT_MAX / 10 - 1) /* Integer overflow */ { overflow = TRUE; @@ -1231,32 +1231,32 @@ else { while (IS_DIGIT(ptr[1])) ptr++; - *errorcodeptr = ERR61; - break; - } + *errorcodeptr = ERR61; + break; + } if (s < 8 || s <= bracount) /* Check for back reference */ - { + { escape = -s; - break; - } - ptr = oldptr; /* Put the pointer back and fall through */ - } - + break; + } + ptr = oldptr; /* Put the pointer back and fall through */ + } + /* Handle a digit following \ when the number is not a back reference. If the first digit is 8 or 9, Perl used to generate a binary zero byte and then treat the digit as a following literal. At least by Perl 5.18 this changed so as not to insert the binary zero. */ - + if ((c = *ptr) >= CHAR_8) break; - + /* Fall through with a digit less than 8 */ - /* \0 always starts an octal number, but we may drop through to here with a - larger first octal digit. The original code used just to take the least - significant 8 bits of octal numbers (I think this is what early Perls used + /* \0 always starts an octal number, but we may drop through to here with a + larger first octal digit. The original code used just to take the least + significant 8 bits of octal numbers (I think this is what early Perls used to do). Nowadays we allow for larger numbers in UTF-8 mode and 16-bit mode, but no more than 3 octal digits. */ - + case CHAR_0: c -= CHAR_0; while(i++ < 2 && ptr[1] >= CHAR_0 && ptr[1] <= CHAR_7) @@ -1264,20 +1264,20 @@ else #ifdef COMPILE_PCRE8 if (!utf && c > 0xff) *errorcodeptr = ERR51; #endif - break; - + break; + /* \o is a relatively new Perl feature, supporting a more general way of specifying character codes in octal. The only supported form is \o{ddd}. */ - + case CHAR_o: if (ptr[1] != CHAR_LEFT_CURLY_BRACKET) *errorcodeptr = ERR81; else if (ptr[2] == CHAR_RIGHT_CURLY_BRACKET) *errorcodeptr = ERR86; else - { + { ptr += 2; - c = 0; + c = 0; overflow = FALSE; while (*ptr >= CHAR_0 && *ptr <= CHAR_7) - { + { register pcre_uint32 cc = *ptr++; if (c == 0 && cc == CHAR_0) continue; /* Leading zeroes */ #ifdef COMPILE_PCRE32 @@ -1304,7 +1304,7 @@ else else *errorcodeptr = ERR80; } break; - + /* \x is complicated. In JavaScript, \x must be followed by two hexadecimal numbers. Otherwise it is a lowercase x letter. */ @@ -1321,14 +1321,14 @@ else #ifndef EBCDIC /* ASCII/UTF-8 coding */ if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */ c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10)); -#else /* EBCDIC coding */ +#else /* EBCDIC coding */ if (cc >= CHAR_a && cc <= CHAR_z) cc += 64; /* Convert to upper case */ c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10)); -#endif +#endif } - } + } } /* End JavaScript handling */ - + /* Handle \x in Perl's style. \x{ddd} is a character number which can be greater than 0xff in utf or non-8bit mode, but only if the ddd are hex digits. If not, { used to be treated as a data character. However, Perl @@ -1339,7 +1339,7 @@ else else { if (ptr[1] == CHAR_LEFT_CURLY_BRACKET) - { + { ptr += 2; if (*ptr == CHAR_RIGHT_CURLY_BRACKET) { @@ -1352,11 +1352,11 @@ else { register pcre_uint32 cc = *ptr++; if (c == 0 && cc == CHAR_0) continue; /* Leading zeroes */ - + #ifdef COMPILE_PCRE32 if (c >= 0x10000000l) { overflow = TRUE; break; } #endif - + #ifndef EBCDIC /* ASCII/UTF-8 coding */ if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */ c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10)); @@ -1364,7 +1364,7 @@ else if (cc >= CHAR_a && cc <= CHAR_z) cc += 64; /* Convert to upper case */ c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10)); #endif - + #if defined COMPILE_PCRE8 if (c > (utf ? 0x10ffffU : 0xffU)) { overflow = TRUE; break; } #elif defined COMPILE_PCRE16 @@ -1405,27 +1405,27 @@ else #ifndef EBCDIC /* ASCII/UTF-8 coding */ if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */ c = c * 16 + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10)); -#else /* EBCDIC coding */ +#else /* EBCDIC coding */ if (cc <= CHAR_z) cc += 64; /* Convert to upper case */ c = c * 16 + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10)); -#endif +#endif } } /* End of \xdd handling */ } /* End of Perl-style \x handling */ - break; - - /* For \c, a following letter is upper-cased; then the 0x40 bit is flipped. + break; + + /* For \c, a following letter is upper-cased; then the 0x40 bit is flipped. An error is given if the byte following \c is not an ASCII character. This coding is ASCII-specific, but then the whole concept of \cx is - ASCII-specific. (However, an EBCDIC equivalent has now been added.) */ - + ASCII-specific. (However, an EBCDIC equivalent has now been added.) */ + case CHAR_c: - c = *(++ptr); + c = *(++ptr); if (c == CHAR_NULL) - { - *errorcodeptr = ERR2; - break; - } + { + *errorcodeptr = ERR2; + break; + } #ifndef EBCDIC /* ASCII/UTF-8 coding */ if (c > 127) /* Excludes all non-ASCII in either mode */ { @@ -1433,7 +1433,7 @@ else break; } if (c >= CHAR_a && c <= CHAR_z) c -= 32; - c ^= 0x40; + c ^= 0x40; #else /* EBCDIC coding */ if (c >= CHAR_a && c <= CHAR_z) c += 64; if (c == CHAR_QUESTION_MARK) @@ -1446,26 +1446,26 @@ else } if (i < 32) c = i; else *errorcodeptr = ERR68; } -#endif - break; - - /* PCRE_EXTRA enables extensions to Perl in the matter of escapes. Any - other alphanumeric following \ is an error if PCRE_EXTRA was set; - otherwise, for Perl compatibility, it is a literal. This code looks a bit - odd, but there used to be some cases other than the default, and there may - be again in future, so I haven't "optimized" it. */ - - default: - if ((options & PCRE_EXTRA) != 0) switch(c) - { - default: - *errorcodeptr = ERR3; - break; - } - break; - } - } - +#endif + break; + + /* PCRE_EXTRA enables extensions to Perl in the matter of escapes. Any + other alphanumeric following \ is an error if PCRE_EXTRA was set; + otherwise, for Perl compatibility, it is a literal. This code looks a bit + odd, but there used to be some cases other than the default, and there may + be again in future, so I haven't "optimized" it. */ + + default: + if ((options & PCRE_EXTRA) != 0) switch(c) + { + default: + *errorcodeptr = ERR3; + break; + } + break; + } + } + /* Perl supports \N{name} for character names, as well as plain \N for "not newline". PCRE does not support \N{name}. However, it does support quantification such as \N{2,3}. */ @@ -1481,314 +1481,314 @@ if ((options & PCRE_UCP) != 0 && escape >= ESC_D && escape <= ESC_w) /* Set the pointer to the final character before returning. */ -*ptrptr = ptr; +*ptrptr = ptr; *chptr = c; return escape; -} - - - -#ifdef SUPPORT_UCP -/************************************************* -* Handle \P and \p * -*************************************************/ - -/* This function is called after \P or \p has been encountered, provided that -PCRE is compiled with support for Unicode properties. On entry, ptrptr is -pointing at the P or p. On exit, it is pointing at the final character of the -escape sequence. - -Argument: - ptrptr points to the pattern position pointer - negptr points to a boolean that is set TRUE for negation else FALSE +} + + + +#ifdef SUPPORT_UCP +/************************************************* +* Handle \P and \p * +*************************************************/ + +/* This function is called after \P or \p has been encountered, provided that +PCRE is compiled with support for Unicode properties. On entry, ptrptr is +pointing at the P or p. On exit, it is pointing at the final character of the +escape sequence. + +Argument: + ptrptr points to the pattern position pointer + negptr points to a boolean that is set TRUE for negation else FALSE ptypeptr points to an unsigned int that is set to the type value pdataptr points to an unsigned int that is set to the detailed property value - errorcodeptr points to the error code variable - + errorcodeptr points to the error code variable + Returns: TRUE if the type value was found, or FALSE for an invalid type -*/ - +*/ + static BOOL get_ucp(const pcre_uchar **ptrptr, BOOL *negptr, unsigned int *ptypeptr, unsigned int *pdataptr, int *errorcodeptr) -{ +{ pcre_uchar c; int i, bot, top; const pcre_uchar *ptr = *ptrptr; pcre_uchar name[32]; - -c = *(++ptr); + +c = *(++ptr); if (c == CHAR_NULL) goto ERROR_RETURN; - -*negptr = FALSE; - -/* \P or \p can be followed by a name in {}, optionally preceded by ^ for -negation. */ - + +*negptr = FALSE; + +/* \P or \p can be followed by a name in {}, optionally preceded by ^ for +negation. */ + if (c == CHAR_LEFT_CURLY_BRACKET) - { + { if (ptr[1] == CHAR_CIRCUMFLEX_ACCENT) - { - *negptr = TRUE; - ptr++; - } + { + *negptr = TRUE; + ptr++; + } for (i = 0; i < (int)(sizeof(name) / sizeof(pcre_uchar)) - 1; i++) - { - c = *(++ptr); + { + c = *(++ptr); if (c == CHAR_NULL) goto ERROR_RETURN; if (c == CHAR_RIGHT_CURLY_BRACKET) break; - name[i] = c; - } + name[i] = c; + } if (c != CHAR_RIGHT_CURLY_BRACKET) goto ERROR_RETURN; - name[i] = 0; - } - -/* Otherwise there is just one following character */ - -else - { - name[0] = c; - name[1] = 0; - } - -*ptrptr = ptr; - -/* Search for a recognized property name using binary chop */ - -bot = 0; + name[i] = 0; + } + +/* Otherwise there is just one following character */ + +else + { + name[0] = c; + name[1] = 0; + } + +*ptrptr = ptr; + +/* Search for a recognized property name using binary chop */ + +bot = 0; top = PRIV(utt_size); - -while (bot < top) - { + +while (bot < top) + { int r; - i = (bot + top) >> 1; + i = (bot + top) >> 1; r = STRCMP_UC_C8(name, PRIV(utt_names) + PRIV(utt)[i].name_offset); if (r == 0) - { + { *ptypeptr = PRIV(utt)[i].type; *pdataptr = PRIV(utt)[i].value; return TRUE; - } + } if (r > 0) bot = i + 1; else top = i; - } - -*errorcodeptr = ERR47; -*ptrptr = ptr; + } + +*errorcodeptr = ERR47; +*ptrptr = ptr; return FALSE; - -ERROR_RETURN: -*errorcodeptr = ERR46; -*ptrptr = ptr; + +ERROR_RETURN: +*errorcodeptr = ERR46; +*ptrptr = ptr; return FALSE; -} -#endif - - - -/************************************************* -* Read repeat counts * -*************************************************/ - -/* Read an item of the form {n,m} and return the values. This is called only -after is_counted_repeat() has confirmed that a repeat-count quantifier exists, -so the syntax is guaranteed to be correct, but we need to check the values. - -Arguments: - p pointer to first char after '{' - minp pointer to int for min - maxp pointer to int for max - returned as -1 if no max - errorcodeptr points to error code variable - -Returns: pointer to '}' on success; - current ptr on error, with errorcodeptr set non-zero -*/ - +} +#endif + + + +/************************************************* +* Read repeat counts * +*************************************************/ + +/* Read an item of the form {n,m} and return the values. This is called only +after is_counted_repeat() has confirmed that a repeat-count quantifier exists, +so the syntax is guaranteed to be correct, but we need to check the values. + +Arguments: + p pointer to first char after '{' + minp pointer to int for min + maxp pointer to int for max + returned as -1 if no max + errorcodeptr points to error code variable + +Returns: pointer to '}' on success; + current ptr on error, with errorcodeptr set non-zero +*/ + static const pcre_uchar * read_repeat_counts(const pcre_uchar *p, int *minp, int *maxp, int *errorcodeptr) -{ -int min = 0; -int max = -1; - +{ +int min = 0; +int max = -1; + while (IS_DIGIT(*p)) - { + { min = min * 10 + (int)(*p++ - CHAR_0); if (min > 65535) { *errorcodeptr = ERR5; return p; } - } - + } + if (*p == CHAR_RIGHT_CURLY_BRACKET) max = min; else - { + { if (*(++p) != CHAR_RIGHT_CURLY_BRACKET) - { - max = 0; + { + max = 0; while(IS_DIGIT(*p)) - { + { max = max * 10 + (int)(*p++ - CHAR_0); if (max > 65535) { *errorcodeptr = ERR5; return p; } - } - if (max < min) - { - *errorcodeptr = ERR4; - return p; - } - } - } - -*minp = min; -*maxp = max; -return p; -} - - - -/************************************************* -* Find first significant op code * -*************************************************/ - -/* This is called by several functions that scan a compiled expression looking -for a fixed first character, or an anchoring op code etc. It skips over things + } + if (max < min) + { + *errorcodeptr = ERR4; + return p; + } + } + } + +*minp = min; +*maxp = max; +return p; +} + + + +/************************************************* +* Find first significant op code * +*************************************************/ + +/* This is called by several functions that scan a compiled expression looking +for a fixed first character, or an anchoring op code etc. It skips over things that do not influence this. For some calls, it makes sense to skip negative forward and all backward assertions, and also the \b assertion; for others it does not. - -Arguments: - code pointer to the start of the group - skipassert TRUE if certain assertions are to be skipped - -Returns: pointer to the first significant opcode -*/ - + +Arguments: + code pointer to the start of the group + skipassert TRUE if certain assertions are to be skipped + +Returns: pointer to the first significant opcode +*/ + static const pcre_uchar* first_significant_code(const pcre_uchar *code, BOOL skipassert) -{ -for (;;) - { - switch ((int)*code) - { - case OP_ASSERT_NOT: - case OP_ASSERTBACK: - case OP_ASSERTBACK_NOT: - if (!skipassert) return code; - do code += GET(code, 1); while (*code == OP_ALT); +{ +for (;;) + { + switch ((int)*code) + { + case OP_ASSERT_NOT: + case OP_ASSERTBACK: + case OP_ASSERTBACK_NOT: + if (!skipassert) return code; + do code += GET(code, 1); while (*code == OP_ALT); code += PRIV(OP_lengths)[*code]; - break; - - case OP_WORD_BOUNDARY: - case OP_NOT_WORD_BOUNDARY: - if (!skipassert) return code; - /* Fall through */ - - case OP_CALLOUT: - case OP_CREF: + break; + + case OP_WORD_BOUNDARY: + case OP_NOT_WORD_BOUNDARY: + if (!skipassert) return code; + /* Fall through */ + + case OP_CALLOUT: + case OP_CREF: case OP_DNCREF: - case OP_RREF: + case OP_RREF: case OP_DNRREF: - case OP_DEF: + case OP_DEF: code += PRIV(OP_lengths)[*code]; - break; - - default: - return code; - } - } -/* Control never reaches here */ -} - - - -/************************************************* + break; + + default: + return code; + } + } +/* Control never reaches here */ +} + + + +/************************************************* * Find the fixed length of a branch * -*************************************************/ - +*************************************************/ + /* Scan a branch and compute the fixed length of subject that will match it, -if the length is fixed. This is needed for dealing with backward assertions. +if the length is fixed. This is needed for dealing with backward assertions. In UTF8 mode, the result is in characters rather than bytes. The branch is temporarily terminated with OP_END when this function is called. - + This function is called when a backward assertion is encountered, so that if it fails, the error message can point to the correct place in the pattern. However, we cannot do this when the assertion contains subroutine calls, because they can be forward references. We solve this by remembering this case and doing the check at the end; a flag specifies which mode we are running in. -Arguments: - code points to the start of the pattern (the bracket) +Arguments: + code points to the start of the pattern (the bracket) utf TRUE in UTF-8 / UTF-16 / UTF-32 mode atend TRUE if called when the pattern is complete cd the "compile data" structure recurses chain of recurse_check to catch mutual recursion - + Returns: the fixed length, or -1 if there is no fixed length, or -2 if \C was encountered (in UTF-8 mode only) or -3 if an OP_RECURSE item was encountered and atend is FALSE or -4 if an unknown opcode was encountered (internal error) -*/ - -static int +*/ + +static int find_fixedlength(pcre_uchar *code, BOOL utf, BOOL atend, compile_data *cd, recurse_check *recurses) -{ -int length = -1; +{ +int length = -1; recurse_check this_recurse; -register int branchlength = 0; +register int branchlength = 0; register pcre_uchar *cc = code + 1 + LINK_SIZE; - -/* Scan along the opcodes for this branch. If we get to the end of the -branch, check the length against that of the other branches. */ - -for (;;) - { - int d; + +/* Scan along the opcodes for this branch. If we get to the end of the +branch, check the length against that of the other branches. */ + +for (;;) + { + int d; pcre_uchar *ce, *cs; register pcre_uchar op = *cc; - switch (op) - { + switch (op) + { /* We only need to continue for OP_CBRA (normal capturing bracket) and OP_BRA (normal non-capturing bracket) because the other variants of these opcodes are all concerned with unlimited repeated groups, which of course are not of fixed length. */ - case OP_CBRA: - case OP_BRA: - case OP_ONCE: + case OP_CBRA: + case OP_BRA: + case OP_ONCE: case OP_ONCE_NC: - case OP_COND: + case OP_COND: d = find_fixedlength(cc + ((op == OP_CBRA)? IMM2_SIZE : 0), utf, atend, cd, recurses); - if (d < 0) return d; - branchlength += d; - do cc += GET(cc, 1); while (*cc == OP_ALT); - cc += 1 + LINK_SIZE; - break; - + if (d < 0) return d; + branchlength += d; + do cc += GET(cc, 1); while (*cc == OP_ALT); + cc += 1 + LINK_SIZE; + break; + /* Reached end of a branch; if it's a ket it is the end of a nested call. If it's ALT it is an alternation in a nested call. An ACCEPT is effectively an ALT. If it is END it's the end of the outer call. All can be handled by the same code. Note that we must not include the OP_KETRxxx opcodes here, because they all imply an unlimited repeat. */ - - case OP_ALT: - case OP_KET: - case OP_END: + + case OP_ALT: + case OP_KET: + case OP_END: case OP_ACCEPT: case OP_ASSERT_ACCEPT: - if (length < 0) length = branchlength; - else if (length != branchlength) return -1; - if (*cc != OP_ALT) return length; - cc += 1 + LINK_SIZE; - branchlength = 0; - break; - + if (length < 0) length = branchlength; + else if (length != branchlength) return -1; + if (*cc != OP_ALT) return length; + cc += 1 + LINK_SIZE; + branchlength = 0; + break; + /* A true recursion implies not fixed length, but a subroutine call may be OK. If the subroutine is a forward reference, we can't deal with it until the end of the pattern, so return -3. */ @@ -1812,18 +1812,18 @@ for (;;) cc += 1 + LINK_SIZE; break; - /* Skip over assertive subpatterns */ - - case OP_ASSERT: - case OP_ASSERT_NOT: - case OP_ASSERTBACK: - case OP_ASSERTBACK_NOT: - do cc += GET(cc, 1); while (*cc == OP_ALT); + /* Skip over assertive subpatterns */ + + case OP_ASSERT: + case OP_ASSERT_NOT: + case OP_ASSERTBACK: + case OP_ASSERTBACK_NOT: + do cc += GET(cc, 1); while (*cc == OP_ALT); cc += 1 + LINK_SIZE; break; - - /* Skip over things that don't match chars */ - + + /* Skip over things that don't match chars */ + case OP_MARK: case OP_PRUNE_ARG: case OP_SKIP_ARG: @@ -1836,16 +1836,16 @@ for (;;) case OP_CIRCM: case OP_CLOSE: case OP_COMMIT: - case OP_CREF: - case OP_DEF: + case OP_CREF: + case OP_DEF: case OP_DNCREF: case OP_DNRREF: case OP_DOLL: case OP_DOLLM: - case OP_EOD: - case OP_EODN: + case OP_EOD: + case OP_EODN: case OP_FAIL: - case OP_NOT_WORD_BOUNDARY: + case OP_NOT_WORD_BOUNDARY: case OP_PRUNE: case OP_REVERSE: case OP_RREF: @@ -1854,27 +1854,27 @@ for (;;) case OP_SOD: case OP_SOM: case OP_THEN: - case OP_WORD_BOUNDARY: + case OP_WORD_BOUNDARY: cc += PRIV(OP_lengths)[*cc]; - break; - - /* Handle literal characters */ - - case OP_CHAR: + break; + + /* Handle literal characters */ + + case OP_CHAR: case OP_CHARI: - case OP_NOT: + case OP_NOT: case OP_NOTI: - branchlength++; - cc += 2; + branchlength++; + cc += 2; #ifdef SUPPORT_UTF if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]); -#endif - break; - - /* Handle exact repetitions. The count is already in characters, but we - need to skip over a multibyte character in UTF8 mode. */ - - case OP_EXACT: +#endif + break; + + /* Handle exact repetitions. The count is already in characters, but we + need to skip over a multibyte character in UTF8 mode. */ + + case OP_EXACT: case OP_EXACTI: case OP_NOTEXACT: case OP_NOTEXACTI: @@ -1882,51 +1882,51 @@ for (;;) cc += 2 + IMM2_SIZE; #ifdef SUPPORT_UTF if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]); -#endif - break; - - case OP_TYPEEXACT: - branchlength += GET2(cc,1); +#endif + break; + + case OP_TYPEEXACT: + branchlength += GET2(cc,1); if (cc[1 + IMM2_SIZE] == OP_PROP || cc[1 + IMM2_SIZE] == OP_NOTPROP) cc += 2; cc += 1 + IMM2_SIZE + 1; - break; - - /* Handle single-char matchers */ - - case OP_PROP: - case OP_NOTPROP: - cc += 2; - /* Fall through */ - + break; + + /* Handle single-char matchers */ + + case OP_PROP: + case OP_NOTPROP: + cc += 2; + /* Fall through */ + case OP_HSPACE: case OP_VSPACE: case OP_NOT_HSPACE: case OP_NOT_VSPACE: - case OP_NOT_DIGIT: - case OP_DIGIT: - case OP_NOT_WHITESPACE: - case OP_WHITESPACE: - case OP_NOT_WORDCHAR: - case OP_WORDCHAR: - case OP_ANY: + case OP_NOT_DIGIT: + case OP_DIGIT: + case OP_NOT_WHITESPACE: + case OP_WHITESPACE: + case OP_NOT_WORDCHAR: + case OP_WORDCHAR: + case OP_ANY: case OP_ALLANY: - branchlength++; - cc++; - break; - + branchlength++; + cc++; + break; + /* The single-byte matcher isn't allowed. This only happens in UTF-8 mode; otherwise \C is coded as OP_ALLANY. */ - - case OP_ANYBYTE: - return -2; - - /* Check a class for variable quantification */ - + + case OP_ANYBYTE: + return -2; + + /* Check a class for variable quantification */ + case OP_CLASS: case OP_NCLASS: #if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32 - case OP_XCLASS: + case OP_XCLASS: /* The original code caused an unsigned overflow in 64 bit systems, so now we use a conditional statement. */ if (op == OP_XCLASS) @@ -1935,36 +1935,36 @@ for (;;) cc += PRIV(OP_lengths)[OP_CLASS]; #else cc += PRIV(OP_lengths)[OP_CLASS]; -#endif - - switch (*cc) - { - case OP_CRSTAR: - case OP_CRMINSTAR: +#endif + + switch (*cc) + { + case OP_CRSTAR: + case OP_CRMINSTAR: case OP_CRPLUS: case OP_CRMINPLUS: - case OP_CRQUERY: - case OP_CRMINQUERY: + case OP_CRQUERY: + case OP_CRMINQUERY: case OP_CRPOSSTAR: case OP_CRPOSPLUS: case OP_CRPOSQUERY: - return -1; - - case OP_CRRANGE: - case OP_CRMINRANGE: + return -1; + + case OP_CRRANGE: + case OP_CRMINRANGE: case OP_CRPOSRANGE: if (GET2(cc,1) != GET2(cc,1+IMM2_SIZE)) return -1; branchlength += (int)GET2(cc,1); cc += 1 + 2 * IMM2_SIZE; - break; - - default: - branchlength++; - } - break; - - /* Anything else is variable length */ - + break; + + default: + branchlength++; + } + break; + + /* Anything else is variable length */ + case OP_ANYNL: case OP_BRAMINZERO: case OP_BRAPOS: @@ -2050,48 +2050,48 @@ for (;;) /* Catch unrecognized opcodes so that when new ones are added they are not forgotten, as has happened in the past. */ - default: + default: return -4; - } - } -/* Control never gets here */ -} - - - -/************************************************* + } + } +/* Control never gets here */ +} + + + +/************************************************* * Scan compiled regex for specific bracket * -*************************************************/ - -/* This little function scans through a compiled pattern until it finds a +*************************************************/ + +/* This little function scans through a compiled pattern until it finds a capturing bracket with the given number, or, if the number is negative, an instance of OP_REVERSE for a lookbehind. The function is global in the C sense so that it can be called from pcre_study() when finding the minimum matching length. - -Arguments: - code points to start of expression + +Arguments: + code points to start of expression utf TRUE in UTF-8 / UTF-16 / UTF-32 mode number the required bracket number or negative to find a lookbehind - -Returns: pointer to the opcode for the bracket, or NULL if not found -*/ - + +Returns: pointer to the opcode for the bracket, or NULL if not found +*/ + const pcre_uchar * PRIV(find_bracket)(const pcre_uchar *code, BOOL utf, int number) -{ -for (;;) - { +{ +for (;;) + { register pcre_uchar c = *code; - if (c == OP_END) return NULL; - - /* XCLASS is used for classes that cannot be represented just by a bit - map. This includes negated single high-valued characters. The length in - the table is zero; the actual length is stored in the compiled code. */ - - if (c == OP_XCLASS) code += GET(code, 1); - + if (c == OP_END) return NULL; + + /* XCLASS is used for classes that cannot be represented just by a bit + map. This includes negated single high-valued characters. The length in + the table is zero; the actual length is stored in the compiled code. */ + + if (c == OP_XCLASS) code += GET(code, 1); + /* Handle recursion */ else if (c == OP_REVERSE) @@ -2100,44 +2100,44 @@ for (;;) code += PRIV(OP_lengths)[c]; } - /* Handle capturing bracket */ - + /* Handle capturing bracket */ + else if (c == OP_CBRA || c == OP_SCBRA || c == OP_CBRAPOS || c == OP_SCBRAPOS) - { + { int n = (int)GET2(code, 1+LINK_SIZE); if (n == number) return (pcre_uchar *)code; code += PRIV(OP_lengths)[c]; - } - - /* Otherwise, we can get the item's length from the table, except that for - repeated character types, we have to test for \p and \P, which have an extra + } + + /* Otherwise, we can get the item's length from the table, except that for + repeated character types, we have to test for \p and \P, which have an extra two bytes of parameters, and for MARK/PRUNE/SKIP/THEN with an argument, we must add in its length. */ - - else - { - switch(c) - { - case OP_TYPESTAR: - case OP_TYPEMINSTAR: - case OP_TYPEPLUS: - case OP_TYPEMINPLUS: - case OP_TYPEQUERY: - case OP_TYPEMINQUERY: - case OP_TYPEPOSSTAR: - case OP_TYPEPOSPLUS: - case OP_TYPEPOSQUERY: - if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2; - break; - - case OP_TYPEUPTO: - case OP_TYPEMINUPTO: - case OP_TYPEEXACT: - case OP_TYPEPOSUPTO: + + else + { + switch(c) + { + case OP_TYPESTAR: + case OP_TYPEMINSTAR: + case OP_TYPEPLUS: + case OP_TYPEMINPLUS: + case OP_TYPEQUERY: + case OP_TYPEMINQUERY: + case OP_TYPEPOSSTAR: + case OP_TYPEPOSPLUS: + case OP_TYPEPOSQUERY: + if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2; + break; + + case OP_TYPEUPTO: + case OP_TYPEMINUPTO: + case OP_TYPEEXACT: + case OP_TYPEPOSUPTO: if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP) code += 2; - break; + break; case OP_MARK: case OP_PRUNE_ARG: @@ -2145,144 +2145,144 @@ for (;;) case OP_THEN_ARG: code += code[1]; break; - } - - /* Add in the fixed length from the table */ - + } + + /* Add in the fixed length from the table */ + code += PRIV(OP_lengths)[c]; - - /* In UTF-8 mode, opcodes that are followed by a character may be followed by - a multi-byte character. The length in the table is a minimum, so we have to - arrange to skip the extra bytes. */ - + + /* In UTF-8 mode, opcodes that are followed by a character may be followed by + a multi-byte character. The length in the table is a minimum, so we have to + arrange to skip the extra bytes. */ + #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf) switch(c) - { - case OP_CHAR: + { + case OP_CHAR: case OP_CHARI: case OP_NOT: case OP_NOTI: - case OP_EXACT: + case OP_EXACT: case OP_EXACTI: case OP_NOTEXACT: case OP_NOTEXACTI: - case OP_UPTO: + case OP_UPTO: case OP_UPTOI: case OP_NOTUPTO: case OP_NOTUPTOI: - case OP_MINUPTO: + case OP_MINUPTO: case OP_MINUPTOI: case OP_NOTMINUPTO: case OP_NOTMINUPTOI: - case OP_POSUPTO: + case OP_POSUPTO: case OP_POSUPTOI: case OP_NOTPOSUPTO: case OP_NOTPOSUPTOI: - case OP_STAR: + case OP_STAR: case OP_STARI: case OP_NOTSTAR: case OP_NOTSTARI: - case OP_MINSTAR: + case OP_MINSTAR: case OP_MINSTARI: case OP_NOTMINSTAR: case OP_NOTMINSTARI: - case OP_POSSTAR: + case OP_POSSTAR: case OP_POSSTARI: case OP_NOTPOSSTAR: case OP_NOTPOSSTARI: - case OP_PLUS: + case OP_PLUS: case OP_PLUSI: case OP_NOTPLUS: case OP_NOTPLUSI: - case OP_MINPLUS: + case OP_MINPLUS: case OP_MINPLUSI: case OP_NOTMINPLUS: case OP_NOTMINPLUSI: - case OP_POSPLUS: + case OP_POSPLUS: case OP_POSPLUSI: case OP_NOTPOSPLUS: case OP_NOTPOSPLUSI: - case OP_QUERY: + case OP_QUERY: case OP_QUERYI: case OP_NOTQUERY: case OP_NOTQUERYI: - case OP_MINQUERY: + case OP_MINQUERY: case OP_MINQUERYI: case OP_NOTMINQUERY: case OP_NOTMINQUERYI: - case OP_POSQUERY: + case OP_POSQUERY: case OP_POSQUERYI: case OP_NOTPOSQUERY: case OP_NOTPOSQUERYI: if (HAS_EXTRALEN(code[-1])) code += GET_EXTRALEN(code[-1]); - break; - } + break; + } #else (void)(utf); /* Keep compiler happy by referencing function argument */ -#endif - } - } -} - - - -/************************************************* -* Scan compiled regex for recursion reference * -*************************************************/ - -/* This little function scans through a compiled pattern until it finds an -instance of OP_RECURSE. - -Arguments: - code points to start of expression +#endif + } + } +} + + + +/************************************************* +* Scan compiled regex for recursion reference * +*************************************************/ + +/* This little function scans through a compiled pattern until it finds an +instance of OP_RECURSE. + +Arguments: + code points to start of expression utf TRUE in UTF-8 / UTF-16 / UTF-32 mode - -Returns: pointer to the opcode for OP_RECURSE, or NULL if not found -*/ - + +Returns: pointer to the opcode for OP_RECURSE, or NULL if not found +*/ + static const pcre_uchar * find_recurse(const pcre_uchar *code, BOOL utf) -{ -for (;;) - { +{ +for (;;) + { register pcre_uchar c = *code; - if (c == OP_END) return NULL; - if (c == OP_RECURSE) return code; - - /* XCLASS is used for classes that cannot be represented just by a bit - map. This includes negated single high-valued characters. The length in - the table is zero; the actual length is stored in the compiled code. */ - - if (c == OP_XCLASS) code += GET(code, 1); - - /* Otherwise, we can get the item's length from the table, except that for - repeated character types, we have to test for \p and \P, which have an extra + if (c == OP_END) return NULL; + if (c == OP_RECURSE) return code; + + /* XCLASS is used for classes that cannot be represented just by a bit + map. This includes negated single high-valued characters. The length in + the table is zero; the actual length is stored in the compiled code. */ + + if (c == OP_XCLASS) code += GET(code, 1); + + /* Otherwise, we can get the item's length from the table, except that for + repeated character types, we have to test for \p and \P, which have an extra two bytes of parameters, and for MARK/PRUNE/SKIP/THEN with an argument, we must add in its length. */ - - else - { - switch(c) - { - case OP_TYPESTAR: - case OP_TYPEMINSTAR: - case OP_TYPEPLUS: - case OP_TYPEMINPLUS: - case OP_TYPEQUERY: - case OP_TYPEMINQUERY: - case OP_TYPEPOSSTAR: - case OP_TYPEPOSPLUS: - case OP_TYPEPOSQUERY: - if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2; - break; - - case OP_TYPEPOSUPTO: - case OP_TYPEUPTO: - case OP_TYPEMINUPTO: - case OP_TYPEEXACT: + + else + { + switch(c) + { + case OP_TYPESTAR: + case OP_TYPEMINSTAR: + case OP_TYPEPLUS: + case OP_TYPEMINPLUS: + case OP_TYPEQUERY: + case OP_TYPEMINQUERY: + case OP_TYPEPOSSTAR: + case OP_TYPEPOSPLUS: + case OP_TYPEPOSQUERY: + if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2; + break; + + case OP_TYPEPOSUPTO: + case OP_TYPEUPTO: + case OP_TYPEMINUPTO: + case OP_TYPEEXACT: if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP) code += 2; - break; + break; case OP_MARK: case OP_PRUNE_ARG: @@ -2290,134 +2290,134 @@ for (;;) case OP_THEN_ARG: code += code[1]; break; - } - - /* Add in the fixed length from the table */ - + } + + /* Add in the fixed length from the table */ + code += PRIV(OP_lengths)[c]; - - /* In UTF-8 mode, opcodes that are followed by a character may be followed - by a multi-byte character. The length in the table is a minimum, so we have - to arrange to skip the extra bytes. */ - + + /* In UTF-8 mode, opcodes that are followed by a character may be followed + by a multi-byte character. The length in the table is a minimum, so we have + to arrange to skip the extra bytes. */ + #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf) switch(c) - { - case OP_CHAR: + { + case OP_CHAR: case OP_CHARI: case OP_NOT: case OP_NOTI: - case OP_EXACT: + case OP_EXACT: case OP_EXACTI: case OP_NOTEXACT: case OP_NOTEXACTI: - case OP_UPTO: + case OP_UPTO: case OP_UPTOI: case OP_NOTUPTO: case OP_NOTUPTOI: - case OP_MINUPTO: + case OP_MINUPTO: case OP_MINUPTOI: case OP_NOTMINUPTO: case OP_NOTMINUPTOI: - case OP_POSUPTO: + case OP_POSUPTO: case OP_POSUPTOI: case OP_NOTPOSUPTO: case OP_NOTPOSUPTOI: - case OP_STAR: + case OP_STAR: case OP_STARI: case OP_NOTSTAR: case OP_NOTSTARI: - case OP_MINSTAR: + case OP_MINSTAR: case OP_MINSTARI: case OP_NOTMINSTAR: case OP_NOTMINSTARI: - case OP_POSSTAR: + case OP_POSSTAR: case OP_POSSTARI: case OP_NOTPOSSTAR: case OP_NOTPOSSTARI: - case OP_PLUS: + case OP_PLUS: case OP_PLUSI: case OP_NOTPLUS: case OP_NOTPLUSI: - case OP_MINPLUS: + case OP_MINPLUS: case OP_MINPLUSI: case OP_NOTMINPLUS: case OP_NOTMINPLUSI: - case OP_POSPLUS: + case OP_POSPLUS: case OP_POSPLUSI: case OP_NOTPOSPLUS: case OP_NOTPOSPLUSI: - case OP_QUERY: + case OP_QUERY: case OP_QUERYI: case OP_NOTQUERY: case OP_NOTQUERYI: - case OP_MINQUERY: + case OP_MINQUERY: case OP_MINQUERYI: case OP_NOTMINQUERY: case OP_NOTMINQUERYI: - case OP_POSQUERY: + case OP_POSQUERY: case OP_POSQUERYI: case OP_NOTPOSQUERY: case OP_NOTPOSQUERYI: if (HAS_EXTRALEN(code[-1])) code += GET_EXTRALEN(code[-1]); - break; - } + break; + } #else (void)(utf); /* Keep compiler happy by referencing function argument */ -#endif - } - } -} - - - -/************************************************* -* Scan compiled branch for non-emptiness * -*************************************************/ - -/* This function scans through a branch of a compiled pattern to see whether it -can match the empty string or not. It is called from could_be_empty() -below and from compile_branch() when checking for an unlimited repeat of a -group that can match nothing. Note that first_significant_code() skips over -backward and negative forward assertions when its final argument is TRUE. If we -hit an unclosed bracket, we return "empty" - this means we've struck an inner -bracket whose current branch will already have been scanned. - -Arguments: - code points to start of search - endcode points to where to stop +#endif + } + } +} + + + +/************************************************* +* Scan compiled branch for non-emptiness * +*************************************************/ + +/* This function scans through a branch of a compiled pattern to see whether it +can match the empty string or not. It is called from could_be_empty() +below and from compile_branch() when checking for an unlimited repeat of a +group that can match nothing. Note that first_significant_code() skips over +backward and negative forward assertions when its final argument is TRUE. If we +hit an unclosed bracket, we return "empty" - this means we've struck an inner +bracket whose current branch will already have been scanned. + +Arguments: + code points to start of search + endcode points to where to stop utf TRUE if in UTF-8 / UTF-16 / UTF-32 mode cd contains pointers to tables etc. recurses chain of recurse_check to catch mutual recursion - -Returns: TRUE if what is matched could be empty -*/ - -static BOOL + +Returns: TRUE if what is matched could be empty +*/ + +static BOOL could_be_empty_branch(const pcre_uchar *code, const pcre_uchar *endcode, BOOL utf, compile_data *cd, recurse_check *recurses) -{ +{ register pcre_uchar c; recurse_check this_recurse; for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE); - code < endcode; + code < endcode; code = first_significant_code(code + PRIV(OP_lengths)[c], TRUE)) - { + { const pcre_uchar *ccode; - - c = *code; - - /* Skip over forward assertions; the other assertions are skipped by - first_significant_code() with a TRUE final argument. */ - - if (c == OP_ASSERT) - { - do code += GET(code, 1); while (*code == OP_ALT); - c = *code; - continue; - } - + + c = *code; + + /* Skip over forward assertions; the other assertions are skipped by + first_significant_code() with a TRUE final argument. */ + + if (c == OP_ASSERT) + { + do code += GET(code, 1); while (*code == OP_ALT); + c = *code; + continue; + } + /* For a recursion/subroutine call, if its end has been reached, which implies a backward reference subroutine call, we can scan it. If it's a forward reference subroutine call, we can't. To detect forward reference @@ -2480,17 +2480,17 @@ for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE); continue; } - /* Groups with zero repeats can of course be empty; skip them. */ - + /* Groups with zero repeats can of course be empty; skip them. */ + if (c == OP_BRAZERO || c == OP_BRAMINZERO || c == OP_SKIPZERO || c == OP_BRAPOSZERO) - { + { code += PRIV(OP_lengths)[c]; - do code += GET(code, 1); while (*code == OP_ALT); - c = *code; - continue; - } - + do code += GET(code, 1); while (*code == OP_ALT); + c = *code; + continue; + } + /* A nested group that is already marked as "could be empty" can just be skipped. */ @@ -2502,24 +2502,24 @@ for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE); continue; } - /* For other groups, scan the branches. */ - + /* For other groups, scan the branches. */ + if (c == OP_BRA || c == OP_BRAPOS || c == OP_CBRA || c == OP_CBRAPOS || c == OP_ONCE || c == OP_ONCE_NC || c == OP_COND || c == OP_SCOND) - { - BOOL empty_branch; - if (GET(code, 1) == 0) return TRUE; /* Hit unclosed bracket */ - + { + BOOL empty_branch; + if (GET(code, 1) == 0) return TRUE; /* Hit unclosed bracket */ + /* If a conditional group has only one branch, there is a second, implied, empty branch, so just skip over the conditional, because it could be empty. Otherwise, scan the individual branches of the group. */ - + if (c == OP_COND && code[GET(code, 1)] != OP_ALT) code += GET(code, 1); else - { + { empty_branch = FALSE; do { @@ -2529,176 +2529,176 @@ for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE); } while (*code == OP_ALT); if (!empty_branch) return FALSE; /* All branches are non-empty */ - } - - c = *code; - continue; - } - - /* Handle the other opcodes */ - - switch (c) - { - /* Check for quantifiers after a class. XCLASS is used for classes that - cannot be represented just by a bit map. This includes negated single + } + + c = *code; + continue; + } + + /* Handle the other opcodes */ + + switch (c) + { + /* Check for quantifiers after a class. XCLASS is used for classes that + cannot be represented just by a bit map. This includes negated single high-valued characters. The length in PRIV(OP_lengths)[] is zero; the - actual length is stored in the compiled code, so we must update "code" - here. */ - + actual length is stored in the compiled code, so we must update "code" + here. */ + #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 - case OP_XCLASS: - ccode = code += GET(code, 1); - goto CHECK_CLASS_REPEAT; -#endif - - case OP_CLASS: - case OP_NCLASS: + case OP_XCLASS: + ccode = code += GET(code, 1); + goto CHECK_CLASS_REPEAT; +#endif + + case OP_CLASS: + case OP_NCLASS: ccode = code + PRIV(OP_lengths)[OP_CLASS]; - + #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 - CHECK_CLASS_REPEAT: -#endif - - switch (*ccode) - { - case OP_CRSTAR: /* These could be empty; continue */ - case OP_CRMINSTAR: - case OP_CRQUERY: - case OP_CRMINQUERY: + CHECK_CLASS_REPEAT: +#endif + + switch (*ccode) + { + case OP_CRSTAR: /* These could be empty; continue */ + case OP_CRMINSTAR: + case OP_CRQUERY: + case OP_CRMINQUERY: case OP_CRPOSSTAR: case OP_CRPOSQUERY: - break; - - default: /* Non-repeat => class must match */ - case OP_CRPLUS: /* These repeats aren't empty */ - case OP_CRMINPLUS: + break; + + default: /* Non-repeat => class must match */ + case OP_CRPLUS: /* These repeats aren't empty */ + case OP_CRMINPLUS: case OP_CRPOSPLUS: - return FALSE; - - case OP_CRRANGE: - case OP_CRMINRANGE: + return FALSE; + + case OP_CRRANGE: + case OP_CRMINRANGE: case OP_CRPOSRANGE: - if (GET2(ccode, 1) > 0) return FALSE; /* Minimum > 0 */ - break; - } - break; - - /* Opcodes that must match a character */ - + if (GET2(ccode, 1) > 0) return FALSE; /* Minimum > 0 */ + break; + } + break; + + /* Opcodes that must match a character */ + case OP_ANY: case OP_ALLANY: case OP_ANYBYTE: - case OP_PROP: - case OP_NOTPROP: + case OP_PROP: + case OP_NOTPROP: case OP_ANYNL: case OP_NOT_HSPACE: case OP_HSPACE: case OP_NOT_VSPACE: case OP_VSPACE: - case OP_EXTUNI: + case OP_EXTUNI: - case OP_NOT_DIGIT: - case OP_DIGIT: - case OP_NOT_WHITESPACE: - case OP_WHITESPACE: - case OP_NOT_WORDCHAR: - case OP_WORDCHAR: + case OP_NOT_DIGIT: + case OP_DIGIT: + case OP_NOT_WHITESPACE: + case OP_WHITESPACE: + case OP_NOT_WORDCHAR: + case OP_WORDCHAR: - case OP_CHAR: + case OP_CHAR: case OP_CHARI: - case OP_NOT: + case OP_NOT: case OP_NOTI: - case OP_PLUS: + case OP_PLUS: case OP_PLUSI: - case OP_MINPLUS: + case OP_MINPLUS: case OP_MINPLUSI: - case OP_NOTPLUS: + case OP_NOTPLUS: case OP_NOTPLUSI: - case OP_NOTMINPLUS: + case OP_NOTMINPLUS: case OP_NOTMINPLUSI: case OP_POSPLUS: case OP_POSPLUSI: - case OP_NOTPOSPLUS: + case OP_NOTPOSPLUS: case OP_NOTPOSPLUSI: case OP_EXACT: case OP_EXACTI: - case OP_NOTEXACT: + case OP_NOTEXACT: case OP_NOTEXACTI: - case OP_TYPEPLUS: - case OP_TYPEMINPLUS: - case OP_TYPEPOSPLUS: - case OP_TYPEEXACT: - - return FALSE; - - /* These are going to continue, as they may be empty, but we have to - fudge the length for the \p and \P cases. */ - - case OP_TYPESTAR: - case OP_TYPEMINSTAR: - case OP_TYPEPOSSTAR: - case OP_TYPEQUERY: - case OP_TYPEMINQUERY: - case OP_TYPEPOSQUERY: - if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2; - break; - - /* Same for these */ - - case OP_TYPEUPTO: - case OP_TYPEMINUPTO: - case OP_TYPEPOSUPTO: + case OP_TYPEPLUS: + case OP_TYPEMINPLUS: + case OP_TYPEPOSPLUS: + case OP_TYPEEXACT: + + return FALSE; + + /* These are going to continue, as they may be empty, but we have to + fudge the length for the \p and \P cases. */ + + case OP_TYPESTAR: + case OP_TYPEMINSTAR: + case OP_TYPEPOSSTAR: + case OP_TYPEQUERY: + case OP_TYPEMINQUERY: + case OP_TYPEPOSQUERY: + if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2; + break; + + /* Same for these */ + + case OP_TYPEUPTO: + case OP_TYPEMINUPTO: + case OP_TYPEPOSUPTO: if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP) code += 2; - break; - - /* End of branch */ - - case OP_KET: - case OP_KETRMAX: - case OP_KETRMIN: + break; + + /* End of branch */ + + case OP_KET: + case OP_KETRMAX: + case OP_KETRMIN: case OP_KETRPOS: - case OP_ALT: - return TRUE; - - /* In UTF-8 mode, STAR, MINSTAR, POSSTAR, QUERY, MINQUERY, POSQUERY, UPTO, + case OP_ALT: + return TRUE; + + /* In UTF-8 mode, STAR, MINSTAR, POSSTAR, QUERY, MINQUERY, POSQUERY, UPTO, MINUPTO, and POSUPTO and their caseless and negative versions may be followed by a multibyte character. */ - + #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 - case OP_STAR: + case OP_STAR: case OP_STARI: case OP_NOTSTAR: case OP_NOTSTARI: - case OP_MINSTAR: + case OP_MINSTAR: case OP_MINSTARI: case OP_NOTMINSTAR: case OP_NOTMINSTARI: - case OP_POSSTAR: + case OP_POSSTAR: case OP_POSSTARI: case OP_NOTPOSSTAR: case OP_NOTPOSSTARI: - case OP_QUERY: + case OP_QUERY: case OP_QUERYI: case OP_NOTQUERY: case OP_NOTQUERYI: - case OP_MINQUERY: + case OP_MINQUERY: case OP_MINQUERYI: case OP_NOTMINQUERY: case OP_NOTMINQUERYI: - case OP_POSQUERY: + case OP_POSQUERY: case OP_POSQUERYI: case OP_NOTPOSQUERY: case OP_NOTPOSQUERYI: @@ -2706,24 +2706,24 @@ for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE); if (utf && HAS_EXTRALEN(code[1])) code += GET_EXTRALEN(code[1]); break; - case OP_UPTO: + case OP_UPTO: case OP_UPTOI: case OP_NOTUPTO: case OP_NOTUPTOI: - case OP_MINUPTO: + case OP_MINUPTO: case OP_MINUPTOI: case OP_NOTMINUPTO: case OP_NOTMINUPTOI: - case OP_POSUPTO: + case OP_POSUPTO: case OP_POSUPTOI: case OP_NOTPOSUPTO: case OP_NOTPOSUPTOI: if (utf && HAS_EXTRALEN(code[1 + IMM2_SIZE])) code += GET_EXTRALEN(code[1 + IMM2_SIZE]); - break; -#endif + break; +#endif /* MARK, and PRUNE/SKIP/THEN with an argument must skip over the argument string. */ @@ -2739,51 +2739,51 @@ for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE); default: break; - } - } - -return TRUE; -} - - - -/************************************************* -* Scan compiled regex for non-emptiness * -*************************************************/ - -/* This function is called to check for left recursive calls. We want to check -the current branch of the current pattern to see if it could match the empty -string. If it could, we must look outwards for branches at other levels, -stopping when we pass beyond the bracket which is the subject of the recursion. + } + } + +return TRUE; +} + + + +/************************************************* +* Scan compiled regex for non-emptiness * +*************************************************/ + +/* This function is called to check for left recursive calls. We want to check +the current branch of the current pattern to see if it could match the empty +string. If it could, we must look outwards for branches at other levels, +stopping when we pass beyond the bracket which is the subject of the recursion. This function is called only during the real compile, not during the pre-compile. - -Arguments: - code points to start of the recursion - endcode points to where to stop (current RECURSE item) - bcptr points to the chain of current (unclosed) branch starts + +Arguments: + code points to start of the recursion + endcode points to where to stop (current RECURSE item) + bcptr points to the chain of current (unclosed) branch starts utf TRUE if in UTF-8 / UTF-16 / UTF-32 mode cd pointers to tables etc - -Returns: TRUE if what is matched could be empty -*/ - -static BOOL + +Returns: TRUE if what is matched could be empty +*/ + +static BOOL could_be_empty(const pcre_uchar *code, const pcre_uchar *endcode, branch_chain *bcptr, BOOL utf, compile_data *cd) -{ +{ while (bcptr != NULL && bcptr->current_branch >= code) - { + { if (!could_be_empty_branch(bcptr->current_branch, endcode, utf, cd, NULL)) return FALSE; - bcptr = bcptr->outer; - } -return TRUE; -} - - - -/************************************************* + bcptr = bcptr->outer; + } +return TRUE; +} + + + +/************************************************* * Base opcode of repeated opcodes * *************************************************/ @@ -3889,29 +3889,29 @@ for (;;) /************************************************* -* Check for POSIX class syntax * -*************************************************/ - -/* This function is called when the sequence "[:" or "[." or "[=" is -encountered in a character class. It checks whether this is followed by a -sequence of characters terminated by a matching ":]" or ".]" or "=]". If we -reach an unescaped ']' without the special preceding character, return FALSE. - -Originally, this function only recognized a sequence of letters between the -terminators, but it seems that Perl recognizes any sequence of characters, -though of course unknown POSIX names are subsequently rejected. Perl gives an -"Unknown POSIX class" error for [:f\oo:] for example, where previously PCRE -didn't consider this to be a POSIX class. Likewise for [:1234:]. - -The problem in trying to be exactly like Perl is in the handling of escapes. We -have to be sure that [abc[:x\]pqr] is *not* treated as containing a POSIX -class, but [abc[:x\]pqr:]] is (so that an error can be generated). The code +* Check for POSIX class syntax * +*************************************************/ + +/* This function is called when the sequence "[:" or "[." or "[=" is +encountered in a character class. It checks whether this is followed by a +sequence of characters terminated by a matching ":]" or ".]" or "=]". If we +reach an unescaped ']' without the special preceding character, return FALSE. + +Originally, this function only recognized a sequence of letters between the +terminators, but it seems that Perl recognizes any sequence of characters, +though of course unknown POSIX names are subsequently rejected. Perl gives an +"Unknown POSIX class" error for [:f\oo:] for example, where previously PCRE +didn't consider this to be a POSIX class. Likewise for [:1234:]. + +The problem in trying to be exactly like Perl is in the handling of escapes. We +have to be sure that [abc[:x\]pqr] is *not* treated as containing a POSIX +class, but [abc[:x\]pqr:]] is (so that an error can be generated). The code below handles the special cases \\ and \], but does not try to do any other escape processing. This makes it different from Perl for cases such as [:l\ower:] where Perl recognizes it as the POSIX class "lower" but PCRE does not recognize "l\ower". This is a lesser evil than not diagnosing bad classes when Perl does, I think. - + A user pointed out that PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the appearance of a nested POSIX class supersedes an apparent external class. For example, [:a[:digit:]b:] matches "a", "b", ":", or @@ -3923,20 +3923,20 @@ example, [:a[:abc]b:] gives unknown POSIX class "[:abc]b:]". However, for seem right at all. PCRE does not allow closing square brackets in POSIX class names. -Arguments: - ptr pointer to the initial [ - endptr where to return the end pointer - -Returns: TRUE or FALSE -*/ - -static BOOL +Arguments: + ptr pointer to the initial [ + endptr where to return the end pointer + +Returns: TRUE or FALSE +*/ + +static BOOL check_posix_syntax(const pcre_uchar *ptr, const pcre_uchar **endptr) -{ +{ pcre_uchar terminator; /* Don't combine these lines; the Solaris cc */ -terminator = *(++ptr); /* compiler warns about "non-constant" initializer. */ +terminator = *(++ptr); /* compiler warns about "non-constant" initializer. */ for (++ptr; *ptr != CHAR_NULL; ptr++) - { + { if (*ptr == CHAR_BACKSLASH && (ptr[1] == CHAR_RIGHT_SQUARE_BRACKET || ptr[1] == CHAR_BACKSLASH)) @@ -3944,107 +3944,107 @@ for (++ptr; *ptr != CHAR_NULL; ptr++) else if ((*ptr == CHAR_LEFT_SQUARE_BRACKET && ptr[1] == terminator) || *ptr == CHAR_RIGHT_SQUARE_BRACKET) return FALSE; else if (*ptr == terminator && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET) - { + { *endptr = ptr; return TRUE; - } - } -return FALSE; -} - - - - -/************************************************* -* Check POSIX class name * -*************************************************/ - -/* This function is called to check the name given in a POSIX-style class entry -such as [:alnum:]. - -Arguments: - ptr points to the first letter - len the length of the name - -Returns: a value representing the name, or -1 if unknown -*/ - -static int + } + } +return FALSE; +} + + + + +/************************************************* +* Check POSIX class name * +*************************************************/ + +/* This function is called to check the name given in a POSIX-style class entry +such as [:alnum:]. + +Arguments: + ptr points to the first letter + len the length of the name + +Returns: a value representing the name, or -1 if unknown +*/ + +static int check_posix_name(const pcre_uchar *ptr, int len) -{ -const char *pn = posix_names; -register int yield = 0; -while (posix_name_lengths[yield] != 0) - { - if (len == posix_name_lengths[yield] && +{ +const char *pn = posix_names; +register int yield = 0; +while (posix_name_lengths[yield] != 0) + { + if (len == posix_name_lengths[yield] && STRNCMP_UC_C8(ptr, pn, (unsigned int)len) == 0) return yield; - pn += posix_name_lengths[yield] + 1; - yield++; - } -return -1; -} - - -/************************************************* -* Adjust OP_RECURSE items in repeated group * -*************************************************/ - -/* OP_RECURSE items contain an offset from the start of the regex to the group -that is referenced. This means that groups can be replicated for fixed -repetition simply by copying (because the recursion is allowed to refer to -earlier groups that are outside the current group). However, when a group is + pn += posix_name_lengths[yield] + 1; + yield++; + } +return -1; +} + + +/************************************************* +* Adjust OP_RECURSE items in repeated group * +*************************************************/ + +/* OP_RECURSE items contain an offset from the start of the regex to the group +that is referenced. This means that groups can be replicated for fixed +repetition simply by copying (because the recursion is allowed to refer to +earlier groups that are outside the current group). However, when a group is optional (i.e. the minimum quantifier is zero), OP_BRAZERO or OP_SKIPZERO is inserted before it, after it has been compiled. This means that any OP_RECURSE items within it that refer to the group itself or any contained groups have to have their offsets adjusted. That one of the jobs of this function. Before it is called, the partially compiled regex must be temporarily terminated with OP_END. - + This function has been extended to cope with forward references for recursions and subroutine calls. It must check the list of such references for the group we are dealing with. If it finds that one of the recursions in the current group is on this list, it does not adjust the value in the reference (which is a group number). After the group has been scanned, all the offsets in the forward reference list for the group are adjusted. - -Arguments: - group points to the start of the group - adjust the amount by which the group is to be moved + +Arguments: + group points to the start of the group + adjust the amount by which the group is to be moved utf TRUE in UTF-8 / UTF-16 / UTF-32 mode - cd contains pointers to tables etc. + cd contains pointers to tables etc. save_hwm_offset the hwm forward reference offset at the start of the group - -Returns: nothing -*/ - -static void + +Returns: nothing +*/ + +static void adjust_recurse(pcre_uchar *group, int adjust, BOOL utf, compile_data *cd, size_t save_hwm_offset) -{ +{ int offset; pcre_uchar *hc; pcre_uchar *ptr = group; - + while ((ptr = (pcre_uchar *)find_recurse(ptr, utf)) != NULL) - { + { for (hc = (pcre_uchar *)cd->start_workspace + save_hwm_offset; hc < cd->hwm; hc += LINK_SIZE) - { + { offset = (int)GET(hc, 0); if (cd->start_code + offset == ptr + 1) break; - } - + } + /* If we have not found this recursion on the forward reference list, adjust the recursion's offset if it's after the start of this group. */ - - if (hc >= cd->hwm) - { + + if (hc >= cd->hwm) + { offset = (int)GET(ptr, 1); - if (cd->start_code + offset >= group) PUT(ptr, 1, offset + adjust); - } - - ptr += 1 + LINK_SIZE; - } + if (cd->start_code + offset >= group) PUT(ptr, 1, offset + adjust); + } + + ptr += 1 + LINK_SIZE; + } /* Now adjust all forward reference offsets for the group. */ @@ -4054,96 +4054,96 @@ for (hc = (pcre_uchar *)cd->start_workspace + save_hwm_offset; hc < cd->hwm; offset = (int)GET(hc, 0); PUT(hc, 0, offset + adjust); } -} - - - -/************************************************* -* Insert an automatic callout point * -*************************************************/ - -/* This function is called when the PCRE_AUTO_CALLOUT option is set, to insert -callout points before each pattern item. - -Arguments: - code current code pointer - ptr current pattern pointer - cd pointers to tables etc - -Returns: new code pointer -*/ - +} + + + +/************************************************* +* Insert an automatic callout point * +*************************************************/ + +/* This function is called when the PCRE_AUTO_CALLOUT option is set, to insert +callout points before each pattern item. + +Arguments: + code current code pointer + ptr current pattern pointer + cd pointers to tables etc + +Returns: new code pointer +*/ + static pcre_uchar * auto_callout(pcre_uchar *code, const pcre_uchar *ptr, compile_data *cd) -{ -*code++ = OP_CALLOUT; -*code++ = 255; +{ +*code++ = OP_CALLOUT; +*code++ = 255; PUT(code, 0, (int)(ptr - cd->start_pattern)); /* Pattern offset */ PUT(code, LINK_SIZE, 0); /* Default length */ return code + 2 * LINK_SIZE; -} - - - -/************************************************* -* Complete a callout item * -*************************************************/ - -/* A callout item contains the length of the next item in the pattern, which -we can't fill in till after we have reached the relevant point. This is used -for both automatic and manual callouts. - -Arguments: - previous_callout points to previous callout item - ptr current pattern pointer - cd pointers to tables etc - -Returns: nothing -*/ - -static void +} + + + +/************************************************* +* Complete a callout item * +*************************************************/ + +/* A callout item contains the length of the next item in the pattern, which +we can't fill in till after we have reached the relevant point. This is used +for both automatic and manual callouts. + +Arguments: + previous_callout points to previous callout item + ptr current pattern pointer + cd pointers to tables etc + +Returns: nothing +*/ + +static void complete_callout(pcre_uchar *previous_callout, const pcre_uchar *ptr, compile_data *cd) -{ +{ int length = (int)(ptr - cd->start_pattern - GET(previous_callout, 2)); -PUT(previous_callout, 2 + LINK_SIZE, length); -} - - - -#ifdef SUPPORT_UCP -/************************************************* -* Get othercase range * -*************************************************/ - -/* This function is passed the start and end of a class range, in UTF-8 mode +PUT(previous_callout, 2 + LINK_SIZE, length); +} + + + +#ifdef SUPPORT_UCP +/************************************************* +* Get othercase range * +*************************************************/ + +/* This function is passed the start and end of a class range, in UTF-8 mode with UCP support. It searches up the characters, looking for ranges of -characters in the "other" case. Each call returns the next one, updating the +characters in the "other" case. Each call returns the next one, updating the start address. A character with multiple other cases is returned on its own with a special return value. - -Arguments: - cptr points to starting character value; updated - d end value - ocptr where to put start of othercase range - odptr where to put end of othercase range - + +Arguments: + cptr points to starting character value; updated + d end value + ocptr where to put start of othercase range + odptr where to put end of othercase range + Yield: -1 when no more 0 when a range is returned >0 the CASESET offset for char with multiple other cases in this case, ocptr contains the original -*/ - +*/ + static int get_othercase_range(pcre_uint32 *cptr, pcre_uint32 d, pcre_uint32 *ocptr, pcre_uint32 *odptr) -{ +{ pcre_uint32 c, othercase, next; unsigned int co; - + /* Find the first character that has an other case. If it has multiple other cases, return its case offset value. */ -for (c = *cptr; c <= d; c++) +for (c = *cptr; c <= d; c++) { if ((co = UCD_CASESET(c)) != 0) { @@ -4153,69 +4153,69 @@ for (c = *cptr; c <= d; c++) } if ((othercase = UCD_OTHERCASE(c)) != c) break; } - + if (c > d) return -1; /* Reached end of range */ - + /* Found a character that has a single other case. Search for the end of the range, which is either the end of the input range, or a character that has zero or more than one other cases. */ -*ocptr = othercase; -next = othercase + 1; - -for (++c; c <= d; c++) - { +*ocptr = othercase; +next = othercase + 1; + +for (++c; c <= d; c++) + { if ((co = UCD_CASESET(c)) != 0 || UCD_OTHERCASE(c) != next) break; - next++; - } - + next++; + } + *odptr = next - 1; /* End of othercase range */ *cptr = c; /* Rest of input range */ return 0; -} -#endif /* SUPPORT_UCP */ - - - -/************************************************* +} +#endif /* SUPPORT_UCP */ + + + +/************************************************* * Add a character or range to a class * -*************************************************/ - +*************************************************/ + /* This function packages up the logic of adding a character or range of characters to a class. The character values in the arguments will be within the valid values for the current mode (8-bit, 16-bit, UTF, etc). This function is mutually recursive with the function immediately below. - -Arguments: + +Arguments: classbits the bit map for characters < 256 uchardptr points to the pointer for extra data options the options word - cd contains pointers to tables etc. + cd contains pointers to tables etc. start start of range character end end of range character - + Returns: the number of < 256 characters added the pointer to extra data is updated -*/ - +*/ + static int add_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr, int options, compile_data *cd, pcre_uint32 start, pcre_uint32 end) -{ +{ pcre_uint32 c; pcre_uint32 classbits_end = (end <= 0xff ? end : 0xff); int n8 = 0; - + /* If caseless matching is required, scan the range and process alternate cases. In Unicode, there are 8-bit characters that have alternate cases that are greater than 255 and vice-versa. Sometimes we can just extend the original range. */ - + if ((options & PCRE_CASELESS) != 0) - { + { #ifdef SUPPORT_UCP if ((options & PCRE_UTF8) != 0) - { + { int rc; pcre_uint32 oc, od; @@ -4223,20 +4223,20 @@ if ((options & PCRE_CASELESS) != 0) c = start; while ((rc = get_othercase_range(&c, end, &oc, &od)) >= 0) - { + { /* Handle a single character that has more than one other case. */ - + if (rc > 0) n8 += add_list_to_class(classbits, uchardptr, options, cd, PRIV(ucd_caseless_sets) + rc, oc); - + /* Do nothing if the other case range is within the original range. */ - + else if (oc >= start && od <= end) continue; - + /* Extend the original range if there is overlap, noting that if oc < c, we can't have od > end because a subrange is always shorter than the basic range. Otherwise, use a recursive call to add the additional range. */ - + else if (oc < start && od >= start - 1) start = oc; /* Extend downwards */ else if (od > end && oc <= end + 1) { @@ -4248,52 +4248,52 @@ if ((options & PCRE_CASELESS) != 0) } else #endif /* SUPPORT_UCP */ - + /* Not UTF-mode, or no UCP */ for (c = start; c <= classbits_end; c++) - { + { SETBIT(classbits, cd->fcc[c]); n8++; - } - } - + } + } + /* Now handle the original range. Adjust the final value according to the bit length - this means that the same lists of (e.g.) horizontal spaces can be used in all cases. */ - + #if defined COMPILE_PCRE8 #ifdef SUPPORT_UTF if ((options & PCRE_UTF8) == 0) #endif if (end > 0xff) end = 0xff; - + #elif defined COMPILE_PCRE16 #ifdef SUPPORT_UTF if ((options & PCRE_UTF16) == 0) #endif if (end > 0xffff) end = 0xffff; - + #endif /* COMPILE_PCRE[8|16] */ - + /* Use the bitmap for characters < 256. Otherwise use extra data.*/ - + for (c = start; c <= classbits_end; c++) - { + { /* Regardless of start, c will always be <= 255. */ SETBIT(classbits, c); n8++; } - + #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 if (start <= 0xff) start = 0xff + 1; - + if (end >= start) { pcre_uchar *uchardata = *uchardptr; #ifdef SUPPORT_UTF if ((options & PCRE_UTF8) != 0) /* All UTFs use the same flag bit */ - { + { if (start < end) { *uchardata++ = XCL_RANGE; @@ -4305,49 +4305,49 @@ if (end >= start) *uchardata++ = XCL_SINGLE; uchardata += PRIV(ord2utf)(start, uchardata); } - } - else + } + else #endif /* SUPPORT_UTF */ - + /* Without UTF support, character values are constrained by the bit length, and can only be > 256 for 16-bit and 32-bit libraries. */ - + #ifdef COMPILE_PCRE8 {} -#else +#else if (start < end) - { + { *uchardata++ = XCL_RANGE; *uchardata++ = start; *uchardata++ = end; - } + } else if (start == end) - { + { *uchardata++ = XCL_SINGLE; *uchardata++ = start; - } + } #endif - + *uchardptr = uchardata; /* Updata extra data pointer */ - } + } #endif /* SUPPORT_UTF || !COMPILE_PCRE8 */ - + return n8; /* Number of 8-bit characters */ } - - - - + + + + /************************************************* * Add a list of characters to a class * *************************************************/ - + /* This function is used for adding a list of case-equivalent characters to a class, and also for adding a list of horizontal or vertical whitespace. If the list is in order (which it should be), ranges of characters are detected and handled appropriately. This function is mutually recursive with the function above. - + Arguments: classbits the bit map for characters < 256 uchardptr points to the pointer for extra data @@ -4357,11 +4357,11 @@ Arguments: except character to omit; this is used when adding lists of case-equivalent characters to avoid including the one we already know about - + Returns: the number of < 256 characters added the pointer to extra data is updated */ - + static int add_list_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr, int options, compile_data *cd, const pcre_uint32 *p, unsigned int except) @@ -4374,32 +4374,32 @@ while (p[0] < NOTACHAR) { while(p[n+1] == p[0] + n + 1) n++; n8 += add_to_class(classbits, uchardptr, options, cd, p[0], p[n]); - } + } p += n + 1; } return n8; } - - - + + + /************************************************* * Add characters not in a list to a class * *************************************************/ - + /* This function is used for adding the complement of a list of horizontal or vertical whitespace to a class. The list must be in order. - + Arguments: classbits the bit map for characters < 256 uchardptr points to the pointer for extra data options the options word cd contains pointers to tables etc. p points to row of 32-bit values, terminated by NOTACHAR - + Returns: the number of < 256 characters added the pointer to extra data is updated */ - + static int add_not_list_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr, int options, compile_data *cd, const pcre_uint32 *p) @@ -4414,23 +4414,23 @@ while (p[0] < NOTACHAR) n8 += add_to_class(classbits, uchardptr, options, cd, p[0] + 1, (p[1] == NOTACHAR) ? (utf ? 0x10ffffu : 0xffffffffu) : p[1] - 1); p++; - } + } return n8; -} - - - -/************************************************* -* Compile one branch * -*************************************************/ - -/* Scan the pattern, compiling it into the a vector. If the options are -changed during the branch, the pointer is used to change the external options -bits. This function is used during the pre-compile phase when we are trying -to find out the amount of memory needed, as well as during the real compile -phase. The value of lengthptr distinguishes the two phases. - -Arguments: +} + + + +/************************************************* +* Compile one branch * +*************************************************/ + +/* Scan the pattern, compiling it into the a vector. If the options are +changed during the branch, the pointer is used to change the external options +bits. This function is used during the pre-compile phase when we are trying +to find out the amount of memory needed, as well as during the real compile +phase. The value of lengthptr distinguishes the two phases. + +Arguments: optionsptr pointer to the option bits codeptr points to the pointer to the current code point ptrptr points to the current pattern pointer @@ -4444,38 +4444,38 @@ Arguments: cd contains pointers to tables etc. lengthptr NULL during the real compile phase points to length accumulator during pre-compile phase - + Returns: TRUE on success FALSE, with *errorcodeptr set non-zero on error -*/ - -static BOOL +*/ + +static BOOL compile_branch(int *optionsptr, pcre_uchar **codeptr, const pcre_uchar **ptrptr, int *errorcodeptr, pcre_uint32 *firstcharptr, pcre_int32 *firstcharflagsptr, pcre_uint32 *reqcharptr, pcre_int32 *reqcharflagsptr, branch_chain *bcptr, int cond_depth, - compile_data *cd, int *lengthptr) -{ -int repeat_type, op_type; -int repeat_min = 0, repeat_max = 0; /* To please picky compilers */ -int bravalue = 0; -int greedy_default, greedy_non_default; + compile_data *cd, int *lengthptr) +{ +int repeat_type, op_type; +int repeat_min = 0, repeat_max = 0; /* To please picky compilers */ +int bravalue = 0; +int greedy_default, greedy_non_default; pcre_uint32 firstchar, reqchar; pcre_int32 firstcharflags, reqcharflags; pcre_uint32 zeroreqchar, zerofirstchar; pcre_int32 zeroreqcharflags, zerofirstcharflags; pcre_int32 req_caseopt, reqvary, tempreqvary; int options = *optionsptr; /* May change dynamically */ -int after_manual_callout = 0; -int length_prevgroup = 0; +int after_manual_callout = 0; +int length_prevgroup = 0; register pcre_uint32 c; int escape; register pcre_uchar *code = *codeptr; pcre_uchar *last_code = code; pcre_uchar *orig_code = code; pcre_uchar *tempcode; -BOOL inescq = FALSE; +BOOL inescq = FALSE; BOOL groupsetfirstchar = FALSE; const pcre_uchar *ptr = *ptrptr; const pcre_uchar *tempptr; @@ -4484,7 +4484,7 @@ pcre_uchar *previous = NULL; pcre_uchar *previous_callout = NULL; size_t item_hwm_offset = 0; pcre_uint8 classbits[32]; - + /* We can fish out the UTF-8 setting once and for all into a BOOL, but we must not do this for other options (e.g. PCRE_EXTENDED) because they may change dynamically as we process the pattern. */ @@ -4495,10 +4495,10 @@ BOOL utf = (options & PCRE_UTF8) != 0; #ifndef COMPILE_PCRE32 pcre_uchar utf_chars[6]; #endif -#else +#else BOOL utf = FALSE; -#endif - +#endif + /* Helper variables for OP_XCLASS opcode (for characters > 255). We define class_uchardata always so that it can be passed to add_to_class() always, though it will not be used in non-UTF 8-bit cases. This avoids having to supply @@ -4511,70 +4511,70 @@ pcre_uchar *class_uchardata_base; #endif #ifdef PCRE_DEBUG -if (lengthptr != NULL) DPRINTF((">> start branch\n")); -#endif - -/* Set up the default and non-default settings for greediness */ - -greedy_default = ((options & PCRE_UNGREEDY) != 0); -greedy_non_default = greedy_default ^ 1; - -/* Initialize no first byte, no required byte. REQ_UNSET means "no char -matching encountered yet". It gets changed to REQ_NONE if we hit something that +if (lengthptr != NULL) DPRINTF((">> start branch\n")); +#endif + +/* Set up the default and non-default settings for greediness */ + +greedy_default = ((options & PCRE_UNGREEDY) != 0); +greedy_non_default = greedy_default ^ 1; + +/* Initialize no first byte, no required byte. REQ_UNSET means "no char +matching encountered yet". It gets changed to REQ_NONE if we hit something that matches a non-fixed char first char; reqchar just remains unset if we never -find one. - -When we hit a repeat whose minimum is zero, we may have to adjust these values -to take the zero repeat into account. This is implemented by setting them to +find one. + +When we hit a repeat whose minimum is zero, we may have to adjust these values +to take the zero repeat into account. This is implemented by setting them to zerofirstbyte and zeroreqchar when such a repeat is encountered. The individual -item types that can be repeated set these backoff variables appropriately. */ - +item types that can be repeated set these backoff variables appropriately. */ + firstchar = reqchar = zerofirstchar = zeroreqchar = 0; firstcharflags = reqcharflags = zerofirstcharflags = zeroreqcharflags = REQ_UNSET; - + /* The variable req_caseopt contains either the REQ_CASELESS value or zero, according to the current setting of the caseless flag. The REQ_CASELESS leaves the lower 28 bit empty. It is added into the firstchar or reqchar variables to record the case status of the value. This is used only for ASCII characters. */ - + req_caseopt = ((options & PCRE_CASELESS) != 0)? REQ_CASELESS:0; - -/* Switch on next character until the end of the branch */ - -for (;; ptr++) - { - BOOL negate_class; - BOOL should_flip_negation; - BOOL possessive_quantifier; - BOOL is_quantifier; - BOOL is_recurse; - BOOL reset_bracount; + +/* Switch on next character until the end of the branch */ + +for (;; ptr++) + { + BOOL negate_class; + BOOL should_flip_negation; + BOOL possessive_quantifier; + BOOL is_quantifier; + BOOL is_recurse; + BOOL reset_bracount; int class_has_8bitchar; int class_one_char; #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 BOOL xclass_has_prop; #endif - int newoptions; - int recno; - int refsign; - int skipbytes; + int newoptions; + int recno; + int refsign; + int skipbytes; pcre_uint32 subreqchar, subfirstchar; pcre_int32 subreqcharflags, subfirstcharflags; - int terminator; + int terminator; unsigned int mclength; unsigned int tempbracount; pcre_uint32 ec; pcre_uchar mcbuffer[8]; - + /* Come here to restart the loop without advancing the pointer. */ - + REDO_LOOP: /* Get next character in the pattern */ - c = *ptr; - + c = *ptr; + /* If we are at the end of a nested substitution, revert to the outer level string. Nesting only happens one level deep. */ @@ -4585,122 +4585,122 @@ for (;; ptr++) c = *ptr; } - /* If we are in the pre-compile phase, accumulate the length used for the - previous cycle of this loop. */ - - if (lengthptr != NULL) - { + /* If we are in the pre-compile phase, accumulate the length used for the + previous cycle of this loop. */ + + if (lengthptr != NULL) + { #ifdef PCRE_DEBUG - if (code > cd->hwm) cd->hwm = code; /* High water info */ -#endif + if (code > cd->hwm) cd->hwm = code; /* High water info */ +#endif if (code > cd->start_workspace + cd->workspace_size - WORK_SIZE_SAFETY_MARGIN) /* Check for overrun */ - { + { *errorcodeptr = (code >= cd->start_workspace + cd->workspace_size)? ERR52 : ERR87; - goto FAILED; - } - - /* There is at least one situation where code goes backwards: this is the - case of a zero quantifier after a class (e.g. [ab]{0}). At compile time, - the class is simply eliminated. However, it is created first, so we have to - allow memory for it. Therefore, don't ever reduce the length at this point. - */ - - if (code < last_code) code = last_code; - - /* Paranoid check for integer overflow */ - - if (OFLOW_MAX - *lengthptr < code - last_code) - { - *errorcodeptr = ERR20; - goto FAILED; - } - + goto FAILED; + } + + /* There is at least one situation where code goes backwards: this is the + case of a zero quantifier after a class (e.g. [ab]{0}). At compile time, + the class is simply eliminated. However, it is created first, so we have to + allow memory for it. Therefore, don't ever reduce the length at this point. + */ + + if (code < last_code) code = last_code; + + /* Paranoid check for integer overflow */ + + if (OFLOW_MAX - *lengthptr < code - last_code) + { + *errorcodeptr = ERR20; + goto FAILED; + } + *lengthptr += (int)(code - last_code); DPRINTF(("length=%d added %d c=%c (0x%x)\n", *lengthptr, (int)(code - last_code), c, c)); - - /* If "previous" is set and it is not at the start of the work space, move - it back to there, in order to avoid filling up the work space. Otherwise, - if "previous" is NULL, reset the current code pointer to the start. */ - - if (previous != NULL) - { - if (previous > orig_code) - { + + /* If "previous" is set and it is not at the start of the work space, move + it back to there, in order to avoid filling up the work space. Otherwise, + if "previous" is NULL, reset the current code pointer to the start. */ + + if (previous != NULL) + { + if (previous > orig_code) + { memmove(orig_code, previous, IN_UCHARS(code - previous)); - code -= previous - orig_code; - previous = orig_code; - } - } - else code = orig_code; - - /* Remember where this code item starts so we can pick up the length - next time round. */ - - last_code = code; - } - - /* In the real compile phase, just check the workspace used by the forward - reference list. */ - + code -= previous - orig_code; + previous = orig_code; + } + } + else code = orig_code; + + /* Remember where this code item starts so we can pick up the length + next time round. */ + + last_code = code; + } + + /* In the real compile phase, just check the workspace used by the forward + reference list. */ + else if (cd->hwm > cd->start_workspace + cd->workspace_size) - { - *errorcodeptr = ERR52; - goto FAILED; - } - + { + *errorcodeptr = ERR52; + goto FAILED; + } + /* If in \Q...\E, check for the end; if not, we have a literal. Otherwise an isolated \E is ignored. */ - + if (c != CHAR_NULL) - { + { if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E) - { - inescq = FALSE; - ptr++; - continue; - } + { + inescq = FALSE; + ptr++; + continue; + } else if (inescq) - { - if (previous_callout != NULL) - { - if (lengthptr == NULL) /* Don't attempt in pre-compile phase */ - complete_callout(previous_callout, ptr, cd); - previous_callout = NULL; - } - if ((options & PCRE_AUTO_CALLOUT) != 0) - { - previous_callout = code; - code = auto_callout(code, ptr, cd); - } - goto NORMAL_CHAR; - } - + { + if (previous_callout != NULL) + { + if (lengthptr == NULL) /* Don't attempt in pre-compile phase */ + complete_callout(previous_callout, ptr, cd); + previous_callout = NULL; + } + if ((options & PCRE_AUTO_CALLOUT) != 0) + { + previous_callout = code; + code = auto_callout(code, ptr, cd); + } + goto NORMAL_CHAR; + } + /* Check for the start of a \Q...\E sequence. We must do this here rather than later in case it is immediately followed by \E, which turns it into a "do nothing" sequence. */ - + if (c == CHAR_BACKSLASH && ptr[1] == CHAR_Q) { inescq = TRUE; ptr++; continue; } - } - + } + /* In extended mode, skip white space and comments. */ - - if ((options & PCRE_EXTENDED) != 0) - { + + if ((options & PCRE_EXTENDED) != 0) + { const pcre_uchar *wscptr = ptr; while (MAX_255(c) && (cd->ctypes[c] & ctype_space) != 0) c = *(++ptr); if (c == CHAR_NUMBER_SIGN) - { + { ptr++; while (*ptr != CHAR_NULL) - { + { if (IS_NEWLINE(ptr)) /* For non-fixed-length newline cases, */ { /* IS_NEWLINE sets cd->nllen. */ ptr += cd->nllen; @@ -4710,9 +4710,9 @@ for (;; ptr++) #ifdef SUPPORT_UTF if (utf) FORWARDCHAR(ptr); #endif - } + } } - + /* If we skipped any characters, restart the loop. Otherwise, we didn't see a comment. */ @@ -4732,12 +4732,12 @@ for (;; ptr++) { *errorcodeptr = ERR18; goto FAILED; - } + } continue; - } - + } + /* See if the next thing is a quantifier. */ - + is_quantifier = c == CHAR_ASTERISK || c == CHAR_PLUS || c == CHAR_QUESTION_MARK || (c == CHAR_LEFT_CURLY_BRACKET && is_counted_repeat(ptr+1)); @@ -4747,7 +4747,7 @@ for (;; ptr++) if (!is_quantifier && previous_callout != NULL && nestptr == NULL && after_manual_callout-- <= 0) - { + { if (lengthptr == NULL) /* Don't attempt in pre-compile phase */ complete_callout(previous_callout, ptr, cd); previous_callout = NULL; @@ -4758,15 +4758,15 @@ for (;; ptr++) if ((options & PCRE_AUTO_CALLOUT) != 0 && !is_quantifier && nestptr == NULL) { - previous_callout = code; - code = auto_callout(code, ptr, cd); - } - + previous_callout = code; + code = auto_callout(code, ptr, cd); + } + /* Process the next pattern item. */ - switch(c) - { - /* ===================================================================*/ + switch(c) + { + /* ===================================================================*/ case CHAR_NULL: /* The branch terminates at string end */ case CHAR_VERTICAL_LINE: /* or | or ) */ case CHAR_RIGHT_PARENTHESIS: @@ -4774,68 +4774,68 @@ for (;; ptr++) *firstcharflagsptr = firstcharflags; *reqcharptr = reqchar; *reqcharflagsptr = reqcharflags; - *codeptr = code; - *ptrptr = ptr; - if (lengthptr != NULL) - { - if (OFLOW_MAX - *lengthptr < code - last_code) - { - *errorcodeptr = ERR20; - goto FAILED; - } + *codeptr = code; + *ptrptr = ptr; + if (lengthptr != NULL) + { + if (OFLOW_MAX - *lengthptr < code - last_code) + { + *errorcodeptr = ERR20; + goto FAILED; + } *lengthptr += (int)(code - last_code); /* To include callout length */ - DPRINTF((">> end branch\n")); - } - return TRUE; - - - /* ===================================================================*/ - /* Handle single-character metacharacters. In multiline mode, ^ disables - the setting of any following char as a first character. */ - + DPRINTF((">> end branch\n")); + } + return TRUE; + + + /* ===================================================================*/ + /* Handle single-character metacharacters. In multiline mode, ^ disables + the setting of any following char as a first character. */ + case CHAR_CIRCUMFLEX_ACCENT: previous = NULL; - if ((options & PCRE_MULTILINE) != 0) - { + if ((options & PCRE_MULTILINE) != 0) + { if (firstcharflags == REQ_UNSET) zerofirstcharflags = firstcharflags = REQ_NONE; *code++ = OP_CIRCM; - } + } else *code++ = OP_CIRC; - break; - + break; + case CHAR_DOLLAR_SIGN: - previous = NULL; + previous = NULL; *code++ = ((options & PCRE_MULTILINE) != 0)? OP_DOLLM : OP_DOLL; - break; - - /* There can never be a first char if '.' is first, whatever happens about + break; + + /* There can never be a first char if '.' is first, whatever happens about repeats. The value of reqchar doesn't change either. */ - + case CHAR_DOT: if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE; zerofirstchar = firstchar; zerofirstcharflags = firstcharflags; zeroreqchar = reqchar; zeroreqcharflags = reqcharflags; - previous = code; + previous = code; item_hwm_offset = cd->hwm - cd->start_workspace; *code++ = ((options & PCRE_DOTALL) != 0)? OP_ALLANY: OP_ANY; - break; - - - /* ===================================================================*/ - /* Character classes. If the included characters are all < 256, we build a - 32-byte bitmap of the permitted characters, except in the special case - where there is only one such character. For negated classes, we build the - map as usual, then invert it at the end. However, we use a different opcode - so that data characters > 255 can be handled correctly. - - If the class contains characters outside the 0-255 range, a different - opcode is compiled. It may optionally have a bit map for characters < 256, - but those above are are explicitly listed afterwards. A flag byte tells - whether the bitmap is present, and whether this is a negated class or not. - + break; + + + /* ===================================================================*/ + /* Character classes. If the included characters are all < 256, we build a + 32-byte bitmap of the permitted characters, except in the special case + where there is only one such character. For negated classes, we build the + map as usual, then invert it at the end. However, we use a different opcode + so that data characters > 255 can be handled correctly. + + If the class contains characters outside the 0-255 range, a different + opcode is compiled. It may optionally have a bit map for characters < 256, + but those above are are explicitly listed afterwards. A flag byte tells + whether the bitmap is present, and whether this is a negated class or not. + In JavaScript compatibility mode, an isolated ']' causes an error. In default (Perl) mode, it is treated as a data character. */ @@ -4870,42 +4870,42 @@ for (;; ptr++) /* Handle a real character class. */ - previous = code; + previous = code; item_hwm_offset = cd->hwm - cd->start_workspace; - - /* PCRE supports POSIX class stuff inside a class. Perl gives an error if - they are encountered at the top level, so we'll do that too. */ - + + /* PCRE supports POSIX class stuff inside a class. Perl gives an error if + they are encountered at the top level, so we'll do that too. */ + if ((ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT || ptr[1] == CHAR_EQUALS_SIGN) && - check_posix_syntax(ptr, &tempptr)) - { + check_posix_syntax(ptr, &tempptr)) + { *errorcodeptr = (ptr[1] == CHAR_COLON)? ERR13 : ERR31; - goto FAILED; - } - - /* If the first character is '^', set the negation flag and skip it. Also, - if the first few characters (either before or after ^) are \Q\E or \E we - skip them too. This makes for compatibility with Perl. */ - - negate_class = FALSE; - for (;;) - { - c = *(++ptr); + goto FAILED; + } + + /* If the first character is '^', set the negation flag and skip it. Also, + if the first few characters (either before or after ^) are \Q\E or \E we + skip them too. This makes for compatibility with Perl. */ + + negate_class = FALSE; + for (;;) + { + c = *(++ptr); if (c == CHAR_BACKSLASH) - { + { if (ptr[1] == CHAR_E) ptr++; else if (STRNCMP_UC_C8(ptr + 1, STR_Q STR_BACKSLASH STR_E, 3) == 0) ptr += 3; else break; - } + } else if (!negate_class && c == CHAR_CIRCUMFLEX_ACCENT) - negate_class = TRUE; - else break; - } - + negate_class = TRUE; + else break; + } + /* Empty classes are allowed in JavaScript compatibility mode. Otherwise, an initial ']' is taken as a data character -- the code below handles that. In JS mode, [] must always fail, so generate OP_FAIL, whereas @@ -4921,21 +4921,21 @@ for (;; ptr++) break; } - /* If a class contains a negative special such as \S, we need to flip the - negation flag at the end, so that support for characters > 255 works - correctly (they are all included in the class). */ - - should_flip_negation = FALSE; - + /* If a class contains a negative special such as \S, we need to flip the + negation flag at the end, so that support for characters > 255 works + correctly (they are all included in the class). */ + + should_flip_negation = FALSE; + /* Extended class (xclass) will be used when characters > 255 might match. */ - + #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 xclass = FALSE; class_uchardata = code + LINK_SIZE + 2; /* For XCLASS items */ class_uchardata_base = class_uchardata; /* Save the start */ #endif - + /* For optimization purposes, we track some properties of the class: class_has_8bitchar will be non-zero if the class contains at least one < 256 character; class_one_char will be 1 if the class contains just one @@ -4948,28 +4948,28 @@ for (;; ptr++) xclass_has_prop = FALSE; #endif - /* Initialize the 32-char bit map to all zeros. We build the map in a + /* Initialize the 32-char bit map to all zeros. We build the map in a temporary bit of memory, in case the class contains fewer than two 8-bit characters because in that case the compiled code doesn't use the bit map. */ - + memset(classbits, 0, 32 * sizeof(pcre_uint8)); - - /* Process characters until ] is reached. By writing this as a "do" it - means that an initial ] is taken as a data character. At the start of the - loop, c contains the first byte of the character. */ - + + /* Process characters until ] is reached. By writing this as a "do" it + means that an initial ] is taken as a data character. At the start of the + loop, c contains the first byte of the character. */ + if (c != CHAR_NULL) do - { + { const pcre_uchar *oldptr; - + #ifdef SUPPORT_UTF if (utf && HAS_EXTRALEN(c)) - { /* Braces are required because the */ - GETCHARLEN(c, ptr, ptr); /* macro generates multiple statements */ - } -#endif - + { /* Braces are required because the */ + GETCHARLEN(c, ptr, ptr); /* macro generates multiple statements */ + } +#endif + #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 /* In the pre-compile phase, accumulate the length of any extra data and reset the pointer. This is so that very large classes that @@ -4986,67 +4986,67 @@ for (;; ptr++) } #endif - /* Inside \Q...\E everything is literal except \E */ - - if (inescq) - { + /* Inside \Q...\E everything is literal except \E */ + + if (inescq) + { if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E) /* If we are at \E */ - { - inescq = FALSE; /* Reset literal state */ - ptr++; /* Skip the 'E' */ - continue; /* Carry on with next */ - } - goto CHECK_RANGE; /* Could be range if \E follows */ - } - - /* Handle POSIX class names. Perl allows a negation extension of the - form [:^name:]. A square bracket that doesn't match the syntax is - treated as a literal. We also recognize the POSIX constructions - [.ch.] and [=ch=] ("collating elements") and fault them, as Perl - 5.6 and 5.8 do. */ - + { + inescq = FALSE; /* Reset literal state */ + ptr++; /* Skip the 'E' */ + continue; /* Carry on with next */ + } + goto CHECK_RANGE; /* Could be range if \E follows */ + } + + /* Handle POSIX class names. Perl allows a negation extension of the + form [:^name:]. A square bracket that doesn't match the syntax is + treated as a literal. We also recognize the POSIX constructions + [.ch.] and [=ch=] ("collating elements") and fault them, as Perl + 5.6 and 5.8 do. */ + if (c == CHAR_LEFT_SQUARE_BRACKET && (ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT || ptr[1] == CHAR_EQUALS_SIGN) && check_posix_syntax(ptr, &tempptr)) - { - BOOL local_negate = FALSE; - int posix_class, taboffset, tabopt; + { + BOOL local_negate = FALSE; + int posix_class, taboffset, tabopt; register const pcre_uint8 *cbits = cd->cbits; pcre_uint8 pbits[32]; - + if (ptr[1] != CHAR_COLON) - { - *errorcodeptr = ERR31; - goto FAILED; - } - - ptr += 2; + { + *errorcodeptr = ERR31; + goto FAILED; + } + + ptr += 2; if (*ptr == CHAR_CIRCUMFLEX_ACCENT) - { - local_negate = TRUE; - should_flip_negation = TRUE; /* Note negative special */ - ptr++; - } - + { + local_negate = TRUE; + should_flip_negation = TRUE; /* Note negative special */ + ptr++; + } + posix_class = check_posix_name(ptr, (int)(tempptr - ptr)); - if (posix_class < 0) - { - *errorcodeptr = ERR30; - goto FAILED; - } - - /* If matching is caseless, upper and lower are converted to - alpha. This relies on the fact that the class table starts with - alpha, lower, upper as the first 3 entries. */ - - if ((options & PCRE_CASELESS) != 0 && posix_class <= 2) - posix_class = 0; - + if (posix_class < 0) + { + *errorcodeptr = ERR30; + goto FAILED; + } + + /* If matching is caseless, upper and lower are converted to + alpha. This relies on the fact that the class table starts with + alpha, lower, upper as the first 3 entries. */ + + if ((options & PCRE_CASELESS) != 0 && posix_class <= 2) + posix_class = 0; + /* When PCRE_UCP is set, some of the POSIX classes are converted to different escape sequences that use Unicode properties \p or \P. Others that are not available via \p or \P generate XCL_PROP/XCL_NOTPROP directly. */ - + #ifdef SUPPORT_UCP if ((options & PCRE_UCP) != 0) { @@ -5115,91 +5115,91 @@ for (;; ptr++) may be in the main map already. At the end we or the result into the bit map that is being built. */ - posix_class *= 3; - - /* Copy in the first table (always present) */ - - memcpy(pbits, cbits + posix_class_maps[posix_class], + posix_class *= 3; + + /* Copy in the first table (always present) */ + + memcpy(pbits, cbits + posix_class_maps[posix_class], 32 * sizeof(pcre_uint8)); - - /* If there is a second table, add or remove it as required. */ - - taboffset = posix_class_maps[posix_class + 1]; - tabopt = posix_class_maps[posix_class + 2]; - - if (taboffset >= 0) - { - if (tabopt >= 0) - for (c = 0; c < 32; c++) pbits[c] |= cbits[c + taboffset]; - else - for (c = 0; c < 32; c++) pbits[c] &= ~cbits[c + taboffset]; - } - + + /* If there is a second table, add or remove it as required. */ + + taboffset = posix_class_maps[posix_class + 1]; + tabopt = posix_class_maps[posix_class + 2]; + + if (taboffset >= 0) + { + if (tabopt >= 0) + for (c = 0; c < 32; c++) pbits[c] |= cbits[c + taboffset]; + else + for (c = 0; c < 32; c++) pbits[c] &= ~cbits[c + taboffset]; + } + /* Now see if we need to remove any special characters. An option - value of 1 removes vertical space and 2 removes underscore. */ - - if (tabopt < 0) tabopt = -tabopt; - if (tabopt == 1) pbits[1] &= ~0x3c; - else if (tabopt == 2) pbits[11] &= 0x7f; - - /* Add the POSIX table or its complement into the main table that is - being built and we are done. */ - - if (local_negate) - for (c = 0; c < 32; c++) classbits[c] |= ~pbits[c]; - else - for (c = 0; c < 32; c++) classbits[c] |= pbits[c]; - - ptr = tempptr + 1; + value of 1 removes vertical space and 2 removes underscore. */ + + if (tabopt < 0) tabopt = -tabopt; + if (tabopt == 1) pbits[1] &= ~0x3c; + else if (tabopt == 2) pbits[11] &= 0x7f; + + /* Add the POSIX table or its complement into the main table that is + being built and we are done. */ + + if (local_negate) + for (c = 0; c < 32; c++) classbits[c] |= ~pbits[c]; + else + for (c = 0; c < 32; c++) classbits[c] |= pbits[c]; + + ptr = tempptr + 1; /* Every class contains at least one < 256 character. */ class_has_8bitchar = 1; /* Every class contains at least two characters. */ class_one_char = 2; - continue; /* End of POSIX syntax handling */ - } - - /* Backslash may introduce a single character, or it may introduce one - of the specials, which just set a flag. The sequence \b is a special + continue; /* End of POSIX syntax handling */ + } + + /* Backslash may introduce a single character, or it may introduce one + of the specials, which just set a flag. The sequence \b is a special case. Inside a class (and only there) it is treated as backspace. We assume that other escapes have more than one character in them, so speculatively set both class_has_8bitchar and class_one_char bigger than one. Unrecognized escapes fall through and are either treated as literal characters (by default), or are faulted if PCRE_EXTRA is set. */ - + if (c == CHAR_BACKSLASH) - { + { escape = check_escape(&ptr, &ec, errorcodeptr, cd->bracount, options, TRUE); - if (*errorcodeptr != 0) goto FAILED; + if (*errorcodeptr != 0) goto FAILED; if (escape == 0) c = ec; else if (escape == ESC_b) c = CHAR_BS; /* \b is backspace in a class */ else if (escape == ESC_N) /* \N is not supported in a class */ - { + { *errorcodeptr = ERR71; goto FAILED; } else if (escape == ESC_Q) /* Handle start of quoted string */ { if (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E) - { - ptr += 2; /* avoid empty string */ - } - else inescq = TRUE; - continue; - } + { + ptr += 2; /* avoid empty string */ + } + else inescq = TRUE; + continue; + } else if (escape == ESC_E) continue; /* Ignore orphan \E */ - + else - { + { register const pcre_uint8 *cbits = cd->cbits; /* Every class contains at least two < 256 characters. */ class_has_8bitchar++; /* Every class contains at least two characters. */ class_one_char += 2; - + switch (escape) - { + { #ifdef SUPPORT_UCP case ESC_du: /* These are the values given for \d etc */ case ESC_DU: /* when PCRE_UCP is set. We replace the */ @@ -5212,24 +5212,24 @@ for (;; ptr++) class_has_8bitchar--; /* Undo! */ continue; #endif - case ESC_d: - for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_digit]; - continue; - - case ESC_D: - should_flip_negation = TRUE; - for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_digit]; - continue; - - case ESC_w: - for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_word]; - continue; - - case ESC_W: - should_flip_negation = TRUE; - for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_word]; - continue; - + case ESC_d: + for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_digit]; + continue; + + case ESC_D: + should_flip_negation = TRUE; + for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_digit]; + continue; + + case ESC_w: + for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_word]; + continue; + + case ESC_W: + should_flip_negation = TRUE; + for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_word]; + continue; + /* Perl 5.004 onwards omitted VT from \s, but restored it at Perl 5.18. Before PCRE 8.34, we had to preserve the VT bit if it was previously set by something earlier in the character class. @@ -5237,41 +5237,41 @@ for (;; ptr++) we could just adjust the appropriate bit. From PCRE 8.34 we no longer treat \s and \S specially. */ - case ESC_s: - for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_space]; - continue; - - case ESC_S: - should_flip_negation = TRUE; - for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_space]; - continue; - + case ESC_s: + for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_space]; + continue; + + case ESC_S: + should_flip_negation = TRUE; + for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_space]; + continue; + /* The rest apply in both UCP and non-UCP cases. */ - + case ESC_h: (void)add_list_to_class(classbits, &class_uchardata, options, cd, PRIV(hspace_list), NOTACHAR); continue; - + case ESC_H: (void)add_not_list_to_class(classbits, &class_uchardata, options, cd, PRIV(hspace_list)); continue; - + case ESC_v: (void)add_list_to_class(classbits, &class_uchardata, options, cd, PRIV(vspace_list), NOTACHAR); - continue; - + continue; + case ESC_V: (void)add_not_list_to_class(classbits, &class_uchardata, options, cd, PRIV(vspace_list)); - continue; - + continue; + case ESC_p: case ESC_P: #ifdef SUPPORT_UCP - { + { BOOL negated; unsigned int ptype = 0, pdata = 0; if (!get_ucp(&ptr, &negated, &ptype, &pdata, errorcodeptr)) @@ -5283,118 +5283,118 @@ for (;; ptr++) xclass_has_prop = TRUE; class_has_8bitchar--; /* Undo! */ continue; - } + } #else *errorcodeptr = ERR45; goto FAILED; -#endif +#endif /* Unrecognized escapes are faulted if PCRE is running in its strict mode. By default, for compatibility with Perl, they are treated as literals. */ - + default: if ((options & PCRE_EXTRA) != 0) - { + { *errorcodeptr = ERR7; goto FAILED; - } + } class_has_8bitchar--; /* Undo the speculative increase. */ class_one_char -= 2; /* Undo the speculative increase. */ c = *ptr; /* Get the final character and fall through */ break; - } + } } - + /* Fall through if the escape just defined a single character (c >= 0). This may be greater than 256. */ - + escape = 0; - - } /* End of backslash handling */ - + + } /* End of backslash handling */ + /* A character may be followed by '-' to form a range. However, Perl does not permit ']' to be the end of the range. A '-' character at the end is treated as a literal. Perl ignores orphaned \E sequences entirely. The code for handling \Q and \E is messy. */ - - CHECK_RANGE: + + CHECK_RANGE: while (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E) - { - inescq = FALSE; - ptr += 2; - } - oldptr = ptr; - + { + inescq = FALSE; + ptr += 2; + } + oldptr = ptr; + /* Remember if \r or \n were explicitly used */ - + if (c == CHAR_CR || c == CHAR_NL) cd->external_flags |= PCRE_HASCRORLF; - - /* Check for range */ - + + /* Check for range */ + if (!inescq && ptr[1] == CHAR_MINUS) - { + { pcre_uint32 d; - ptr += 2; + ptr += 2; while (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_E) ptr += 2; - - /* If we hit \Q (not followed by \E) at this point, go into escaped - mode. */ - + + /* If we hit \Q (not followed by \E) at this point, go into escaped + mode. */ + while (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_Q) - { - ptr += 2; + { + ptr += 2; if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_E) { ptr += 2; continue; } - inescq = TRUE; - break; - } - + inescq = TRUE; + break; + } + /* Minus (hyphen) at the end of a class is treated as a literal, so put back the pointer and jump to handle the character that preceded it. */ if (*ptr == CHAR_NULL || (!inescq && *ptr == CHAR_RIGHT_SQUARE_BRACKET)) - { - ptr = oldptr; + { + ptr = oldptr; goto CLASS_SINGLE_CHARACTER; - } - + } + /* Otherwise, we have a potential range; pick up the next character */ #ifdef SUPPORT_UTF if (utf) - { /* Braces are required because the */ - GETCHARLEN(d, ptr, ptr); /* macro generates multiple statements */ - } - else -#endif - d = *ptr; /* Not UTF-8 mode */ - + { /* Braces are required because the */ + GETCHARLEN(d, ptr, ptr); /* macro generates multiple statements */ + } + else +#endif + d = *ptr; /* Not UTF-8 mode */ + /* The second part of a range can be a single-character escape sequence, but not any of the other escapes. Perl treats a hyphen as a literal in such circumstances. However, in Perl's warning mode, a warning is given, so PCRE now faults it as it is almost certainly a mistake on the user's part. */ - + if (!inescq) - { + { if (d == CHAR_BACKSLASH) { int descape; descape = check_escape(&ptr, &d, errorcodeptr, cd->bracount, options, TRUE); if (*errorcodeptr != 0) goto FAILED; - + /* 0 means a character was put into d; \b is backspace; any other special causes an error. */ - + if (descape != 0) - { + { if (descape == ESC_b) d = CHAR_BS; else { *errorcodeptr = ERR83; goto FAILED; } - } - } + } + } /* A hyphen followed by a POSIX class is treated in the same way. */ @@ -5406,43 +5406,43 @@ for (;; ptr++) *errorcodeptr = ERR83; goto FAILED; } - } - - /* Check that the two values are in the correct order. Optimize + } + + /* Check that the two values are in the correct order. Optimize one-character ranges. */ - - if (d < c) - { - *errorcodeptr = ERR8; - goto FAILED; - } + + if (d < c) + { + *errorcodeptr = ERR8; + goto FAILED; + } if (d == c) goto CLASS_SINGLE_CHARACTER; /* A few lines below */ - + /* We have found a character range, so single character optimizations cannot be done anymore. Any value greater than 1 indicates that there is more than one character. */ - + class_one_char = 2; - + /* Remember an explicit \r or \n, and add the range to the class. */ - + if (d == CHAR_CR || d == CHAR_NL) cd->external_flags |= PCRE_HASCRORLF; - + class_has_8bitchar += add_to_class(classbits, &class_uchardata, options, cd, c, d); - + continue; /* Go get the next char in the class */ } - + /* Handle a single character - we can get here for a normal non-escape char, or after \ that introduces a single character or for an apparent range that isn't. Only the value 1 matters for class_one_char, so don't increase it if it is already 2 or more ... just in case there's a class with a zillion characters in it. */ - + CLASS_SINGLE_CHARACTER: if (class_one_char < 2) class_one_char++; - + /* If xclass_has_prop is false and class_one_char is 1, we have the first single character in the class, and there have been no prior ranges, or XCLASS items generated by escapes. If this is the final character in the @@ -5451,7 +5451,7 @@ for (;; ptr++) can cause firstchar to be set. Otherwise, there can be no first char if this item is first, whatever repeat count may follow. In the case of reqchar, save the previous value for reinstating. */ - + if (!inescq && #ifdef SUPPORT_UCP !xclass_has_prop && @@ -5461,7 +5461,7 @@ for (;; ptr++) ptr++; zeroreqchar = reqchar; zeroreqcharflags = reqcharflags; - + if (negate_class) { #ifdef SUPPORT_UCP @@ -5470,12 +5470,12 @@ for (;; ptr++) if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE; zerofirstchar = firstchar; zerofirstcharflags = firstcharflags; - + /* For caseless UTF-8 mode when UCP support is available, check whether this character has more than one other case. If so, generate a special OP_NOTPROP item instead of OP_NOTI. */ - -#ifdef SUPPORT_UCP + +#ifdef SUPPORT_UCP if (utf && (options & PCRE_CASELESS) != 0 && (d = UCD_CASESET(c)) != 0) { @@ -5486,8 +5486,8 @@ for (;; ptr++) else #endif /* Char has only one other case, or UCP not available */ - - { + + { *code++ = ((options & PCRE_CASELESS) != 0)? OP_NOTI: OP_NOT; #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf && c > MAX_VALUE_FOR_SINGLE_CHAR) @@ -5495,52 +5495,52 @@ for (;; ptr++) else #endif *code++ = c; - } - + } + /* We are finished with this character class */ - + goto END_CLASS; } - + /* For a single, positive character, get the value into mcbuffer, and then we can handle this with the normal one-character code. */ - + #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf && c > MAX_VALUE_FOR_SINGLE_CHAR) mclength = PRIV(ord2utf)(c, mcbuffer); else #endif - { + { mcbuffer[0] = c; mclength = 1; - } + } goto ONE_CHAR; } /* End of 1-char optimization */ - + /* There is more than one character in the class, or an XCLASS item has been generated. Add this character to the class. */ - + class_has_8bitchar += add_to_class(classbits, &class_uchardata, options, cd, c, c); - } - + } + /* Loop until ']' reached. This "while" is the end of the "do" far above. If we are at the end of an internal nested string, revert to the outer string. */ - + while (((c = *(++ptr)) != CHAR_NULL || (nestptr != NULL && (ptr = nestptr, nestptr = NULL, c = *(++ptr)) != CHAR_NULL)) && (c != CHAR_RIGHT_SQUARE_BRACKET || inescq)); - + /* Check for missing terminating ']' */ if (c == CHAR_NULL) - { - *errorcodeptr = ERR6; - goto FAILED; - } - + { + *errorcodeptr = ERR6; + goto FAILED; + } + /* We will need an XCLASS if data has been placed in class_uchardata. In the second phase this is a sufficient test. However, in the pre-compile phase, class_uchardata gets emptied to prevent workspace overflow, so it @@ -5548,21 +5548,21 @@ for (;; ptr++) anything at this point. For this reason, xclass gets set TRUE above when uchar_classdata is emptied, and that's why this code is the way it is here instead of just doing a test on class_uchardata below. */ - + #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 if (class_uchardata > class_uchardata_base) xclass = TRUE; -#endif - +#endif + /* If this is the first thing in the branch, there can be no first char setting, whatever the repeat count. Any reqchar setting must remain unchanged after any kind of repeat. */ - + if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE; zerofirstchar = firstchar; zerofirstcharflags = firstcharflags; zeroreqchar = reqchar; zeroreqcharflags = reqcharflags; - + /* If there are characters with values > 255, we have to compile an extended class, with its own opcode, unless there was a negated special such as \S in the class, and PCRE_UCP is not set, because in that case all @@ -5570,25 +5570,25 @@ for (;; ptr++) well can be ignored. If (when there are explicit characters > 255 that must be listed) there are no characters < 256, we can omit the bitmap in the actual compiled code. */ - + #ifdef SUPPORT_UTF if (xclass && (xclass_has_prop || !should_flip_negation || (options & PCRE_UCP) != 0)) #elif !defined COMPILE_PCRE8 if (xclass && (xclass_has_prop || !should_flip_negation)) -#endif +#endif #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 - { + { /* For non-UCP wide characters, in a non-negative class containing \S or similar (should_flip_negation is set), all characters greater than 255 must be in the class. */ - + if ( #if defined COMPILE_PCRE8 utf && #endif should_flip_negation && !negate_class && (options & PCRE_UCP) == 0) - { + { *class_uchardata++ = XCL_RANGE; if (utf) /* Will always be utf in the 8-bit library */ { @@ -5603,114 +5603,114 @@ for (;; ptr++) #elif defined COMPILE_PCRE32 *class_uchardata++ = 0x100; *class_uchardata++ = 0xffffffffu; -#endif +#endif } - } - + } + *class_uchardata++ = XCL_END; /* Marks the end of extra data */ - *code++ = OP_XCLASS; - code += LINK_SIZE; + *code++ = OP_XCLASS; + code += LINK_SIZE; *code = negate_class? XCL_NOT:0; if (xclass_has_prop) *code |= XCL_HASPROP; - - /* If the map is required, move up the extra data to make room for it; - otherwise just move the code pointer to the end of the extra data. */ - + + /* If the map is required, move up the extra data to make room for it; + otherwise just move the code pointer to the end of the extra data. */ + if (class_has_8bitchar > 0) - { - *code++ |= XCL_MAP; + { + *code++ |= XCL_MAP; memmove(code + (32 / sizeof(pcre_uchar)), code, IN_UCHARS(class_uchardata - code)); if (negate_class && !xclass_has_prop) for (c = 0; c < 32; c++) classbits[c] = ~classbits[c]; - memcpy(code, classbits, 32); + memcpy(code, classbits, 32); code = class_uchardata + (32 / sizeof(pcre_uchar)); - } + } else code = class_uchardata; - - /* Now fill in the complete length of the item */ - + + /* Now fill in the complete length of the item */ + PUT(previous, 1, (int)(code - previous)); - break; /* End of class handling */ - } + break; /* End of class handling */ + } /* Even though any XCLASS list is now discarded, we must allow for its memory. */ if (lengthptr != NULL) *lengthptr += (int)(class_uchardata - class_uchardata_base); -#endif - +#endif + /* If there are no characters > 255, or they are all to be included or excluded, set the opcode to OP_CLASS or OP_NCLASS, depending on whether the whole class was negated and whether there were negative specials such as \S (non-UCP) in the class. Then copy the 32-byte map into the code vector, negating it if necessary. */ - - *code++ = (negate_class == should_flip_negation) ? OP_CLASS : OP_NCLASS; + + *code++ = (negate_class == should_flip_negation) ? OP_CLASS : OP_NCLASS; if (lengthptr == NULL) /* Save time in the pre-compile phase */ - { + { if (negate_class) for (c = 0; c < 32; c++) classbits[c] = ~classbits[c]; - memcpy(code, classbits, 32); - } + memcpy(code, classbits, 32); + } code += 32 / sizeof(pcre_uchar); END_CLASS: - break; - - - /* ===================================================================*/ - /* Various kinds of repeat; '{' is not necessarily a quantifier, but this - has been tested above. */ - + break; + + + /* ===================================================================*/ + /* Various kinds of repeat; '{' is not necessarily a quantifier, but this + has been tested above. */ + case CHAR_LEFT_CURLY_BRACKET: - if (!is_quantifier) goto NORMAL_CHAR; - ptr = read_repeat_counts(ptr+1, &repeat_min, &repeat_max, errorcodeptr); - if (*errorcodeptr != 0) goto FAILED; - goto REPEAT; - + if (!is_quantifier) goto NORMAL_CHAR; + ptr = read_repeat_counts(ptr+1, &repeat_min, &repeat_max, errorcodeptr); + if (*errorcodeptr != 0) goto FAILED; + goto REPEAT; + case CHAR_ASTERISK: - repeat_min = 0; - repeat_max = -1; - goto REPEAT; - + repeat_min = 0; + repeat_max = -1; + goto REPEAT; + case CHAR_PLUS: - repeat_min = 1; - repeat_max = -1; - goto REPEAT; - + repeat_min = 1; + repeat_max = -1; + goto REPEAT; + case CHAR_QUESTION_MARK: - repeat_min = 0; - repeat_max = 1; - - REPEAT: - if (previous == NULL) - { - *errorcodeptr = ERR9; - goto FAILED; - } - - if (repeat_min == 0) - { + repeat_min = 0; + repeat_max = 1; + + REPEAT: + if (previous == NULL) + { + *errorcodeptr = ERR9; + goto FAILED; + } + + if (repeat_min == 0) + { firstchar = zerofirstchar; /* Adjust for zero repeat */ firstcharflags = zerofirstcharflags; reqchar = zeroreqchar; /* Ditto */ reqcharflags = zeroreqcharflags; - } - - /* Remember whether this is a variable length repeat */ - - reqvary = (repeat_min == repeat_max)? 0 : REQ_VARY; - - op_type = 0; /* Default single-char op codes */ - possessive_quantifier = FALSE; /* Default not possessive quantifier */ - + } + + /* Remember whether this is a variable length repeat */ + + reqvary = (repeat_min == repeat_max)? 0 : REQ_VARY; + + op_type = 0; /* Default single-char op codes */ + possessive_quantifier = FALSE; /* Default not possessive quantifier */ + /* Save start of previous item, in case we have to move it up in order to insert something before it. */ - - tempcode = previous; - + + tempcode = previous; + /* Before checking for a possessive quantifier, we must skip over whitespace and comments in extended mode because Perl allows white space at this point. */ @@ -5754,33 +5754,33 @@ for (;; ptr++) } } - /* If the next character is '+', we have a possessive quantifier. This - implies greediness, whatever the setting of the PCRE_UNGREEDY option. - If the next character is '?' this is a minimizing repeat, by default, - but if PCRE_UNGREEDY is set, it works the other way round. We change the - repeat type to the non-default. */ - + /* If the next character is '+', we have a possessive quantifier. This + implies greediness, whatever the setting of the PCRE_UNGREEDY option. + If the next character is '?' this is a minimizing repeat, by default, + but if PCRE_UNGREEDY is set, it works the other way round. We change the + repeat type to the non-default. */ + if (ptr[1] == CHAR_PLUS) - { - repeat_type = 0; /* Force greedy */ - possessive_quantifier = TRUE; - ptr++; - } + { + repeat_type = 0; /* Force greedy */ + possessive_quantifier = TRUE; + ptr++; + } else if (ptr[1] == CHAR_QUESTION_MARK) - { - repeat_type = greedy_non_default; - ptr++; - } - else repeat_type = greedy_default; - + { + repeat_type = greedy_non_default; + ptr++; + } + else repeat_type = greedy_default; + /* If previous was a recursion call, wrap it in atomic brackets so that previous becomes the atomic group. All recursions were so wrapped in the past, but it no longer happens for non-repeated recursions. In fact, the repeated ones could be re-implemented independently so as not to need this, but for the moment we rely on the code for repeating groups. */ - + if (*previous == OP_RECURSE) - { + { memmove(previous + 1 + LINK_SIZE, previous, IN_UCHARS(1 + LINK_SIZE)); *previous = OP_ONCE; PUT(previous, 1, 2 + 2*LINK_SIZE); @@ -5788,20 +5788,20 @@ for (;; ptr++) PUT(previous, 3 + 2*LINK_SIZE, 2 + 2*LINK_SIZE); code += 2 + 2 * LINK_SIZE; length_prevgroup = 3 + 3*LINK_SIZE; - + /* When actually compiling, we need to check whether this was a forward reference, and if so, adjust the offset. */ if (lengthptr == NULL && cd->hwm >= cd->start_workspace + LINK_SIZE) - { + { int offset = GET(cd->hwm, -LINK_SIZE); if (offset == previous + 1 - cd->start_code) PUT(cd->hwm, -LINK_SIZE, offset + 1 + LINK_SIZE); - } + } } - + /* Now handle repetition for the different types of item. */ - + /* If previous was a character or negated character match, abolish the item and generate a repeat item instead. If a char item has a minimum of more than one, ensure that it is set in reqchar - it might not be if a sequence @@ -5812,236 +5812,236 @@ for (;; ptr++) || *previous == OP_NOT || *previous == OP_NOTI) { switch (*previous) - { + { default: /* Make compiler happy. */ case OP_CHAR: op_type = OP_STAR - OP_STAR; break; case OP_CHARI: op_type = OP_STARI - OP_STAR; break; case OP_NOT: op_type = OP_NOTSTAR - OP_STAR; break; case OP_NOTI: op_type = OP_NOTSTARI - OP_STAR; break; - } - + } + /* Deal with UTF characters that take up more than one character. It's easier to write this out separately than try to macrify it. Use c to hold the length of the character in bytes, plus UTF_LENGTH to flag that it's a length rather than a small character. */ - + #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf && NOT_FIRSTCHAR(code[-1])) - { + { pcre_uchar *lastchar = code - 1; BACKCHAR(lastchar); c = (int)(code - lastchar); /* Length of UTF-8 character */ memcpy(utf_chars, lastchar, IN_UCHARS(c)); /* Save the char */ c |= UTF_LENGTH; /* Flag c as a length */ - } + } else #endif /* SUPPORT_UTF */ - + /* Handle the case of a single charater - either with no UTF support, or with UTF disabled, or for a single character UTF character. */ - { + { c = code[-1]; if (*previous <= OP_CHARI && repeat_min > 1) { reqchar = c; reqcharflags = req_caseopt | cd->req_varyopt; } - } + } goto OUTPUT_SINGLE_REPEAT; /* Code shared with single character types */ - } - - /* If previous was a character type match (\d or similar), abolish it and - create a suitable repeat item. The code is shared with single-character - repeats by setting op_type to add a suitable offset into repeat_type. Note - the the Unicode property types will be present only when SUPPORT_UCP is - defined, but we don't wrap the little bits of code here because it just - makes it horribly messy. */ - - else if (*previous < OP_EODN) - { + } + + /* If previous was a character type match (\d or similar), abolish it and + create a suitable repeat item. The code is shared with single-character + repeats by setting op_type to add a suitable offset into repeat_type. Note + the the Unicode property types will be present only when SUPPORT_UCP is + defined, but we don't wrap the little bits of code here because it just + makes it horribly messy. */ + + else if (*previous < OP_EODN) + { pcre_uchar *oldcode; - int prop_type, prop_value; - op_type = OP_TYPESTAR - OP_STAR; /* Use type opcodes */ - c = *previous; - - OUTPUT_SINGLE_REPEAT: - if (*previous == OP_PROP || *previous == OP_NOTPROP) - { - prop_type = previous[1]; - prop_value = previous[2]; - } - else prop_type = prop_value = -1; - - oldcode = code; - code = previous; /* Usually overwrite previous item */ - - /* If the maximum is zero then the minimum must also be zero; Perl allows - this case, so we do too - by simply omitting the item altogether. */ - - if (repeat_max == 0) goto END_REPEAT; - - /* Combine the op_type with the repeat_type */ - - repeat_type += op_type; - - /* A minimum of zero is handled either as the special case * or ?, or as - an UPTO, with the maximum given. */ - - if (repeat_min == 0) - { - if (repeat_max == -1) *code++ = OP_STAR + repeat_type; - else if (repeat_max == 1) *code++ = OP_QUERY + repeat_type; - else - { - *code++ = OP_UPTO + repeat_type; - PUT2INC(code, 0, repeat_max); - } - } - - /* A repeat minimum of 1 is optimized into some special cases. If the - maximum is unlimited, we use OP_PLUS. Otherwise, the original item is - left in place and, if the maximum is greater than 1, we use OP_UPTO with - one less than the maximum. */ - - else if (repeat_min == 1) - { - if (repeat_max == -1) - *code++ = OP_PLUS + repeat_type; - else - { - code = oldcode; /* leave previous item in place */ - if (repeat_max == 1) goto END_REPEAT; - *code++ = OP_UPTO + repeat_type; - PUT2INC(code, 0, repeat_max - 1); - } - } - - /* The case {n,n} is just an EXACT, while the general case {n,m} is - handled as an EXACT followed by an UPTO. */ - - else - { - *code++ = OP_EXACT + op_type; /* NB EXACT doesn't have repeat_type */ - PUT2INC(code, 0, repeat_min); - - /* If the maximum is unlimited, insert an OP_STAR. Before doing so, - we have to insert the character for the previous code. For a repeated - Unicode property match, there are two extra bytes that define the - required property. In UTF-8 mode, long characters have their length in + int prop_type, prop_value; + op_type = OP_TYPESTAR - OP_STAR; /* Use type opcodes */ + c = *previous; + + OUTPUT_SINGLE_REPEAT: + if (*previous == OP_PROP || *previous == OP_NOTPROP) + { + prop_type = previous[1]; + prop_value = previous[2]; + } + else prop_type = prop_value = -1; + + oldcode = code; + code = previous; /* Usually overwrite previous item */ + + /* If the maximum is zero then the minimum must also be zero; Perl allows + this case, so we do too - by simply omitting the item altogether. */ + + if (repeat_max == 0) goto END_REPEAT; + + /* Combine the op_type with the repeat_type */ + + repeat_type += op_type; + + /* A minimum of zero is handled either as the special case * or ?, or as + an UPTO, with the maximum given. */ + + if (repeat_min == 0) + { + if (repeat_max == -1) *code++ = OP_STAR + repeat_type; + else if (repeat_max == 1) *code++ = OP_QUERY + repeat_type; + else + { + *code++ = OP_UPTO + repeat_type; + PUT2INC(code, 0, repeat_max); + } + } + + /* A repeat minimum of 1 is optimized into some special cases. If the + maximum is unlimited, we use OP_PLUS. Otherwise, the original item is + left in place and, if the maximum is greater than 1, we use OP_UPTO with + one less than the maximum. */ + + else if (repeat_min == 1) + { + if (repeat_max == -1) + *code++ = OP_PLUS + repeat_type; + else + { + code = oldcode; /* leave previous item in place */ + if (repeat_max == 1) goto END_REPEAT; + *code++ = OP_UPTO + repeat_type; + PUT2INC(code, 0, repeat_max - 1); + } + } + + /* The case {n,n} is just an EXACT, while the general case {n,m} is + handled as an EXACT followed by an UPTO. */ + + else + { + *code++ = OP_EXACT + op_type; /* NB EXACT doesn't have repeat_type */ + PUT2INC(code, 0, repeat_min); + + /* If the maximum is unlimited, insert an OP_STAR. Before doing so, + we have to insert the character for the previous code. For a repeated + Unicode property match, there are two extra bytes that define the + required property. In UTF-8 mode, long characters have their length in c, with the UTF_LENGTH bit as a flag. */ - - if (repeat_max < 0) - { + + if (repeat_max < 0) + { #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf && (c & UTF_LENGTH) != 0) - { + { memcpy(code, utf_chars, IN_UCHARS(c & 7)); - code += c & 7; - } - else -#endif - { - *code++ = c; - if (prop_type >= 0) - { - *code++ = prop_type; - *code++ = prop_value; - } - } - *code++ = OP_STAR + repeat_type; - } - - /* Else insert an UPTO if the max is greater than the min, again - preceded by the character, for the previously inserted code. If the - UPTO is just for 1 instance, we can use QUERY instead. */ - - else if (repeat_max != repeat_min) - { + code += c & 7; + } + else +#endif + { + *code++ = c; + if (prop_type >= 0) + { + *code++ = prop_type; + *code++ = prop_value; + } + } + *code++ = OP_STAR + repeat_type; + } + + /* Else insert an UPTO if the max is greater than the min, again + preceded by the character, for the previously inserted code. If the + UPTO is just for 1 instance, we can use QUERY instead. */ + + else if (repeat_max != repeat_min) + { #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf && (c & UTF_LENGTH) != 0) - { + { memcpy(code, utf_chars, IN_UCHARS(c & 7)); - code += c & 7; - } - else -#endif - *code++ = c; - if (prop_type >= 0) - { - *code++ = prop_type; - *code++ = prop_value; - } - repeat_max -= repeat_min; - - if (repeat_max == 1) - { - *code++ = OP_QUERY + repeat_type; - } - else - { - *code++ = OP_UPTO + repeat_type; - PUT2INC(code, 0, repeat_max); - } - } - } - - /* The character or character type itself comes last in all cases. */ - + code += c & 7; + } + else +#endif + *code++ = c; + if (prop_type >= 0) + { + *code++ = prop_type; + *code++ = prop_value; + } + repeat_max -= repeat_min; + + if (repeat_max == 1) + { + *code++ = OP_QUERY + repeat_type; + } + else + { + *code++ = OP_UPTO + repeat_type; + PUT2INC(code, 0, repeat_max); + } + } + } + + /* The character or character type itself comes last in all cases. */ + #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf && (c & UTF_LENGTH) != 0) - { + { memcpy(code, utf_chars, IN_UCHARS(c & 7)); - code += c & 7; - } - else -#endif - *code++ = c; - - /* For a repeated Unicode property match, there are two extra bytes that - define the required property. */ - -#ifdef SUPPORT_UCP - if (prop_type >= 0) - { - *code++ = prop_type; - *code++ = prop_value; - } -#endif - } - - /* If previous was a character class or a back reference, we put the repeat - stuff after it, but just skip the item if the repeat was {0,0}. */ - + code += c & 7; + } + else +#endif + *code++ = c; + + /* For a repeated Unicode property match, there are two extra bytes that + define the required property. */ + +#ifdef SUPPORT_UCP + if (prop_type >= 0) + { + *code++ = prop_type; + *code++ = prop_value; + } +#endif + } + + /* If previous was a character class or a back reference, we put the repeat + stuff after it, but just skip the item if the repeat was {0,0}. */ + else if (*previous == OP_CLASS || *previous == OP_NCLASS || #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 - *previous == OP_XCLASS || -#endif + *previous == OP_XCLASS || +#endif *previous == OP_REF || *previous == OP_REFI || *previous == OP_DNREF || *previous == OP_DNREFI) - { - if (repeat_max == 0) - { - code = previous; - goto END_REPEAT; - } - - if (repeat_min == 0 && repeat_max == -1) - *code++ = OP_CRSTAR + repeat_type; - else if (repeat_min == 1 && repeat_max == -1) - *code++ = OP_CRPLUS + repeat_type; - else if (repeat_min == 0 && repeat_max == 1) - *code++ = OP_CRQUERY + repeat_type; - else - { - *code++ = OP_CRRANGE + repeat_type; - PUT2INC(code, 0, repeat_min); - if (repeat_max == -1) repeat_max = 0; /* 2-byte encoding for max */ - PUT2INC(code, 0, repeat_max); - } - } - - /* If previous was a bracket group, we may have to replicate it in certain + { + if (repeat_max == 0) + { + code = previous; + goto END_REPEAT; + } + + if (repeat_min == 0 && repeat_max == -1) + *code++ = OP_CRSTAR + repeat_type; + else if (repeat_min == 1 && repeat_max == -1) + *code++ = OP_CRPLUS + repeat_type; + else if (repeat_min == 0 && repeat_max == 1) + *code++ = OP_CRQUERY + repeat_type; + else + { + *code++ = OP_CRRANGE + repeat_type; + PUT2INC(code, 0, repeat_min); + if (repeat_max == -1) repeat_max = 0; /* 2-byte encoding for max */ + PUT2INC(code, 0, repeat_max); + } + } + + /* If previous was a bracket group, we may have to replicate it in certain cases. Note that at this point we can encounter only the "basic" bracket opcodes such as BRA and CBRA, as this is the place where they get converted into the more special varieties such as BRAPOS and SBRA. A test for >= @@ -6049,56 +6049,56 @@ for (;; ptr++) ASSERTBACK_NOT, ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND. Originally, PCRE did not allow repetition of assertions, but now it does, for Perl compatibility. */ - + else if (*previous >= OP_ASSERT && *previous <= OP_COND) - { + { register int i; int len = (int)(code - previous); size_t base_hwm_offset = item_hwm_offset; pcre_uchar *bralink = NULL; pcre_uchar *brazeroptr = NULL; - + /* Repeating a DEFINE group is pointless, but Perl allows the syntax, so we just ignore the repeat. */ - - if (*previous == OP_COND && previous[LINK_SIZE+1] == OP_DEF) + + if (*previous == OP_COND && previous[LINK_SIZE+1] == OP_DEF) goto END_REPEAT; - + /* There is no sense in actually repeating assertions. The only potential use of repetition is in cases when the assertion is optional. Therefore, if the minimum is greater than zero, just ignore the repeat. If the maximum is not zero or one, set it to 1. */ - + if (*previous < OP_ONCE) /* Assertion */ - { + { if (repeat_min > 0) goto END_REPEAT; if (repeat_max < 0 || repeat_max > 1) repeat_max = 1; - } - - /* The case of a zero minimum is special because of the need to stick - OP_BRAZERO in front of it, and because the group appears once in the - data, whereas in other cases it appears the minimum number of times. For - this reason, it is simplest to treat this case separately, as otherwise - the code gets far too messy. There are several special subcases when the - minimum is zero. */ - - if (repeat_min == 0) - { + } + + /* The case of a zero minimum is special because of the need to stick + OP_BRAZERO in front of it, and because the group appears once in the + data, whereas in other cases it appears the minimum number of times. For + this reason, it is simplest to treat this case separately, as otherwise + the code gets far too messy. There are several special subcases when the + minimum is zero. */ + + if (repeat_min == 0) + { /* If the maximum is also zero, we used to just omit the group from the output altogether, like this: - + ** if (repeat_max == 0) ** { ** code = previous; ** goto END_REPEAT; ** } - + However, that fails when a group or a subgroup within it is referenced as a subroutine from elsewhere in the pattern, so now we stick in OP_SKIPZERO in front of it so that it is skipped on execution. As we don't have a list of which groups are referenced, we cannot do this selectively. - + If the maximum is 1 or unlimited, we just have to stick in the BRAZERO and do no more at this point. However, we do need to adjust any OP_RECURSE calls inside the group that refer to the group itself or any @@ -6107,94 +6107,94 @@ for (;; ptr++) this. */ if (repeat_max <= 1) /* Covers 0, 1, and unlimited */ - { - *code = OP_END; + { + *code = OP_END; adjust_recurse(previous, 1, utf, cd, item_hwm_offset); memmove(previous + 1, previous, IN_UCHARS(len)); - code++; + code++; if (repeat_max == 0) { *previous++ = OP_SKIPZERO; goto END_REPEAT; } brazeroptr = previous; /* Save for possessive optimizing */ - *previous++ = OP_BRAZERO + repeat_type; - } - - /* If the maximum is greater than 1 and limited, we have to replicate - in a nested fashion, sticking OP_BRAZERO before each set of brackets. - The first one has to be handled carefully because it's the original - copy, which has to be moved up. The remainder can be handled by code - that is common with the non-zero minimum case below. We have to - adjust the value or repeat_max, since one less copy is required. Once - again, we may have to adjust any OP_RECURSE calls inside the group. */ - - else - { - int offset; - *code = OP_END; + *previous++ = OP_BRAZERO + repeat_type; + } + + /* If the maximum is greater than 1 and limited, we have to replicate + in a nested fashion, sticking OP_BRAZERO before each set of brackets. + The first one has to be handled carefully because it's the original + copy, which has to be moved up. The remainder can be handled by code + that is common with the non-zero minimum case below. We have to + adjust the value or repeat_max, since one less copy is required. Once + again, we may have to adjust any OP_RECURSE calls inside the group. */ + + else + { + int offset; + *code = OP_END; adjust_recurse(previous, 2 + LINK_SIZE, utf, cd, item_hwm_offset); memmove(previous + 2 + LINK_SIZE, previous, IN_UCHARS(len)); - code += 2 + LINK_SIZE; - *previous++ = OP_BRAZERO + repeat_type; - *previous++ = OP_BRA; - - /* We chain together the bracket offset fields that have to be - filled in later when the ends of the brackets are reached. */ - + code += 2 + LINK_SIZE; + *previous++ = OP_BRAZERO + repeat_type; + *previous++ = OP_BRA; + + /* We chain together the bracket offset fields that have to be + filled in later when the ends of the brackets are reached. */ + offset = (bralink == NULL)? 0 : (int)(previous - bralink); - bralink = previous; - PUTINC(previous, 0, offset); - } - - repeat_max--; - } - - /* If the minimum is greater than zero, replicate the group as many - times as necessary, and adjust the maximum to the number of subsequent - copies that we need. If we set a first char from the group, and didn't - set a required char, copy the latter from the former. If there are any - forward reference subroutine calls in the group, there will be entries on - the workspace list; replicate these with an appropriate increment. */ - - else - { - if (repeat_min > 1) - { - /* In the pre-compile phase, we don't actually do the replication. We - just adjust the length as if we had. Do some paranoid checks for + bralink = previous; + PUTINC(previous, 0, offset); + } + + repeat_max--; + } + + /* If the minimum is greater than zero, replicate the group as many + times as necessary, and adjust the maximum to the number of subsequent + copies that we need. If we set a first char from the group, and didn't + set a required char, copy the latter from the former. If there are any + forward reference subroutine calls in the group, there will be entries on + the workspace list; replicate these with an appropriate increment. */ + + else + { + if (repeat_min > 1) + { + /* In the pre-compile phase, we don't actually do the replication. We + just adjust the length as if we had. Do some paranoid checks for potential integer overflow. The INT64_OR_DOUBLE type is a 64-bit integer type when available, otherwise double. */ - - if (lengthptr != NULL) - { - int delta = (repeat_min - 1)*length_prevgroup; + + if (lengthptr != NULL) + { + int delta = (repeat_min - 1)*length_prevgroup; if ((INT64_OR_DOUBLE)(repeat_min - 1)* (INT64_OR_DOUBLE)length_prevgroup > (INT64_OR_DOUBLE)INT_MAX || - OFLOW_MAX - *lengthptr < delta) - { - *errorcodeptr = ERR20; - goto FAILED; - } - *lengthptr += delta; - } - + OFLOW_MAX - *lengthptr < delta) + { + *errorcodeptr = ERR20; + goto FAILED; + } + *lengthptr += delta; + } + /* This is compiling for real. If there is a set first byte for the group, and we have not yet set a "required byte", set it. Make sure there is enough workspace for copying forward references before doing the copy. */ - - else - { + + else + { if (groupsetfirstchar && reqcharflags < 0) { reqchar = firstchar; reqcharflags = firstcharflags; } - for (i = 1; i < repeat_min; i++) - { + for (i = 1; i < repeat_min; i++) + { pcre_uchar *hc; size_t this_hwm_offset = cd->hwm - cd->start_workspace; memcpy(code, previous, IN_UCHARS(len)); @@ -6202,7 +6202,7 @@ for (;; ptr++) while (cd->hwm > cd->start_workspace + cd->workspace_size - WORK_SIZE_SAFETY_MARGIN - (this_hwm_offset - base_hwm_offset)) - { + { *errorcodeptr = expand_workspace(cd); if (*errorcodeptr != 0) goto FAILED; } @@ -6211,70 +6211,70 @@ for (;; ptr++) hc < (pcre_uchar *)cd->start_workspace + this_hwm_offset; hc += LINK_SIZE) { - PUT(cd->hwm, 0, GET(hc, 0) + len); - cd->hwm += LINK_SIZE; - } + PUT(cd->hwm, 0, GET(hc, 0) + len); + cd->hwm += LINK_SIZE; + } base_hwm_offset = this_hwm_offset; - code += len; - } - } - } - - if (repeat_max > 0) repeat_max -= repeat_min; - } - - /* This code is common to both the zero and non-zero minimum cases. If - the maximum is limited, it replicates the group in a nested fashion, - remembering the bracket starts on a stack. In the case of a zero minimum, - the first one was set up above. In all cases the repeat_max now specifies - the number of additional copies needed. Again, we must remember to - replicate entries on the forward reference list. */ - - if (repeat_max >= 0) - { - /* In the pre-compile phase, we don't actually do the replication. We - just adjust the length as if we had. For each repetition we must add 1 - to the length for BRAZERO and for all but the last repetition we must - add 2 + 2*LINKSIZE to allow for the nesting that occurs. Do some + code += len; + } + } + } + + if (repeat_max > 0) repeat_max -= repeat_min; + } + + /* This code is common to both the zero and non-zero minimum cases. If + the maximum is limited, it replicates the group in a nested fashion, + remembering the bracket starts on a stack. In the case of a zero minimum, + the first one was set up above. In all cases the repeat_max now specifies + the number of additional copies needed. Again, we must remember to + replicate entries on the forward reference list. */ + + if (repeat_max >= 0) + { + /* In the pre-compile phase, we don't actually do the replication. We + just adjust the length as if we had. For each repetition we must add 1 + to the length for BRAZERO and for all but the last repetition we must + add 2 + 2*LINKSIZE to allow for the nesting that occurs. Do some paranoid checks to avoid integer overflow. The INT64_OR_DOUBLE type is a 64-bit integer type when available, otherwise double. */ - - if (lengthptr != NULL && repeat_max > 0) - { - int delta = repeat_max * (length_prevgroup + 1 + 2 + 2*LINK_SIZE) - - 2 - 2*LINK_SIZE; /* Last one doesn't nest */ + + if (lengthptr != NULL && repeat_max > 0) + { + int delta = repeat_max * (length_prevgroup + 1 + 2 + 2*LINK_SIZE) - + 2 - 2*LINK_SIZE; /* Last one doesn't nest */ if ((INT64_OR_DOUBLE)repeat_max * (INT64_OR_DOUBLE)(length_prevgroup + 1 + 2 + 2*LINK_SIZE) > (INT64_OR_DOUBLE)INT_MAX || - OFLOW_MAX - *lengthptr < delta) - { - *errorcodeptr = ERR20; - goto FAILED; - } - *lengthptr += delta; - } - - /* This is compiling for real */ - - else for (i = repeat_max - 1; i >= 0; i--) - { + OFLOW_MAX - *lengthptr < delta) + { + *errorcodeptr = ERR20; + goto FAILED; + } + *lengthptr += delta; + } + + /* This is compiling for real */ + + else for (i = repeat_max - 1; i >= 0; i--) + { pcre_uchar *hc; size_t this_hwm_offset = cd->hwm - cd->start_workspace; - - *code++ = OP_BRAZERO + repeat_type; - - /* All but the final copy start a new nesting, maintaining the - chain of brackets outstanding. */ - - if (i != 0) - { - int offset; - *code++ = OP_BRA; + + *code++ = OP_BRAZERO + repeat_type; + + /* All but the final copy start a new nesting, maintaining the + chain of brackets outstanding. */ + + if (i != 0) + { + int offset; + *code++ = OP_BRA; offset = (bralink == NULL)? 0 : (int)(code - bralink); - bralink = code; - PUTINC(code, 0, offset); - } - + bralink = code; + PUTINC(code, 0, offset); + } + memcpy(code, previous, IN_UCHARS(len)); /* Ensure there is enough workspace for forward references before @@ -6283,7 +6283,7 @@ for (;; ptr++) while (cd->hwm > cd->start_workspace + cd->workspace_size - WORK_SIZE_SAFETY_MARGIN - (this_hwm_offset - base_hwm_offset)) - { + { *errorcodeptr = expand_workspace(cd); if (*errorcodeptr != 0) goto FAILED; } @@ -6292,41 +6292,41 @@ for (;; ptr++) hc < (pcre_uchar *)cd->start_workspace + this_hwm_offset; hc += LINK_SIZE) { - PUT(cd->hwm, 0, GET(hc, 0) + len + ((i != 0)? 2+LINK_SIZE : 1)); - cd->hwm += LINK_SIZE; - } + PUT(cd->hwm, 0, GET(hc, 0) + len + ((i != 0)? 2+LINK_SIZE : 1)); + cd->hwm += LINK_SIZE; + } base_hwm_offset = this_hwm_offset; - code += len; - } - - /* Now chain through the pending brackets, and fill in their length - fields (which are holding the chain links pro tem). */ - - while (bralink != NULL) - { - int oldlinkoffset; + code += len; + } + + /* Now chain through the pending brackets, and fill in their length + fields (which are holding the chain links pro tem). */ + + while (bralink != NULL) + { + int oldlinkoffset; int offset = (int)(code - bralink + 1); pcre_uchar *bra = code - offset; - oldlinkoffset = GET(bra, 1); - bralink = (oldlinkoffset == 0)? NULL : bralink - oldlinkoffset; - *code++ = OP_KET; - PUTINC(code, 0, offset); - PUT(bra, 1, offset); - } - } - + oldlinkoffset = GET(bra, 1); + bralink = (oldlinkoffset == 0)? NULL : bralink - oldlinkoffset; + *code++ = OP_KET; + PUTINC(code, 0, offset); + PUT(bra, 1, offset); + } + } + /* If the maximum is unlimited, set a repeater in the final copy. For ONCE brackets, that's all we need to do. However, possessively repeated ONCE brackets can be converted into non-capturing brackets, as the behaviour of (?:xx)++ is the same as (?>xx)++ and this saves having to deal with possessive ONCEs specially. - + Otherwise, when we are doing the actual compile phase, check to see whether this group is one that could match an empty string. If so, - convert the initial operator to the S form (e.g. OP_BRA -> OP_SBRA) so + convert the initial operator to the S form (e.g. OP_BRA -> OP_SBRA) so that runtime checking can be done. [This check is also applied to ONCE groups at runtime, but in a different way.] - + Then, if the quantifier was possessive and the bracket is not a conditional, we convert the BRA code to the POS form, and the KET code to KETRPOS. (It turns out to be convenient at runtime to detect this kind of @@ -6340,8 +6340,8 @@ for (;; ptr++) there will be earlier copies of the group, and so we still have to wrap the whole thing. */ - else - { + else + { pcre_uchar *ketcode = code - 1 - LINK_SIZE; pcre_uchar *bracode = ketcode - GET(ketcode, 1); @@ -6360,23 +6360,23 @@ for (;; ptr++) converted to non-capturing above). */ else - { + { /* In the compile phase, check for empty string matching. */ if (lengthptr == NULL) - { + { pcre_uchar *scode = bracode; do - { + { if (could_be_empty_branch(scode, ketcode, utf, cd, NULL)) { *bracode += OP_SBRA - OP_BRA; break; } scode += GET(scode, 1); - } + } while (*scode == OP_ALT); - } + } /* A conditional group with only one branch has an implicit empty alternative branch. */ @@ -6425,10 +6425,10 @@ for (;; ptr++) /* Non-possessive quantifier */ else *ketcode = OP_KETRMAX + repeat_type; - } - } - } - + } + } + } + /* If previous is OP_FAIL, it was generated by an empty class [] in JavaScript mode. The other ways in which OP_FAIL can be generated, that is by (*FAIL) or (?!) set previous to NULL, which gives a "nothing to repeat" @@ -6436,14 +6436,14 @@ for (;; ptr++) else if (*previous == OP_FAIL) goto END_REPEAT; - /* Else there's some kind of shambles */ - - else - { - *errorcodeptr = ERR11; - goto FAILED; - } - + /* Else there's some kind of shambles */ + + else + { + *errorcodeptr = ERR11; + goto FAILED; + } + /* If the character following a repeat is '+', possessive_quantifier is TRUE. For some opcodes, there are special alternative opcodes for this case. For anything else, we wrap the entire repeated item inside OP_ONCE @@ -6453,12 +6453,12 @@ for (;; ptr++) Some (but not all) possessively repeated subpatterns have already been completely handled in the code just above. For them, possessive_quantifier is always FALSE at this stage. Note that the repeated item starts at - tempcode, not at previous, which might be the first part of a string whose + tempcode, not at previous, which might be the first part of a string whose (former) last char we repeated. */ - - if (possessive_quantifier) - { - int len; + + if (possessive_quantifier) + { + int len; /* Possessifying an EXACT quantifier has no effect, so we can ignore it. However, QUERY, STAR, or UPTO may follow (for quantifiers such as {5,6}, @@ -6543,23 +6543,23 @@ for (;; ptr++) } #ifdef NEVER - if (len > 0) switch (*tempcode) - { - case OP_STAR: *tempcode = OP_POSSTAR; break; - case OP_PLUS: *tempcode = OP_POSPLUS; break; - case OP_QUERY: *tempcode = OP_POSQUERY; break; - case OP_UPTO: *tempcode = OP_POSUPTO; break; - + if (len > 0) switch (*tempcode) + { + case OP_STAR: *tempcode = OP_POSSTAR; break; + case OP_PLUS: *tempcode = OP_POSPLUS; break; + case OP_QUERY: *tempcode = OP_POSQUERY; break; + case OP_UPTO: *tempcode = OP_POSUPTO; break; + case OP_STARI: *tempcode = OP_POSSTARI; break; case OP_PLUSI: *tempcode = OP_POSPLUSI; break; case OP_QUERYI: *tempcode = OP_POSQUERYI; break; case OP_UPTOI: *tempcode = OP_POSUPTOI; break; - - case OP_NOTSTAR: *tempcode = OP_NOTPOSSTAR; break; - case OP_NOTPLUS: *tempcode = OP_NOTPOSPLUS; break; - case OP_NOTQUERY: *tempcode = OP_NOTPOSQUERY; break; - case OP_NOTUPTO: *tempcode = OP_NOTPOSUPTO; break; - + + case OP_NOTSTAR: *tempcode = OP_NOTPOSSTAR; break; + case OP_NOTPLUS: *tempcode = OP_NOTPOSPLUS; break; + case OP_NOTQUERY: *tempcode = OP_NOTPOSQUERY; break; + case OP_NOTUPTO: *tempcode = OP_NOTPOSUPTO; break; + case OP_NOTSTARI: *tempcode = OP_NOTPOSSTARI; break; case OP_NOTPLUSI: *tempcode = OP_NOTPOSPLUSI; break; case OP_NOTQUERYI: *tempcode = OP_NOTPOSQUERYI; break; @@ -6578,50 +6578,50 @@ for (;; ptr++) /* Because we are moving code along, we must ensure that any pending recursive references are updated. */ - default: + default: *code = OP_END; adjust_recurse(tempcode, 1 + LINK_SIZE, utf, cd, item_hwm_offset); memmove(tempcode + 1 + LINK_SIZE, tempcode, IN_UCHARS(len)); - code += 1 + LINK_SIZE; - len += 1 + LINK_SIZE; - tempcode[0] = OP_ONCE; - *code++ = OP_KET; - PUTINC(code, 0, len); - PUT(tempcode, 1, len); - break; - } + code += 1 + LINK_SIZE; + len += 1 + LINK_SIZE; + tempcode[0] = OP_ONCE; + *code++ = OP_KET; + PUTINC(code, 0, len); + PUT(tempcode, 1, len); + break; + } #endif - } - - /* In all case we no longer have a previous item. We also set the + } + + /* In all case we no longer have a previous item. We also set the "follows varying string" flag for subsequently encountered reqchars if - it isn't already set and we have just passed a varying length item. */ - - END_REPEAT: - previous = NULL; - cd->req_varyopt |= reqvary; - break; - - - /* ===================================================================*/ - /* Start of nested parenthesized sub-expression, or comment or lookahead or - lookbehind or option setting or condition or all the other extended - parenthesis forms. */ - + it isn't already set and we have just passed a varying length item. */ + + END_REPEAT: + previous = NULL; + cd->req_varyopt |= reqvary; + break; + + + /* ===================================================================*/ + /* Start of nested parenthesized sub-expression, or comment or lookahead or + lookbehind or option setting or condition or all the other extended + parenthesis forms. */ + case CHAR_LEFT_PARENTHESIS: ptr++; - + /* Now deal with various "verbs" that can be introduced by '*'. */ - + if (ptr[0] == CHAR_ASTERISK && (ptr[1] == ':' || (MAX_255(ptr[1]) && ((cd->ctypes[ptr[1]] & ctype_letter) != 0)))) - { - int i, namelen; + { + int i, namelen; int arglen = 0; - const char *vn = verbnames; + const char *vn = verbnames; const pcre_uchar *name = ptr + 1; const pcre_uchar *arg = NULL; - previous = NULL; + previous = NULL; ptr++; while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_letter) != 0) ptr++; namelen = (int)(ptr - name); @@ -6631,7 +6631,7 @@ for (;; ptr++) letters, digits, and underscores. */ if (*ptr == CHAR_COLON) - { + { arg = ++ptr; while (*ptr != CHAR_NULL && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++; arglen = (int)(ptr - arg); @@ -6640,21 +6640,21 @@ for (;; ptr++) *errorcodeptr = ERR75; goto FAILED; } - } + } if (*ptr != CHAR_RIGHT_PARENTHESIS) - { - *errorcodeptr = ERR60; - goto FAILED; - } + { + *errorcodeptr = ERR60; + goto FAILED; + } /* Scan the table of verb names */ - for (i = 0; i < verbcount; i++) - { - if (namelen == verbs[i].len && + for (i = 0; i < verbcount; i++) + { + if (namelen == verbs[i].len && STRNCMP_UC_C8(name, vn, namelen) == 0) - { + { int setverb; /* Check for open captures before ACCEPT and convert it to @@ -6744,16 +6744,16 @@ for (;; ptr++) } break; /* Found verb, exit loop */ - } + } - vn += verbs[i].len + 1; - } + vn += verbs[i].len + 1; + } if (i < verbcount) continue; /* Successfully handled a verb */ *errorcodeptr = ERR60; /* Verb not recognized */ - goto FAILED; - } - + goto FAILED; + } + /* Initialize for "real" parentheses */ newoptions = options; @@ -6762,48 +6762,48 @@ for (;; ptr++) item_hwm_offset = cd->hwm - cd->start_workspace; reset_bracount = FALSE; - /* Deal with the extended parentheses; all are introduced by '?', and the - appearance of any of them means that this is not a capturing group. */ - + /* Deal with the extended parentheses; all are introduced by '?', and the + appearance of any of them means that this is not a capturing group. */ + if (*ptr == CHAR_QUESTION_MARK) - { - int i, set, unset, namelen; - int *optset; + { + int i, set, unset, namelen; + int *optset; const pcre_uchar *name; pcre_uchar *slot; - - switch (*(++ptr)) - { - /* ------------------------------------------------------------ */ + + switch (*(++ptr)) + { + /* ------------------------------------------------------------ */ case CHAR_VERTICAL_LINE: /* Reset capture count for each branch */ - reset_bracount = TRUE; + reset_bracount = TRUE; cd->dupgroups = TRUE; /* Record (?| encountered */ - /* Fall through */ - - /* ------------------------------------------------------------ */ + /* Fall through */ + + /* ------------------------------------------------------------ */ case CHAR_COLON: /* Non-capturing bracket */ - bravalue = OP_BRA; - ptr++; - break; - - - /* ------------------------------------------------------------ */ + bravalue = OP_BRA; + ptr++; + break; + + + /* ------------------------------------------------------------ */ case CHAR_LEFT_PARENTHESIS: - bravalue = OP_COND; /* Conditional group */ + bravalue = OP_COND; /* Conditional group */ tempptr = ptr; - - /* A condition can be an assertion, a number (referring to a numbered + + /* A condition can be an assertion, a number (referring to a numbered group's having been set), a name (referring to a named group), or 'R', referring to recursion. R<digits> and R&name are also permitted for recursion tests. - + There are ways of testing a named group: (?(name)) is used by Python; Perl 5.10 onwards uses (?(<name>) or (?('name')). - + There is one unfortunate ambiguity, caused by history. 'R' can be the recursive thing or the name 'R' (and similarly for 'R' followed by digits). We look for a name first; if not found, we try the other case. - + For compatibility with auto-callouts, we allow a callout to be specified before a condition that is an assertion. First, check for the syntax of a callout; if found, adjust the temporary pointer that is @@ -6825,10 +6825,10 @@ for (;; ptr++) } } - /* For conditions that are assertions, check the syntax, and then exit - the switch. This will take control down to where bracketed groups, - including assertions, are processed. */ - + /* For conditions that are assertions, check the syntax, and then exit + the switch. This will take control down to where bracketed groups, + including assertions, are processed. */ + if (tempptr[1] == CHAR_QUESTION_MARK && (tempptr[2] == CHAR_EQUALS_SIGN || tempptr[2] == CHAR_EXCLAMATION_MARK || @@ -6837,54 +6837,54 @@ for (;; ptr++) tempptr[3] == CHAR_EXCLAMATION_MARK)))) { cd->iscondassert = TRUE; - break; + break; } - + /* Other conditions use OP_CREF/OP_DNCREF/OP_RREF/OP_DNRREF, and all need to skip at least 1+IMM2_SIZE bytes at the start of the group. */ - - code[1+LINK_SIZE] = OP_CREF; + + code[1+LINK_SIZE] = OP_CREF; skipbytes = 1+IMM2_SIZE; refsign = -1; /* => not a number */ namelen = -1; /* => not a name; must set to avoid warning */ name = NULL; /* Always set to avoid warning */ recno = 0; /* Always set to avoid warning */ - - /* Check for a test for recursion in a named group. */ - + + /* Check for a test for recursion in a named group. */ + ptr++; if (*ptr == CHAR_R && ptr[1] == CHAR_AMPERSAND) - { - terminator = -1; - ptr += 2; - code[1+LINK_SIZE] = OP_RREF; /* Change the type of test */ - } - - /* Check for a test for a named group's having been set, using the Perl + { + terminator = -1; + ptr += 2; + code[1+LINK_SIZE] = OP_RREF; /* Change the type of test */ + } + + /* Check for a test for a named group's having been set, using the Perl syntax (?(<name>) or (?('name'), and also allow for the original PCRE syntax of (?(name) or for (?(+n), (?(-n), and just (?(n). */ - + else if (*ptr == CHAR_LESS_THAN_SIGN) - { + { terminator = CHAR_GREATER_THAN_SIGN; - ptr++; - } + ptr++; + } else if (*ptr == CHAR_APOSTROPHE) - { + { terminator = CHAR_APOSTROPHE; - ptr++; - } - else - { + ptr++; + } + else + { terminator = CHAR_NULL; if (*ptr == CHAR_MINUS || *ptr == CHAR_PLUS) refsign = *ptr++; else if (IS_DIGIT(*ptr)) refsign = 0; - } - + } + /* Handle a number */ - + if (refsign >= 0) - { + { while (IS_DIGIT(*ptr)) { if (recno > INT_MAX / 10 - 1) /* Integer overflow */ @@ -6896,15 +6896,15 @@ for (;; ptr++) recno = recno * 10 + (int)(*ptr - CHAR_0); ptr++; } - } - + } + /* Otherwise we expect to read a name; anything else is an error. When a name is one of a number of duplicates, a different opcode is used and it needs more memory. Unfortunately we cannot tell whether a name is a duplicate in the first pass, so we have to allow for more memory. */ - + else - { + { if (IS_DIGIT(*ptr)) { *errorcodeptr = ERR84; @@ -6922,62 +6922,62 @@ for (;; ptr++) } namelen = (int)(ptr - name); if (lengthptr != NULL) skipbytes += IMM2_SIZE; - } - + } + /* Check the terminator */ if ((terminator > 0 && *ptr++ != (pcre_uchar)terminator) || *ptr++ != CHAR_RIGHT_PARENTHESIS) - { + { ptr--; /* Error offset */ *errorcodeptr = ERR26; /* Malformed number or name */ - goto FAILED; - } - - /* Do no further checking in the pre-compile phase. */ - - if (lengthptr != NULL) break; - - /* In the real compile we do the work of looking for the actual + goto FAILED; + } + + /* Do no further checking in the pre-compile phase. */ + + if (lengthptr != NULL) break; + + /* In the real compile we do the work of looking for the actual reference. If refsign is not negative, it means we have a number in recno. */ - + if (refsign >= 0) - { - if (recno <= 0) - { + { + if (recno <= 0) + { *errorcodeptr = ERR35; - goto FAILED; - } + goto FAILED; + } if (refsign != 0) recno = (refsign == CHAR_MINUS)? cd->bracount - recno + 1 : recno + cd->bracount; - if (recno <= 0 || recno > cd->final_bracount) - { - *errorcodeptr = ERR15; - goto FAILED; - } - PUT2(code, 2+LINK_SIZE, recno); + if (recno <= 0 || recno > cd->final_bracount) + { + *errorcodeptr = ERR15; + goto FAILED; + } + PUT2(code, 2+LINK_SIZE, recno); if (recno > cd->top_backref) cd->top_backref = recno; - break; - } - + break; + } + /* Otherwise look for the name. */ - - slot = cd->name_table; - for (i = 0; i < cd->names_found; i++) - { + + slot = cd->name_table; + for (i = 0; i < cd->names_found; i++) + { if (STRNCMP_UC_UC(name, slot+IMM2_SIZE, namelen) == 0 && slot[IMM2_SIZE+namelen] == 0) break; - slot += cd->name_entry_size; - } - + slot += cd->name_entry_size; + } + /* Found the named subpattern. If the name is duplicated, add one to the opcode to change CREF/RREF into DNCREF/DNRREF and insert appropriate data values. Otherwise, just insert the unique subpattern number. */ - - if (i < cd->names_found) - { + + if (i < cd->names_found) + { int offset = i++; int count = 1; recno = GET2(slot, 0); /* Number from first found */ @@ -6989,7 +6989,7 @@ for (;; ptr++) (slot+IMM2_SIZE)[namelen] != 0) break; count++; } - + if (count > 1) { PUT2(code, 2+LINK_SIZE, offset); @@ -7001,133 +7001,133 @@ for (;; ptr++) { PUT2(code, 2+LINK_SIZE, recno); } - } - + } + /* If terminator == CHAR_NULL it means that the name followed directly after the opening parenthesis [e.g. (?(abc)...] and in this case there are some further alternatives to try. For the cases where terminator != CHAR_NULL [things like (?(<name>... or (?('name')... or (?(R&name)... ] we have now checked all the possibilities, so give an error. */ - + else if (terminator != CHAR_NULL) - { - *errorcodeptr = ERR15; - goto FAILED; - } - - /* Check for (?(R) for recursion. Allow digits after R to specify a - specific group number. */ - + { + *errorcodeptr = ERR15; + goto FAILED; + } + + /* Check for (?(R) for recursion. Allow digits after R to specify a + specific group number. */ + else if (*name == CHAR_R) - { - recno = 0; - for (i = 1; i < namelen; i++) - { + { + recno = 0; + for (i = 1; i < namelen; i++) + { if (!IS_DIGIT(name[i])) - { - *errorcodeptr = ERR15; - goto FAILED; - } + { + *errorcodeptr = ERR15; + goto FAILED; + } if (recno > INT_MAX / 10 - 1) /* Integer overflow */ { *errorcodeptr = ERR61; goto FAILED; } recno = recno * 10 + name[i] - CHAR_0; - } - if (recno == 0) recno = RREF_ANY; - code[1+LINK_SIZE] = OP_RREF; /* Change test type */ - PUT2(code, 2+LINK_SIZE, recno); - } - - /* Similarly, check for the (?(DEFINE) "condition", which is always - false. */ - + } + if (recno == 0) recno = RREF_ANY; + code[1+LINK_SIZE] = OP_RREF; /* Change test type */ + PUT2(code, 2+LINK_SIZE, recno); + } + + /* Similarly, check for the (?(DEFINE) "condition", which is always + false. */ + else if (namelen == 6 && STRNCMP_UC_C8(name, STRING_DEFINE, 6) == 0) - { - code[1+LINK_SIZE] = OP_DEF; - skipbytes = 1; - } - + { + code[1+LINK_SIZE] = OP_DEF; + skipbytes = 1; + } + /* Reference to an unidentified subpattern. */ - - else - { + + else + { *errorcodeptr = ERR15; - goto FAILED; - } - break; - - - /* ------------------------------------------------------------ */ + goto FAILED; + } + break; + + + /* ------------------------------------------------------------ */ case CHAR_EQUALS_SIGN: /* Positive lookahead */ - bravalue = OP_ASSERT; + bravalue = OP_ASSERT; cd->assert_depth += 1; - ptr++; - break; - + ptr++; + break; + /* Optimize (?!) to (*FAIL) unless it is quantified - which is a weird thing to do, but Perl allows all assertions to be quantified, and when they contain capturing parentheses there may be a potential use for this feature. Not that that applies to a quantified (?!) but we allow it for uniformity. */ - - /* ------------------------------------------------------------ */ + + /* ------------------------------------------------------------ */ case CHAR_EXCLAMATION_MARK: /* Negative lookahead */ - ptr++; + ptr++; if (*ptr == CHAR_RIGHT_PARENTHESIS && ptr[1] != CHAR_ASTERISK && ptr[1] != CHAR_PLUS && ptr[1] != CHAR_QUESTION_MARK && (ptr[1] != CHAR_LEFT_CURLY_BRACKET || !is_counted_repeat(ptr+2))) - { - *code++ = OP_FAIL; - previous = NULL; - continue; - } - bravalue = OP_ASSERT_NOT; + { + *code++ = OP_FAIL; + previous = NULL; + continue; + } + bravalue = OP_ASSERT_NOT; cd->assert_depth += 1; - break; - - - /* ------------------------------------------------------------ */ + break; + + + /* ------------------------------------------------------------ */ case CHAR_LESS_THAN_SIGN: /* Lookbehind or named define */ - switch (ptr[1]) - { + switch (ptr[1]) + { case CHAR_EQUALS_SIGN: /* Positive lookbehind */ - bravalue = OP_ASSERTBACK; + bravalue = OP_ASSERTBACK; cd->assert_depth += 1; - ptr += 2; - break; - + ptr += 2; + break; + case CHAR_EXCLAMATION_MARK: /* Negative lookbehind */ - bravalue = OP_ASSERTBACK_NOT; + bravalue = OP_ASSERTBACK_NOT; cd->assert_depth += 1; - ptr += 2; - break; - - default: /* Could be name define, else bad */ + ptr += 2; + break; + + default: /* Could be name define, else bad */ if (MAX_255(ptr[1]) && (cd->ctypes[ptr[1]] & ctype_word) != 0) goto DEFINE_NAME; - ptr++; /* Correct offset for error */ - *errorcodeptr = ERR24; - goto FAILED; - } - break; - - - /* ------------------------------------------------------------ */ + ptr++; /* Correct offset for error */ + *errorcodeptr = ERR24; + goto FAILED; + } + break; + + + /* ------------------------------------------------------------ */ case CHAR_GREATER_THAN_SIGN: /* One-time brackets */ - bravalue = OP_ONCE; - ptr++; - break; - - - /* ------------------------------------------------------------ */ + bravalue = OP_ONCE; + ptr++; + break; + + + /* ------------------------------------------------------------ */ case CHAR_C: /* Callout - may be followed by digits; */ previous_callout = code; /* Save for later completion */ after_manual_callout = 1; /* Skip one item before completing */ - *code++ = OP_CALLOUT; - { - int n = 0; + *code++ = OP_CALLOUT; + { + int n = 0; ptr++; while(IS_DIGIT(*ptr)) { @@ -7139,63 +7139,63 @@ for (;; ptr++) } } if (*ptr != CHAR_RIGHT_PARENTHESIS) - { - *errorcodeptr = ERR39; - goto FAILED; - } - *code++ = n; + { + *errorcodeptr = ERR39; + goto FAILED; + } + *code++ = n; PUT(code, 0, (int)(ptr - cd->start_pattern + 1)); /* Pattern offset */ PUT(code, LINK_SIZE, 0); /* Default length */ - code += 2 * LINK_SIZE; - } - previous = NULL; - continue; - - - /* ------------------------------------------------------------ */ + code += 2 * LINK_SIZE; + } + previous = NULL; + continue; + + + /* ------------------------------------------------------------ */ case CHAR_P: /* Python-style named subpattern handling */ if (*(++ptr) == CHAR_EQUALS_SIGN || *ptr == CHAR_GREATER_THAN_SIGN) /* Reference or recursion */ - { + { is_recurse = *ptr == CHAR_GREATER_THAN_SIGN; terminator = CHAR_RIGHT_PARENTHESIS; - goto NAMED_REF_OR_RECURSE; - } + goto NAMED_REF_OR_RECURSE; + } else if (*ptr != CHAR_LESS_THAN_SIGN) /* Test for Python-style defn */ - { - *errorcodeptr = ERR41; - goto FAILED; - } - /* Fall through to handle (?P< as (?< is handled */ - - - /* ------------------------------------------------------------ */ - DEFINE_NAME: /* Come here from (?< handling */ + { + *errorcodeptr = ERR41; + goto FAILED; + } + /* Fall through to handle (?P< as (?< is handled */ + + + /* ------------------------------------------------------------ */ + DEFINE_NAME: /* Come here from (?< handling */ case CHAR_APOSTROPHE: terminator = (*ptr == CHAR_LESS_THAN_SIGN)? CHAR_GREATER_THAN_SIGN : CHAR_APOSTROPHE; name = ++ptr; if (IS_DIGIT(*ptr)) - { + { *errorcodeptr = ERR84; /* Group name must start with non-digit */ goto FAILED; } while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0) ptr++; namelen = (int)(ptr - name); - + /* In the pre-compile phase, do a syntax check, remember the longest name, and then remember the group in a vector, expanding it if necessary. Duplicates for the same number are skipped; other duplicates are checked for validity. In the actual compile, there is nothing to do. */ - + if (lengthptr != NULL) { named_group *ng; pcre_uint32 number = cd->bracount + 1; - + if (*ptr != (pcre_uchar)terminator) - { + { *errorcodeptr = ERR42; goto FAILED; } @@ -7210,10 +7210,10 @@ for (;; ptr++) { cd->name_entry_size = namelen + IMM2_SIZE + 1; if (namelen > MAX_NAME_SIZE) - { + { *errorcodeptr = ERR48; - goto FAILED; - } + goto FAILED; + } } /* Scan the list to check for duplicates. For duplicate names, if the @@ -7228,37 +7228,37 @@ for (;; ptr++) { if (namelen == ng->length && STRNCMP_UC_UC(name, ng->name, namelen) == 0) - { + { if (ng->number == number) break; if ((options & PCRE_DUPNAMES) == 0) - { + { *errorcodeptr = ERR43; - goto FAILED; - } + goto FAILED; + } cd->dupnames = TRUE; /* Duplicate names exist */ - } + } else if (ng->number == number) { *errorcodeptr = ERR65; goto FAILED; } - } - + } + if (i >= cd->names_found) /* Not a duplicate with same number */ { /* Increase the list size if necessary */ - + if (cd->names_found >= cd->named_group_list_size) - { + { int newsize = cd->named_group_list_size * 2; named_group *newspace = (PUBL(malloc)) (newsize * sizeof(named_group)); if (newspace == NULL) - { + { *errorcodeptr = ERR21; goto FAILED; - } + } memcpy(newspace, cd->named_groups, cd->named_group_list_size * sizeof(named_group)); @@ -7266,33 +7266,33 @@ for (;; ptr++) (PUBL(free))((void *)cd->named_groups); cd->named_groups = newspace; cd->named_group_list_size = newsize; - } - + } + cd->named_groups[cd->names_found].name = name; cd->named_groups[cd->names_found].length = namelen; cd->named_groups[cd->names_found].number = number; cd->names_found++; - } - } - + } + } + ptr++; /* Move past > or ' in both passes. */ - goto NUMBERED_GROUP; - - - /* ------------------------------------------------------------ */ + goto NUMBERED_GROUP; + + + /* ------------------------------------------------------------ */ case CHAR_AMPERSAND: /* Perl recursion/subroutine syntax */ terminator = CHAR_RIGHT_PARENTHESIS; - is_recurse = TRUE; - /* Fall through */ - - /* We come here from the Python syntax above that handles both - references (?P=name) and recursion (?P>name), as well as falling - through from the Perl recursion syntax (?&name). We also come here from - the Perl \k<name> or \k'name' back reference syntax and the \k{name} + is_recurse = TRUE; + /* Fall through */ + + /* We come here from the Python syntax above that handles both + references (?P=name) and recursion (?P>name), as well as falling + through from the Perl recursion syntax (?&name). We also come here from + the Perl \k<name> or \k'name' back reference syntax and the \k{name} .NET syntax, and the Oniguruma \g<...> and \g'...' subroutine syntax. */ - - NAMED_REF_OR_RECURSE: - name = ++ptr; + + NAMED_REF_OR_RECURSE: + name = ++ptr; if (IS_DIGIT(*ptr)) { *errorcodeptr = ERR84; /* Group name must start with non-digit */ @@ -7300,34 +7300,34 @@ for (;; ptr++) } while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0) ptr++; namelen = (int)(ptr - name); - + /* In the pre-compile phase, do a syntax check. We used to just set a dummy reference number, because it was not used in the first pass. However, with the change of recursive back references to be atomic, we have to look for the number so that this state can be identified, as otherwise the incorrect length is computed. If it's not a backwards reference, the dummy number will do. */ - - if (lengthptr != NULL) - { + + if (lengthptr != NULL) + { named_group *ng; recno = 0; - if (namelen == 0) - { - *errorcodeptr = ERR62; - goto FAILED; - } + if (namelen == 0) + { + *errorcodeptr = ERR62; + goto FAILED; + } if (*ptr != (pcre_uchar)terminator) - { - *errorcodeptr = ERR42; - goto FAILED; - } - if (namelen > MAX_NAME_SIZE) - { - *errorcodeptr = ERR48; - goto FAILED; - } + { + *errorcodeptr = ERR42; + goto FAILED; + } + if (namelen > MAX_NAME_SIZE) + { + *errorcodeptr = ERR48; + goto FAILED; + } /* Count named back references. */ @@ -7393,43 +7393,43 @@ for (;; ptr++) } } } - } - + } + /* In the real compile, search the name table. We check the name - first, and then check that we have reached the end of the name in the + first, and then check that we have reached the end of the name in the table. That way, if the name is longer than any in the table, the comparison will fail without reading beyond the table entry. */ - - else - { - slot = cd->name_table; - for (i = 0; i < cd->names_found; i++) - { + + else + { + slot = cd->name_table; + for (i = 0; i < cd->names_found; i++) + { if (STRNCMP_UC_UC(name, slot+IMM2_SIZE, namelen) == 0 && slot[IMM2_SIZE+namelen] == 0) - break; - slot += cd->name_entry_size; - } - + break; + slot += cd->name_entry_size; + } + if (i < cd->names_found) - { - recno = GET2(slot, 0); - } + { + recno = GET2(slot, 0); + } else - { - *errorcodeptr = ERR15; - goto FAILED; - } - } - + { + *errorcodeptr = ERR15; + goto FAILED; + } + } + /* In both phases, for recursions, we can now go to the code than handles numerical recursion. */ - - if (is_recurse) goto HANDLE_RECURSION; - + + if (is_recurse) goto HANDLE_RECURSION; + /* In the second pass we must see if the name is duplicated. If so, we generate a different opcode. */ - + if (lengthptr == NULL && cd->dupnames) { int count = 1; @@ -7484,7 +7484,7 @@ for (;; ptr++) goto HANDLE_REFERENCE; - /* ------------------------------------------------------------ */ + /* ------------------------------------------------------------ */ case CHAR_R: /* Recursion, same as (?0) */ recno = 0; if (*(++ptr) != CHAR_RIGHT_PARENTHESIS) @@ -7493,16 +7493,16 @@ for (;; ptr++) goto FAILED; } goto HANDLE_RECURSION; - - - /* ------------------------------------------------------------ */ + + + /* ------------------------------------------------------------ */ case CHAR_MINUS: case CHAR_PLUS: /* Recursion or subroutine */ case CHAR_0: case CHAR_1: case CHAR_2: case CHAR_3: case CHAR_4: case CHAR_5: case CHAR_6: case CHAR_7: case CHAR_8: case CHAR_9: - { + { const pcre_uchar *called; terminator = CHAR_RIGHT_PARENTHESIS; - + /* Come here from the \g<...> and \g'...' code (Oniguruma compatibility). However, the syntax has been checked to ensure that the ... are a (signed) number, so that neither ERR63 nor ERR29 will @@ -7512,22 +7512,22 @@ for (;; ptr++) HANDLE_NUMERICAL_RECURSION: if ((refsign = *ptr) == CHAR_PLUS) - { - ptr++; + { + ptr++; if (!IS_DIGIT(*ptr)) - { - *errorcodeptr = ERR63; - goto FAILED; - } - } + { + *errorcodeptr = ERR63; + goto FAILED; + } + } else if (refsign == CHAR_MINUS) - { + { if (!IS_DIGIT(ptr[1])) - goto OTHER_CHAR_AFTER_QUERY; - ptr++; - } - - recno = 0; + goto OTHER_CHAR_AFTER_QUERY; + ptr++; + } + + recno = 0; while(IS_DIGIT(*ptr)) { if (recno > INT_MAX / 10 - 1) /* Integer overflow */ @@ -7538,73 +7538,73 @@ for (;; ptr++) } recno = recno * 10 + *ptr++ - CHAR_0; } - + if (*ptr != (pcre_uchar)terminator) - { - *errorcodeptr = ERR29; - goto FAILED; - } - + { + *errorcodeptr = ERR29; + goto FAILED; + } + if (refsign == CHAR_MINUS) - { - if (recno == 0) - { - *errorcodeptr = ERR58; - goto FAILED; - } - recno = cd->bracount - recno + 1; - if (recno <= 0) - { - *errorcodeptr = ERR15; - goto FAILED; - } - } + { + if (recno == 0) + { + *errorcodeptr = ERR58; + goto FAILED; + } + recno = cd->bracount - recno + 1; + if (recno <= 0) + { + *errorcodeptr = ERR15; + goto FAILED; + } + } else if (refsign == CHAR_PLUS) - { - if (recno == 0) - { - *errorcodeptr = ERR58; - goto FAILED; - } - recno += cd->bracount; - } - - /* Come here from code above that handles a named recursion */ - - HANDLE_RECURSION: - - previous = code; + { + if (recno == 0) + { + *errorcodeptr = ERR58; + goto FAILED; + } + recno += cd->bracount; + } + + /* Come here from code above that handles a named recursion */ + + HANDLE_RECURSION: + + previous = code; item_hwm_offset = cd->hwm - cd->start_workspace; - called = cd->start_code; - - /* When we are actually compiling, find the bracket that is being - referenced. Temporarily end the regex in case it doesn't exist before - this point. If we end up with a forward reference, first check that - the bracket does occur later so we can give the error (and position) - now. Then remember this forward reference in the workspace so it can - be filled in at the end. */ - - if (lengthptr == NULL) - { - *code = OP_END; + called = cd->start_code; + + /* When we are actually compiling, find the bracket that is being + referenced. Temporarily end the regex in case it doesn't exist before + this point. If we end up with a forward reference, first check that + the bracket does occur later so we can give the error (and position) + now. Then remember this forward reference in the workspace so it can + be filled in at the end. */ + + if (lengthptr == NULL) + { + *code = OP_END; if (recno != 0) called = PRIV(find_bracket)(cd->start_code, utf, recno); - - /* Forward reference */ - - if (called == NULL) - { + + /* Forward reference */ + + if (called == NULL) + { if (recno > cd->final_bracount) - { - *errorcodeptr = ERR15; - goto FAILED; - } + { + *errorcodeptr = ERR15; + goto FAILED; + } /* Fudge the value of "called" so that when it is inserted as an offset below, what it actually inserted is the reference number of the group. Then remember the forward reference. */ - called = cd->start_code + recno; + called = cd->start_code + recno; if (cd->hwm >= cd->start_workspace + cd->workspace_size - WORK_SIZE_SAFETY_MARGIN) { @@ -7612,128 +7612,128 @@ for (;; ptr++) if (*errorcodeptr != 0) goto FAILED; } PUTINC(cd->hwm, 0, (int)(code + 1 - cd->start_code)); - } - - /* If not a forward reference, and the subpattern is still open, - this is a recursive call. We check to see if this is a left + } + + /* If not a forward reference, and the subpattern is still open, + this is a recursive call. We check to see if this is a left recursion that could loop for ever, and diagnose that case. We must not, however, do this check if we are in a conditional subpattern because the condition might be testing for recursion in a pattern such as /(?(R)a+|(?R)b)/, which is perfectly valid. Forever loops are also detected at runtime, so those that occur in conditional subpatterns will be picked up then. */ - + else if (GET(called, 1) == 0 && cond_depth <= 0 && could_be_empty(called, code, bcptr, utf, cd)) - { - *errorcodeptr = ERR40; - goto FAILED; - } - } - + { + *errorcodeptr = ERR40; + goto FAILED; + } + } + /* Insert the recursion/subroutine item. It does not have a set first character (relevant if it is repeated, because it will then be wrapped with ONCE brackets). */ - - *code = OP_RECURSE; + + *code = OP_RECURSE; PUT(code, 1, (int)(called - cd->start_code)); - code += 1 + LINK_SIZE; + code += 1 + LINK_SIZE; groupsetfirstchar = FALSE; - } - - /* Can't determine a first byte now */ - + } + + /* Can't determine a first byte now */ + if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE; zerofirstchar = firstchar; zerofirstcharflags = firstcharflags; - continue; - - - /* ------------------------------------------------------------ */ - default: /* Other characters: check option setting */ - OTHER_CHAR_AFTER_QUERY: - set = unset = 0; - optset = &set; - + continue; + + + /* ------------------------------------------------------------ */ + default: /* Other characters: check option setting */ + OTHER_CHAR_AFTER_QUERY: + set = unset = 0; + optset = &set; + while (*ptr != CHAR_RIGHT_PARENTHESIS && *ptr != CHAR_COLON) - { - switch (*ptr++) - { + { + switch (*ptr++) + { case CHAR_MINUS: optset = &unset; break; - + case CHAR_J: /* Record that it changed in the external options */ - *optset |= PCRE_DUPNAMES; - cd->external_flags |= PCRE_JCHANGED; - break; - + *optset |= PCRE_DUPNAMES; + cd->external_flags |= PCRE_JCHANGED; + break; + case CHAR_i: *optset |= PCRE_CASELESS; break; case CHAR_m: *optset |= PCRE_MULTILINE; break; case CHAR_s: *optset |= PCRE_DOTALL; break; case CHAR_x: *optset |= PCRE_EXTENDED; break; case CHAR_U: *optset |= PCRE_UNGREEDY; break; case CHAR_X: *optset |= PCRE_EXTRA; break; - - default: *errorcodeptr = ERR12; - ptr--; /* Correct the offset */ - goto FAILED; - } - } - - /* Set up the changed option bits, but don't change anything yet. */ - - newoptions = (options | set) & (~unset); - - /* If the options ended with ')' this is not the start of a nested + + default: *errorcodeptr = ERR12; + ptr--; /* Correct the offset */ + goto FAILED; + } + } + + /* Set up the changed option bits, but don't change anything yet. */ + + newoptions = (options | set) & (~unset); + + /* If the options ended with ')' this is not the start of a nested group with option changes, so the options change at this level. If we are not at the pattern start, reset the greedy defaults and the case value for firstchar and reqchar. */ - + if (*ptr == CHAR_RIGHT_PARENTHESIS) - { + { greedy_default = ((newoptions & PCRE_UNGREEDY) != 0); greedy_non_default = greedy_default ^ 1; req_caseopt = ((newoptions & PCRE_CASELESS) != 0)? REQ_CASELESS:0; - + /* Change options at this level, and pass them back for use in subsequent branches. */ - + *optionsptr = options = newoptions; - previous = NULL; /* This item can't be repeated */ - continue; /* It is complete */ - } - - /* If the options ended with ':' we are heading into a nested group - with possible change of options. Such groups are non-capturing and are - not assertions of any kind. All we need to do is skip over the ':'; - the newoptions value is handled below. */ - - bravalue = OP_BRA; - ptr++; - } /* End of switch for character following (? */ - } /* End of (? handling */ - + previous = NULL; /* This item can't be repeated */ + continue; /* It is complete */ + } + + /* If the options ended with ':' we are heading into a nested group + with possible change of options. Such groups are non-capturing and are + not assertions of any kind. All we need to do is skip over the ':'; + the newoptions value is handled below. */ + + bravalue = OP_BRA; + ptr++; + } /* End of switch for character following (? */ + } /* End of (? handling */ + /* Opening parenthesis not followed by '*' or '?'. If PCRE_NO_AUTO_CAPTURE is set, all unadorned brackets become non-capturing and behave like (?:...) - brackets. */ - - else if ((options & PCRE_NO_AUTO_CAPTURE) != 0) - { - bravalue = OP_BRA; - } - - /* Else we have a capturing group. */ - - else - { - NUMBERED_GROUP: - cd->bracount += 1; - PUT2(code, 1+LINK_SIZE, cd->bracount); + brackets. */ + + else if ((options & PCRE_NO_AUTO_CAPTURE) != 0) + { + bravalue = OP_BRA; + } + + /* Else we have a capturing group. */ + + else + { + NUMBERED_GROUP: + cd->bracount += 1; + PUT2(code, 1+LINK_SIZE, cd->bracount); skipbytes = IMM2_SIZE; - } - + } + /* Process nested bracketed regex. First check for parentheses nested too deeply. */ - + if ((cd->parens_depth += 1) > PARENS_NEST_LIMIT) { *errorcodeptr = ERR82; @@ -7760,19 +7760,19 @@ for (;; ptr++) item_hwm_offset = cd->hwm - cd->start_workspace; } - *code = bravalue; - tempcode = code; + *code = bravalue; + tempcode = code; tempreqvary = cd->req_varyopt; /* Save value before bracket */ tempbracount = cd->bracount; /* Save value before bracket */ length_prevgroup = 0; /* Initialize for pre-compile phase */ - - if (!compile_regex( + + if (!compile_regex( newoptions, /* The complete new option state */ &tempcode, /* Where to put code (updated) */ &ptr, /* Input pointer (updated) */ errorcodeptr, /* Where to put an error message */ - (bravalue == OP_ASSERTBACK || - bravalue == OP_ASSERTBACK_NOT), /* TRUE if back assert */ + (bravalue == OP_ASSERTBACK || + bravalue == OP_ASSERTBACK_NOT), /* TRUE if back assert */ reset_bracount, /* True if (?| group */ skipbytes, /* Skip over bracket number */ cond_depth + @@ -7785,9 +7785,9 @@ for (;; ptr++) cd, /* Tables block */ (lengthptr == NULL)? NULL : /* Actual compile phase */ &length_prevgroup /* Pre-compile phase */ - )) - goto FAILED; - + )) + goto FAILED; + cd->parens_depth -= 1; /* If this was an atomic group and there are no capturing groups within it, @@ -7799,144 +7799,144 @@ for (;; ptr++) if (bravalue >= OP_ASSERT && bravalue <= OP_ASSERTBACK_NOT) cd->assert_depth -= 1; - /* At the end of compiling, code is still pointing to the start of the + /* At the end of compiling, code is still pointing to the start of the group, while tempcode has been updated to point past the end of the group. The pattern pointer (ptr) is on the bracket. - + If this is a conditional bracket, check that there are no more than - two branches in the group, or just one if it's a DEFINE group. We do this - in the real compile phase, not in the pre-pass, where the whole group may - not be available. */ - - if (bravalue == OP_COND && lengthptr == NULL) - { + two branches in the group, or just one if it's a DEFINE group. We do this + in the real compile phase, not in the pre-pass, where the whole group may + not be available. */ + + if (bravalue == OP_COND && lengthptr == NULL) + { pcre_uchar *tc = code; - int condcount = 0; - - do { - condcount++; - tc += GET(tc,1); - } - while (*tc != OP_KET); - - /* A DEFINE group is never obeyed inline (the "condition" is always - false). It must have only one branch. */ - - if (code[LINK_SIZE+1] == OP_DEF) - { - if (condcount > 1) - { - *errorcodeptr = ERR54; - goto FAILED; - } - bravalue = OP_DEF; /* Just a flag to suppress char handling below */ - } - - /* A "normal" conditional group. If there is just one branch, we must not + int condcount = 0; + + do { + condcount++; + tc += GET(tc,1); + } + while (*tc != OP_KET); + + /* A DEFINE group is never obeyed inline (the "condition" is always + false). It must have only one branch. */ + + if (code[LINK_SIZE+1] == OP_DEF) + { + if (condcount > 1) + { + *errorcodeptr = ERR54; + goto FAILED; + } + bravalue = OP_DEF; /* Just a flag to suppress char handling below */ + } + + /* A "normal" conditional group. If there is just one branch, we must not make use of its firstchar or reqchar, because this is equivalent to an - empty second branch. */ - - else - { - if (condcount > 2) - { - *errorcodeptr = ERR27; - goto FAILED; - } + empty second branch. */ + + else + { + if (condcount > 2) + { + *errorcodeptr = ERR27; + goto FAILED; + } if (condcount == 1) subfirstcharflags = subreqcharflags = REQ_NONE; - } - } - - /* Error if hit end of pattern */ - + } + } + + /* Error if hit end of pattern */ + if (*ptr != CHAR_RIGHT_PARENTHESIS) - { - *errorcodeptr = ERR14; - goto FAILED; - } - - /* In the pre-compile phase, update the length by the length of the group, - less the brackets at either end. Then reduce the compiled code to just a - set of non-capturing brackets so that it doesn't use much memory if it is - duplicated by a quantifier.*/ - - if (lengthptr != NULL) - { - if (OFLOW_MAX - *lengthptr < length_prevgroup - 2 - 2*LINK_SIZE) - { - *errorcodeptr = ERR20; - goto FAILED; - } - *lengthptr += length_prevgroup - 2 - 2*LINK_SIZE; + { + *errorcodeptr = ERR14; + goto FAILED; + } + + /* In the pre-compile phase, update the length by the length of the group, + less the brackets at either end. Then reduce the compiled code to just a + set of non-capturing brackets so that it doesn't use much memory if it is + duplicated by a quantifier.*/ + + if (lengthptr != NULL) + { + if (OFLOW_MAX - *lengthptr < length_prevgroup - 2 - 2*LINK_SIZE) + { + *errorcodeptr = ERR20; + goto FAILED; + } + *lengthptr += length_prevgroup - 2 - 2*LINK_SIZE; code++; /* This already contains bravalue */ - PUTINC(code, 0, 1 + LINK_SIZE); - *code++ = OP_KET; - PUTINC(code, 0, 1 + LINK_SIZE); - break; /* No need to waste time with special character handling */ - } - - /* Otherwise update the main code pointer to the end of the group. */ - - code = tempcode; - - /* For a DEFINE group, required and first character settings are not - relevant. */ - - if (bravalue == OP_DEF) break; - - /* Handle updating of the required and first characters for other types of - group. Update for normal brackets of all kinds, and conditions with two - branches (see code above). If the bracket is followed by a quantifier with + PUTINC(code, 0, 1 + LINK_SIZE); + *code++ = OP_KET; + PUTINC(code, 0, 1 + LINK_SIZE); + break; /* No need to waste time with special character handling */ + } + + /* Otherwise update the main code pointer to the end of the group. */ + + code = tempcode; + + /* For a DEFINE group, required and first character settings are not + relevant. */ + + if (bravalue == OP_DEF) break; + + /* Handle updating of the required and first characters for other types of + group. Update for normal brackets of all kinds, and conditions with two + branches (see code above). If the bracket is followed by a quantifier with zero repeat, we have to back off. Hence the definition of zeroreqchar and zerofirstchar outside the main loop so that they can be accessed for the - back off. */ - + back off. */ + zeroreqchar = reqchar; zeroreqcharflags = reqcharflags; zerofirstchar = firstchar; zerofirstcharflags = firstcharflags; groupsetfirstchar = FALSE; - - if (bravalue >= OP_ONCE) - { + + if (bravalue >= OP_ONCE) + { /* If we have not yet set a firstchar in this branch, take it from the - subpattern, remembering that it was set here so that a repeat of more + subpattern, remembering that it was set here so that a repeat of more than one can replicate it as reqchar if necessary. If the subpattern has no firstchar, set "none" for the whole branch. In both cases, a zero repeat forces firstchar to "none". */ - + if (firstcharflags == REQ_UNSET) - { + { if (subfirstcharflags >= 0) - { + { firstchar = subfirstchar; firstcharflags = subfirstcharflags; groupsetfirstchar = TRUE; - } + } else firstcharflags = REQ_NONE; zerofirstcharflags = REQ_NONE; - } - + } + /* If firstchar was previously set, convert the subpattern's firstchar into reqchar if there wasn't one, using the vary flag that was in - existence beforehand. */ - + existence beforehand. */ + else if (subfirstcharflags >= 0 && subreqcharflags < 0) { subreqchar = subfirstchar; subreqcharflags = subfirstcharflags | tempreqvary; } - - /* If the subpattern set a required byte (or set a first byte that isn't - really the first byte - see above), set it. */ - + + /* If the subpattern set a required byte (or set a first byte that isn't + really the first byte - see above), set it. */ + if (subreqcharflags >= 0) { reqchar = subreqchar; reqcharflags = subreqcharflags; } - } - + } + /* For a forward assertion, we take the reqchar, if set, provided that the group has also set a first char. This can be helpful if the pattern that follows the assertion doesn't set a different char. For example, it's @@ -7945,47 +7945,47 @@ for (;; ptr++) the "real" "a" would then become a reqchar instead of a firstchar. This is overcome by a scan at the end if there's no firstchar, looking for an asserted first char. */ - + else if (bravalue == OP_ASSERT && subreqcharflags >= 0 && subfirstcharflags >= 0) { reqchar = subreqchar; reqcharflags = subreqcharflags; } - break; /* End of processing '(' */ - - - /* ===================================================================*/ - /* Handle metasequences introduced by \. For ones like \d, the ESC_ values + break; /* End of processing '(' */ + + + /* ===================================================================*/ + /* Handle metasequences introduced by \. For ones like \d, the ESC_ values are arranged to be the negation of the corresponding OP_values in the default case when PCRE_UCP is not set. For the back references, the values are negative the reference number. Only back references and those types that consume a character may be repeated. We can test for values between ESC_b and ESC_Z for the latter; this may have to change if any new ones are ever created. */ - + case CHAR_BACKSLASH: - tempptr = ptr; + tempptr = ptr; escape = check_escape(&ptr, &ec, errorcodeptr, cd->bracount, options, FALSE); - if (*errorcodeptr != 0) goto FAILED; - + if (*errorcodeptr != 0) goto FAILED; + if (escape == 0) /* The escape coded a single character */ c = ec; else - { - /* For metasequences that actually match a character, we disable the - setting of a first character if it hasn't already been set. */ - + { + /* For metasequences that actually match a character, we disable the + setting of a first character if it hasn't already been set. */ + if (firstcharflags == REQ_UNSET && escape > ESC_b && escape < ESC_Z) firstcharflags = REQ_NONE; - - /* Set values to reset to if this is followed by a zero repeat. */ - + + /* Set values to reset to if this is followed by a zero repeat. */ + zerofirstchar = firstchar; zerofirstcharflags = firstcharflags; zeroreqchar = reqchar; zeroreqcharflags = reqcharflags; - + /* \g<name> or \g'name' is a subroutine call by name and \g<n> or \g'n' is a subroutine call by number (Oniguruma syntax). In fact, the value ESC_g is returned only for these cases. So we don't need to check for < @@ -8033,44 +8033,44 @@ for (;; ptr++) goto HANDLE_NUMERICAL_RECURSION; } - /* \k<name> or \k'name' is a back reference by name (Perl syntax). + /* \k<name> or \k'name' is a back reference by name (Perl syntax). We also support \k{name} (.NET syntax). */ - + if (escape == ESC_k) - { + { if ((ptr[1] != CHAR_LESS_THAN_SIGN && ptr[1] != CHAR_APOSTROPHE && ptr[1] != CHAR_LEFT_CURLY_BRACKET)) { *errorcodeptr = ERR69; goto FAILED; } - is_recurse = FALSE; + is_recurse = FALSE; terminator = (*(++ptr) == CHAR_LESS_THAN_SIGN)? CHAR_GREATER_THAN_SIGN : (*ptr == CHAR_APOSTROPHE)? CHAR_APOSTROPHE : CHAR_RIGHT_CURLY_BRACKET; - goto NAMED_REF_OR_RECURSE; - } - + goto NAMED_REF_OR_RECURSE; + } + /* Back references are handled specially; must disable firstchar if - not set to cope with cases like (?=(\w+))\1: which would otherwise set - ':' later. */ - + not set to cope with cases like (?=(\w+))\1: which would otherwise set + ':' later. */ + if (escape < 0) - { + { open_capitem *oc; recno = -escape; - + /* Come here from named backref handling when the reference is to a single group (i.e. not to a duplicated name. */ HANDLE_REFERENCE: if (firstcharflags == REQ_UNSET) zerofirstcharflags = firstcharflags = REQ_NONE; - previous = code; + previous = code; item_hwm_offset = cd->hwm - cd->start_workspace; *code++ = ((options & PCRE_CASELESS) != 0)? OP_REFI : OP_REF; - PUT2INC(code, 0, recno); + PUT2INC(code, 0, recno); cd->backref_map |= (recno < 32)? (1U << recno) : 1; - if (recno > cd->top_backref) cd->top_backref = recno; + if (recno > cd->top_backref) cd->top_backref = recno; /* Check to see if this back reference is recursive, that it, it is inside the group that it references. A flag is set so that the @@ -8084,43 +8084,43 @@ for (;; ptr++) break; } } - } - - /* So are Unicode property matches, if supported. */ - -#ifdef SUPPORT_UCP + } + + /* So are Unicode property matches, if supported. */ + +#ifdef SUPPORT_UCP else if (escape == ESC_P || escape == ESC_p) - { - BOOL negated; + { + BOOL negated; unsigned int ptype = 0, pdata = 0; if (!get_ucp(&ptr, &negated, &ptype, &pdata, errorcodeptr)) goto FAILED; - previous = code; + previous = code; item_hwm_offset = cd->hwm - cd->start_workspace; *code++ = ((escape == ESC_p) != negated)? OP_PROP : OP_NOTPROP; - *code++ = ptype; - *code++ = pdata; - } -#else - - /* If Unicode properties are not supported, \X, \P, and \p are not - allowed. */ - + *code++ = ptype; + *code++ = pdata; + } +#else + + /* If Unicode properties are not supported, \X, \P, and \p are not + allowed. */ + else if (escape == ESC_X || escape == ESC_P || escape == ESC_p) - { - *errorcodeptr = ERR45; - goto FAILED; - } -#endif - - /* For the rest (including \X when Unicode properties are supported), we + { + *errorcodeptr = ERR45; + goto FAILED; + } +#endif + + /* For the rest (including \X when Unicode properties are supported), we can obtain the OP value by negating the escape value in the default situation when PCRE_UCP is not set. When it *is* set, we substitute Unicode property tests. Note that \b and \B do a one-character lookbehind, and \A also behaves as if it does. */ - - else - { + + else + { if ((escape == ESC_b || escape == ESC_B || escape == ESC_A) && cd->max_lookbehind == 0) cd->max_lookbehind = 1; @@ -8140,47 +8140,47 @@ for (;; ptr++) item_hwm_offset = cd->hwm - cd->start_workspace; *code++ = (!utf && escape == ESC_C)? OP_ALLANY : escape; } - } - continue; - } - - /* We have a data character whose value is in c. In UTF-8 mode it may have - a value > 127. We set its representation in the length/buffer, and then - handle it as a data character. */ - + } + continue; + } + + /* We have a data character whose value is in c. In UTF-8 mode it may have + a value > 127. We set its representation in the length/buffer, and then + handle it as a data character. */ + #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf && c > MAX_VALUE_FOR_SINGLE_CHAR) mclength = PRIV(ord2utf)(c, mcbuffer); - else -#endif - - { - mcbuffer[0] = c; - mclength = 1; - } - goto ONE_CHAR; - - - /* ===================================================================*/ - /* Handle a literal character. It is guaranteed not to be whitespace or # + else +#endif + + { + mcbuffer[0] = c; + mclength = 1; + } + goto ONE_CHAR; + + + /* ===================================================================*/ + /* Handle a literal character. It is guaranteed not to be whitespace or # when the extended flag is set. If we are in a UTF mode, it may be a multi-unit literal character. */ - - default: - NORMAL_CHAR: - mclength = 1; - mcbuffer[0] = c; - + + default: + NORMAL_CHAR: + mclength = 1; + mcbuffer[0] = c; + #ifdef SUPPORT_UTF if (utf && HAS_EXTRALEN(c)) ACROSSCHAR(TRUE, ptr[1], mcbuffer[mclength++] = *(++ptr)); -#endif - - /* At this point we have the character's bytes in mcbuffer, and the length - in mclength. When not in UTF-8 mode, the length is always 1. */ - - ONE_CHAR: - previous = code; +#endif + + /* At this point we have the character's bytes in mcbuffer, and the length + in mclength. When not in UTF-8 mode, the length is always 1. */ + + ONE_CHAR: + previous = code; item_hwm_offset = cd->hwm - cd->start_workspace; /* For caseless UTF-8 mode when UCP support is available, check whether @@ -8206,29 +8206,29 @@ for (;; ptr++) /* Caseful matches, or not one of the multicase characters. */ *code++ = ((options & PCRE_CASELESS) != 0)? OP_CHARI : OP_CHAR; - for (c = 0; c < mclength; c++) *code++ = mcbuffer[c]; - - /* Remember if \r or \n were seen */ - + for (c = 0; c < mclength; c++) *code++ = mcbuffer[c]; + + /* Remember if \r or \n were seen */ + if (mcbuffer[0] == CHAR_CR || mcbuffer[0] == CHAR_NL) - cd->external_flags |= PCRE_HASCRORLF; - - /* Set the first and required bytes appropriately. If no previous first - byte, set it from this character, but revert to none on a zero repeat. + cd->external_flags |= PCRE_HASCRORLF; + + /* Set the first and required bytes appropriately. If no previous first + byte, set it from this character, but revert to none on a zero repeat. Otherwise, leave the firstchar value alone, and don't change it on a zero - repeat. */ - + repeat. */ + if (firstcharflags == REQ_UNSET) - { + { zerofirstcharflags = REQ_NONE; zeroreqchar = reqchar; zeroreqcharflags = reqcharflags; - + /* If the character is more than one byte long, we can set firstchar - only if it is not to be matched caselessly. */ - - if (mclength == 1 || req_caseopt == 0) - { + only if it is not to be matched caselessly. */ + + if (mclength == 1 || req_caseopt == 0) + { firstchar = mcbuffer[0]; firstcharflags = req_caseopt; @@ -8237,54 +8237,54 @@ for (;; ptr++) reqchar = code[-1]; reqcharflags = cd->req_varyopt; } - } + } else firstcharflags = reqcharflags = REQ_NONE; - } - + } + /* firstchar was previously set; we can set reqchar only if the length is - 1 or the matching is caseful. */ - - else - { + 1 or the matching is caseful. */ + + else + { zerofirstchar = firstchar; zerofirstcharflags = firstcharflags; zeroreqchar = reqchar; zeroreqcharflags = reqcharflags; - if (mclength == 1 || req_caseopt == 0) + if (mclength == 1 || req_caseopt == 0) { reqchar = code[-1]; reqcharflags = req_caseopt | cd->req_varyopt; } - } - - break; /* End of literal character handling */ - } - } /* end of big loop */ - - -/* Control never reaches here by falling through, only by a goto for all the -error states. Pass back the position in the pattern so that it can be displayed -to the user for diagnosing the error. */ - -FAILED: -*ptrptr = ptr; -return FALSE; -} - - - -/************************************************* -* Compile sequence of alternatives * -*************************************************/ - -/* On entry, ptr is pointing past the bracket character, but on return it -points to the closing bracket, or vertical bar, or end of string. The code -variable is pointing at the byte into which the BRA operator has been stored. -This function is used during the pre-compile phase when we are trying to find -out the amount of memory needed, as well as during the real compile phase. The -value of lengthptr distinguishes the two phases. - -Arguments: + } + + break; /* End of literal character handling */ + } + } /* end of big loop */ + + +/* Control never reaches here by falling through, only by a goto for all the +error states. Pass back the position in the pattern so that it can be displayed +to the user for diagnosing the error. */ + +FAILED: +*ptrptr = ptr; +return FALSE; +} + + + +/************************************************* +* Compile sequence of alternatives * +*************************************************/ + +/* On entry, ptr is pointing past the bracket character, but on return it +points to the closing bracket, or vertical bar, or end of string. The code +variable is pointing at the byte into which the BRA operator has been stored. +This function is used during the pre-compile phase when we are trying to find +out the amount of memory needed, as well as during the real compile phase. The +value of lengthptr distinguishes the two phases. + +Arguments: options option bits, including any changes for this subpattern codeptr -> the address of the current code pointer ptrptr -> the address of the current pattern pointer @@ -8301,18 +8301,18 @@ Arguments: cd points to the data block with tables pointers etc. lengthptr NULL during the real compile phase points to length accumulator during pre-compile phase - + Returns: TRUE on success -*/ - -static BOOL +*/ + +static BOOL compile_regex(int options, pcre_uchar **codeptr, const pcre_uchar **ptrptr, - int *errorcodeptr, BOOL lookbehind, BOOL reset_bracount, int skipbytes, + int *errorcodeptr, BOOL lookbehind, BOOL reset_bracount, int skipbytes, int cond_depth, pcre_uint32 *firstcharptr, pcre_int32 *firstcharflagsptr, pcre_uint32 *reqcharptr, pcre_int32 *reqcharflagsptr, branch_chain *bcptr, compile_data *cd, int *lengthptr) -{ +{ const pcre_uchar *ptr = *ptrptr; pcre_uchar *code = *codeptr; pcre_uchar *last_branch = code; @@ -8324,12 +8324,12 @@ pcre_uint32 firstchar, reqchar; pcre_int32 firstcharflags, reqcharflags; pcre_uint32 branchfirstchar, branchreqchar; pcre_int32 branchfirstcharflags, branchreqcharflags; -int length; +int length; unsigned int orig_bracount; unsigned int max_bracount; -branch_chain bc; +branch_chain bc; size_t save_hwm_offset; - + /* If set, call the external function that checks for stack availability. */ if (PUBL(stack_guard) != NULL && PUBL(stack_guard)()) @@ -8340,28 +8340,28 @@ if (PUBL(stack_guard) != NULL && PUBL(stack_guard)()) /* Miscellaneous initialization */ -bc.outer = bcptr; +bc.outer = bcptr; bc.current_branch = code; - + firstchar = reqchar = 0; firstcharflags = reqcharflags = REQ_UNSET; - + save_hwm_offset = cd->hwm - cd->start_workspace; -/* Accumulate the length for use in the pre-compile phase. Start with the -length of the BRA and KET and any extra bytes that are required at the -beginning. We accumulate in a local variable to save frequent testing of -lenthptr for NULL. We cannot do this by looking at the value of code at the -start and end of each alternative, because compiled items are discarded during -the pre-compile phase so that the work space is not exceeded. */ - -length = 2 + 2*LINK_SIZE + skipbytes; - -/* WARNING: If the above line is changed for any reason, you must also change -the code that abstracts option settings at the start of the pattern and makes -them global. It tests the value of length for (2 + 2*LINK_SIZE) in the -pre-compile phase to find out whether anything has yet been compiled or not. */ - +/* Accumulate the length for use in the pre-compile phase. Start with the +length of the BRA and KET and any extra bytes that are required at the +beginning. We accumulate in a local variable to save frequent testing of +lenthptr for NULL. We cannot do this by looking at the value of code at the +start and end of each alternative, because compiled items are discarded during +the pre-compile phase so that the work space is not exceeded. */ + +length = 2 + 2*LINK_SIZE + skipbytes; + +/* WARNING: If the above line is changed for any reason, you must also change +the code that abstracts option settings at the start of the pattern and makes +them global. It tests the value of length for (2 + 2*LINK_SIZE) in the +pre-compile phase to find out whether anything has yet been compiled or not. */ + /* If this is a capturing subpattern, add to the chain of open capturing items so that we can detect them if (*ACCEPT) is encountered. This is also used to detect groups that contain recursive back references to themselves. Note that @@ -8377,95 +8377,95 @@ if (*code == OP_CBRA) cd->open_caps = &capitem; } -/* Offset is set zero to mark that this bracket is still open */ - -PUT(code, 1, 0); -code += 1 + LINK_SIZE + skipbytes; - -/* Loop for each alternative branch */ - -orig_bracount = max_bracount = cd->bracount; -for (;;) - { - /* For a (?| group, reset the capturing bracket count so that each branch - uses the same numbers. */ - - if (reset_bracount) cd->bracount = orig_bracount; - - /* Set up dummy OP_REVERSE if lookbehind assertion */ - - if (lookbehind) - { - *code++ = OP_REVERSE; - reverse_count = code; - PUTINC(code, 0, 0); - length += 1 + LINK_SIZE; - } - - /* Now compile the branch; in the pre-compile phase its length gets added - into the length. */ - +/* Offset is set zero to mark that this bracket is still open */ + +PUT(code, 1, 0); +code += 1 + LINK_SIZE + skipbytes; + +/* Loop for each alternative branch */ + +orig_bracount = max_bracount = cd->bracount; +for (;;) + { + /* For a (?| group, reset the capturing bracket count so that each branch + uses the same numbers. */ + + if (reset_bracount) cd->bracount = orig_bracount; + + /* Set up dummy OP_REVERSE if lookbehind assertion */ + + if (lookbehind) + { + *code++ = OP_REVERSE; + reverse_count = code; + PUTINC(code, 0, 0); + length += 1 + LINK_SIZE; + } + + /* Now compile the branch; in the pre-compile phase its length gets added + into the length. */ + if (!compile_branch(&options, &code, &ptr, errorcodeptr, &branchfirstchar, &branchfirstcharflags, &branchreqchar, &branchreqcharflags, &bc, cond_depth, cd, (lengthptr == NULL)? NULL : &length)) - { - *ptrptr = ptr; - return FALSE; - } - - /* Keep the highest bracket count in case (?| was used and some branch - has fewer than the rest. */ - - if (cd->bracount > max_bracount) max_bracount = cd->bracount; - - /* In the real compile phase, there is some post-processing to be done. */ - - if (lengthptr == NULL) - { + { + *ptrptr = ptr; + return FALSE; + } + + /* Keep the highest bracket count in case (?| was used and some branch + has fewer than the rest. */ + + if (cd->bracount > max_bracount) max_bracount = cd->bracount; + + /* In the real compile phase, there is some post-processing to be done. */ + + if (lengthptr == NULL) + { /* If this is the first branch, the firstchar and reqchar values for the - branch become the values for the regex. */ - - if (*last_branch != OP_ALT) - { + branch become the values for the regex. */ + + if (*last_branch != OP_ALT) + { firstchar = branchfirstchar; firstcharflags = branchfirstcharflags; reqchar = branchreqchar; reqcharflags = branchreqcharflags; - } - + } + /* If this is not the first branch, the first char and reqchar have to - match the values from all the previous branches, except that if the + match the values from all the previous branches, except that if the previous value for reqchar didn't have REQ_VARY set, it can still match, - and we set REQ_VARY for the regex. */ - - else - { + and we set REQ_VARY for the regex. */ + + else + { /* If we previously had a firstchar, but it doesn't match the new branch, we have to abandon the firstchar for the regex, but if there was previously no reqchar, it takes on the value of the old firstchar. */ - + if (firstcharflags >= 0 && (firstcharflags != branchfirstcharflags || firstchar != branchfirstchar)) - { + { if (reqcharflags < 0) { reqchar = firstchar; reqcharflags = firstcharflags; } firstcharflags = REQ_NONE; - } - + } + /* If we (now or from before) have no firstchar, a firstchar from the branch becomes a reqchar if there isn't a branch reqchar. */ - + if (firstcharflags < 0 && branchfirstcharflags >= 0 && branchreqcharflags < 0) { branchreqchar = branchfirstchar; branchreqcharflags = branchfirstcharflags; } - + /* Now ensure that the reqchars match */ - + if (((reqcharflags & ~REQ_VARY) != (branchreqcharflags & ~REQ_VARY)) || reqchar != branchreqchar) reqcharflags = REQ_NONE; @@ -8474,78 +8474,78 @@ for (;;) reqchar = branchreqchar; reqcharflags |= branchreqcharflags; /* To "or" REQ_VARY */ } - } - - /* If lookbehind, check that this branch matches a fixed-length string, and - put the length into the OP_REVERSE item. Temporarily mark the end of the + } + + /* If lookbehind, check that this branch matches a fixed-length string, and + put the length into the OP_REVERSE item. Temporarily mark the end of the branch with OP_END. If the branch contains OP_RECURSE, the result is -3 because there may be forward references that we can't check here. Set a flag to cause another lookbehind check at the end. Why not do it all at the end? Because common, erroneous checks are picked up here and the offset of the problem can be shown. */ - - if (lookbehind) - { - int fixed_length; - *code = OP_END; + + if (lookbehind) + { + int fixed_length; + *code = OP_END; fixed_length = find_fixedlength(last_branch, (options & PCRE_UTF8) != 0, FALSE, cd, NULL); - DPRINTF(("fixed length = %d\n", fixed_length)); + DPRINTF(("fixed length = %d\n", fixed_length)); if (fixed_length == -3) - { + { cd->check_lookbehind = TRUE; } else if (fixed_length < 0) { *errorcodeptr = (fixed_length == -2)? ERR36 : (fixed_length == -4)? ERR70: ERR25; - *ptrptr = ptr; - return FALSE; - } + *ptrptr = ptr; + return FALSE; + } else { if (fixed_length > cd->max_lookbehind) cd->max_lookbehind = fixed_length; PUT(reverse_count, 0, fixed_length); } - } - } - - /* Reached end of expression, either ')' or end of pattern. In the real - compile phase, go back through the alternative branches and reverse the chain - of offsets, with the field in the BRA item now becoming an offset to the - first alternative. If there are no alternatives, it points to the end of the - group. The length in the terminating ket is always the length of the whole + } + } + + /* Reached end of expression, either ')' or end of pattern. In the real + compile phase, go back through the alternative branches and reverse the chain + of offsets, with the field in the BRA item now becoming an offset to the + first alternative. If there are no alternatives, it points to the end of the + group. The length in the terminating ket is always the length of the whole bracketed item. Return leaving the pointer at the terminating char. */ - + if (*ptr != CHAR_VERTICAL_LINE) - { - if (lengthptr == NULL) - { + { + if (lengthptr == NULL) + { int branch_length = (int)(code - last_branch); - do - { - int prev_length = GET(last_branch, 1); - PUT(last_branch, 1, branch_length); - branch_length = prev_length; - last_branch -= branch_length; - } - while (branch_length > 0); - } - - /* Fill in the ket */ - - *code = OP_KET; + do + { + int prev_length = GET(last_branch, 1); + PUT(last_branch, 1, branch_length); + branch_length = prev_length; + last_branch -= branch_length; + } + while (branch_length > 0); + } + + /* Fill in the ket */ + + *code = OP_KET; PUT(code, 1, (int)(code - start_bracket)); - code += 1 + LINK_SIZE; - + code += 1 + LINK_SIZE; + /* If it was a capturing subpattern, check to see if it contained any recursive back references. If so, we must wrap it in atomic brackets. Because we are moving code along, we must ensure that any pending recursive references are updated. In any event, remove the block from the chain. */ - + if (capnumber > 0) - { + { if (cd->open_caps->flag) { *code = OP_END; @@ -8562,139 +8562,139 @@ for (;;) length += 2 + 2*LINK_SIZE; } cd->open_caps = cd->open_caps->next; - } - - /* Retain the highest bracket number, in case resetting was used. */ - - cd->bracount = max_bracount; - - /* Set values to pass back */ - - *codeptr = code; - *ptrptr = ptr; + } + + /* Retain the highest bracket number, in case resetting was used. */ + + cd->bracount = max_bracount; + + /* Set values to pass back */ + + *codeptr = code; + *ptrptr = ptr; *firstcharptr = firstchar; *firstcharflagsptr = firstcharflags; *reqcharptr = reqchar; *reqcharflagsptr = reqcharflags; - if (lengthptr != NULL) - { - if (OFLOW_MAX - *lengthptr < length) - { - *errorcodeptr = ERR20; - return FALSE; - } - *lengthptr += length; - } - return TRUE; - } - - /* Another branch follows. In the pre-compile phase, we can move the code - pointer back to where it was for the start of the first branch. (That is, - pretend that each branch is the only one.) - - In the real compile phase, insert an ALT node. Its length field points back - to the previous branch while the bracket remains open. At the end the chain - is reversed. It's done like this so that the start of the bracket has a - zero offset until it is closed, making it possible to detect recursion. */ - - if (lengthptr != NULL) - { - code = *codeptr + 1 + LINK_SIZE + skipbytes; - length += 1 + LINK_SIZE; - } - else - { - *code = OP_ALT; + if (lengthptr != NULL) + { + if (OFLOW_MAX - *lengthptr < length) + { + *errorcodeptr = ERR20; + return FALSE; + } + *lengthptr += length; + } + return TRUE; + } + + /* Another branch follows. In the pre-compile phase, we can move the code + pointer back to where it was for the start of the first branch. (That is, + pretend that each branch is the only one.) + + In the real compile phase, insert an ALT node. Its length field points back + to the previous branch while the bracket remains open. At the end the chain + is reversed. It's done like this so that the start of the bracket has a + zero offset until it is closed, making it possible to detect recursion. */ + + if (lengthptr != NULL) + { + code = *codeptr + 1 + LINK_SIZE + skipbytes; + length += 1 + LINK_SIZE; + } + else + { + *code = OP_ALT; PUT(code, 1, (int)(code - last_branch)); bc.current_branch = last_branch = code; - code += 1 + LINK_SIZE; - } - - ptr++; - } -/* Control never reaches here */ -} - - - - -/************************************************* -* Check for anchored expression * -*************************************************/ - -/* Try to find out if this is an anchored regular expression. Consider each -alternative branch. If they all start with OP_SOD or OP_CIRC, or with a bracket -all of whose alternatives start with OP_SOD or OP_CIRC (recurse ad lib), then + code += 1 + LINK_SIZE; + } + + ptr++; + } +/* Control never reaches here */ +} + + + + +/************************************************* +* Check for anchored expression * +*************************************************/ + +/* Try to find out if this is an anchored regular expression. Consider each +alternative branch. If they all start with OP_SOD or OP_CIRC, or with a bracket +all of whose alternatives start with OP_SOD or OP_CIRC (recurse ad lib), then it's anchored. However, if this is a multiline pattern, then only OP_SOD will be found, because ^ generates OP_CIRCM in that mode. - -We can also consider a regex to be anchored if OP_SOM starts all its branches. -This is the code for \G, which means "match at start of match position, taking -into account the match offset". - -A branch is also implicitly anchored if it starts with .* and DOTALL is set, -because that will try the rest of the pattern at all possible matching points, -so there is no point trying again.... er .... - -.... except when the .* appears inside capturing parentheses, and there is a -subsequent back reference to those parentheses. We haven't enough information -to catch that case precisely. - -At first, the best we could do was to detect when .* was in capturing brackets -and the highest back reference was greater than or equal to that level. -However, by keeping a bitmap of the first 31 back references, we can catch some -of the more common cases more precisely. - + +We can also consider a regex to be anchored if OP_SOM starts all its branches. +This is the code for \G, which means "match at start of match position, taking +into account the match offset". + +A branch is also implicitly anchored if it starts with .* and DOTALL is set, +because that will try the rest of the pattern at all possible matching points, +so there is no point trying again.... er .... + +.... except when the .* appears inside capturing parentheses, and there is a +subsequent back reference to those parentheses. We haven't enough information +to catch that case precisely. + +At first, the best we could do was to detect when .* was in capturing brackets +and the highest back reference was greater than or equal to that level. +However, by keeping a bitmap of the first 31 back references, we can catch some +of the more common cases more precisely. + ... A second exception is when the .* appears inside an atomic group, because this prevents the number of characters it matches from being adjusted. -Arguments: - code points to start of expression (the bracket) - bracket_map a bitmap of which brackets we are inside while testing; this - handles up to substring 31; after that we just have to take - the less precise approach +Arguments: + code points to start of expression (the bracket) + bracket_map a bitmap of which brackets we are inside while testing; this + handles up to substring 31; after that we just have to take + the less precise approach cd points to the compile data block atomcount atomic group level - -Returns: TRUE or FALSE -*/ - -static BOOL + +Returns: TRUE or FALSE +*/ + +static BOOL is_anchored(register const pcre_uchar *code, unsigned int bracket_map, compile_data *cd, int atomcount) -{ -do { +{ +do { const pcre_uchar *scode = first_significant_code( code + PRIV(OP_lengths)[*code], FALSE); register int op = *scode; - - /* Non-capturing brackets */ - + + /* Non-capturing brackets */ + if (op == OP_BRA || op == OP_BRAPOS || op == OP_SBRA || op == OP_SBRAPOS) - { + { if (!is_anchored(scode, bracket_map, cd, atomcount)) return FALSE; - } - - /* Capturing brackets */ - + } + + /* Capturing brackets */ + else if (op == OP_CBRA || op == OP_CBRAPOS || op == OP_SCBRA || op == OP_SCBRAPOS) - { - int n = GET2(scode, 1+LINK_SIZE); + { + int n = GET2(scode, 1+LINK_SIZE); int new_map = bracket_map | ((n < 32)? (1U << n) : 1); if (!is_anchored(scode, new_map, cd, atomcount)) return FALSE; - } - + } + /* Positive forward assertion */ - + else if (op == OP_ASSERT) - { + { if (!is_anchored(scode, bracket_map, cd, atomcount)) return FALSE; - } - + } + /* Condition; not anchored if no second branch */ - + else if (op == OP_COND) { if (scode[GET(scode,1)] != OP_ALT) return FALSE; @@ -8713,60 +8713,60 @@ do { it isn't in brackets that are or may be referenced or inside an atomic group. */ - else if ((op == OP_TYPESTAR || op == OP_TYPEMINSTAR || + else if ((op == OP_TYPESTAR || op == OP_TYPEMINSTAR || op == OP_TYPEPOSSTAR)) - { + { if (scode[1] != OP_ALLANY || (bracket_map & cd->backref_map) != 0 || atomcount > 0 || cd->had_pruneorskip) return FALSE; - } - - /* Check for explicit anchoring */ - + } + + /* Check for explicit anchoring */ + else if (op != OP_SOD && op != OP_SOM && op != OP_CIRC) return FALSE; - code += GET(code, 1); - } -while (*code == OP_ALT); /* Loop for each alternative */ -return TRUE; -} - - - -/************************************************* -* Check for starting with ^ or .* * -*************************************************/ - -/* This is called to find out if every branch starts with ^ or .* so that -"first char" processing can be done to speed things up in multiline -matching and for non-DOTALL patterns that start with .* (which must start at -the beginning or after \n). As in the case of is_anchored() (see above), we -have to take account of back references to capturing brackets that contain .* + code += GET(code, 1); + } +while (*code == OP_ALT); /* Loop for each alternative */ +return TRUE; +} + + + +/************************************************* +* Check for starting with ^ or .* * +*************************************************/ + +/* This is called to find out if every branch starts with ^ or .* so that +"first char" processing can be done to speed things up in multiline +matching and for non-DOTALL patterns that start with .* (which must start at +the beginning or after \n). As in the case of is_anchored() (see above), we +have to take account of back references to capturing brackets that contain .* because in that case we can't make the assumption. Also, the appearance of .* inside atomic brackets or in an assertion, or in a pattern that contains *PRUNE or *SKIP does not count, because once again the assumption no longer holds. - -Arguments: - code points to start of expression (the bracket) - bracket_map a bitmap of which brackets we are inside while testing; this - handles up to substring 31; after that we just have to take - the less precise approach + +Arguments: + code points to start of expression (the bracket) + bracket_map a bitmap of which brackets we are inside while testing; this + handles up to substring 31; after that we just have to take + the less precise approach cd points to the compile data atomcount atomic group level inassert TRUE if in an assertion - -Returns: TRUE or FALSE -*/ - -static BOOL + +Returns: TRUE or FALSE +*/ + +static BOOL is_startline(const pcre_uchar *code, unsigned int bracket_map, compile_data *cd, int atomcount, BOOL inassert) -{ -do { +{ +do { const pcre_uchar *scode = first_significant_code( code + PRIV(OP_lengths)[*code], FALSE); register int op = *scode; - + /* If we are at the start of a conditional assertion group, *both* the conditional assertion *and* what follows the condition must satisfy the test for start of line. Other kinds of condition fail. Note that there may be an @@ -8796,33 +8796,33 @@ do { op = *scode; } - /* Non-capturing brackets */ - + /* Non-capturing brackets */ + if (op == OP_BRA || op == OP_BRAPOS || op == OP_SBRA || op == OP_SBRAPOS) - { + { if (!is_startline(scode, bracket_map, cd, atomcount, inassert)) return FALSE; - } - - /* Capturing brackets */ - + } + + /* Capturing brackets */ + else if (op == OP_CBRA || op == OP_CBRAPOS || op == OP_SCBRA || op == OP_SCBRAPOS) - { - int n = GET2(scode, 1+LINK_SIZE); + { + int n = GET2(scode, 1+LINK_SIZE); int new_map = bracket_map | ((n < 32)? (1U << n) : 1); if (!is_startline(scode, new_map, cd, atomcount, inassert)) return FALSE; - } - + } + /* Positive forward assertions */ - + else if (op == OP_ASSERT) { if (!is_startline(scode, bracket_map, cd, atomcount, TRUE)) return FALSE; } - + /* Atomic brackets */ - + else if (op == OP_ONCE || op == OP_ONCE_NC) { if (!is_startline(scode, bracket_map, cd, atomcount + 1, inassert)) return FALSE; @@ -8834,60 +8834,60 @@ do { example, /.*?a(*PRUNE)b/ with the subject "aab", which matches "ab", i.e. not at the start of a line. */ - else if (op == OP_TYPESTAR || op == OP_TYPEMINSTAR || op == OP_TYPEPOSSTAR) - { + else if (op == OP_TYPESTAR || op == OP_TYPEMINSTAR || op == OP_TYPEPOSSTAR) + { if (scode[1] != OP_ANY || (bracket_map & cd->backref_map) != 0 || atomcount > 0 || cd->had_pruneorskip || inassert) return FALSE; - } - + } + /* Check for explicit circumflex; anything else gives a FALSE result. Note in particular that this includes atomic brackets OP_ONCE and OP_ONCE_NC because the number of characters matched by .* cannot be adjusted inside them. */ - + else if (op != OP_CIRC && op != OP_CIRCM) return FALSE; - - /* Move on to the next alternative */ - - code += GET(code, 1); - } -while (*code == OP_ALT); /* Loop for each alternative */ -return TRUE; -} - - - -/************************************************* -* Check for asserted fixed first char * -*************************************************/ - -/* During compilation, the "first char" settings from forward assertions are -discarded, because they can cause conflicts with actual literals that follow. -However, if we end up without a first char setting for an unanchored pattern, -it is worth scanning the regex to see if there is an initial asserted first + + /* Move on to the next alternative */ + + code += GET(code, 1); + } +while (*code == OP_ALT); /* Loop for each alternative */ +return TRUE; +} + + + +/************************************************* +* Check for asserted fixed first char * +*************************************************/ + +/* During compilation, the "first char" settings from forward assertions are +discarded, because they can cause conflicts with actual literals that follow. +However, if we end up without a first char setting for an unanchored pattern, +it is worth scanning the regex to see if there is an initial asserted first char. If all branches start with the same asserted char, or with a non-conditional bracket all of whose alternatives start with the same asserted char (recurse ad lib), then we return that char, with the flags set to zero or REQ_CASELESS; otherwise return zero with REQ_NONE in the flags. - -Arguments: - code points to start of expression (the bracket) + +Arguments: + code points to start of expression (the bracket) flags points to the first char flags, or to REQ_NONE - inassert TRUE if in an assertion - + inassert TRUE if in an assertion + Returns: the fixed first char, or 0 with REQ_NONE in flags -*/ - +*/ + static pcre_uint32 find_firstassertedchar(const pcre_uchar *code, pcre_int32 *flags, BOOL inassert) -{ +{ register pcre_uint32 c = 0; int cflags = REQ_NONE; *flags = REQ_NONE; -do { +do { pcre_uint32 d; int dflags; int xl = (*code == OP_CBRA || *code == OP_SCBRA || @@ -8895,39 +8895,39 @@ do { const pcre_uchar *scode = first_significant_code(code + 1+LINK_SIZE + xl, TRUE); register pcre_uchar op = *scode; - - switch(op) - { - default: + + switch(op) + { + default: return 0; - - case OP_BRA: + + case OP_BRA: case OP_BRAPOS: - case OP_CBRA: + case OP_CBRA: case OP_SCBRA: case OP_CBRAPOS: case OP_SCBRAPOS: - case OP_ASSERT: - case OP_ONCE: + case OP_ASSERT: + case OP_ONCE: case OP_ONCE_NC: d = find_firstassertedchar(scode, &dflags, op == OP_ASSERT); if (dflags < 0) return 0; if (cflags < 0) { c = d; cflags = dflags; } else if (c != d || cflags != dflags) return 0; - break; - + break; + case OP_EXACT: scode += IMM2_SIZE; /* Fall through */ - - case OP_CHAR: - case OP_PLUS: - case OP_MINPLUS: - case OP_POSPLUS: + + case OP_CHAR: + case OP_PLUS: + case OP_MINPLUS: + case OP_POSPLUS: if (!inassert) return 0; if (cflags < 0) { c = scode[1]; cflags = 0; } else if (c != scode[1]) return 0; - break; + break; case OP_EXACTI: scode += IMM2_SIZE; @@ -8941,19 +8941,19 @@ do { if (cflags < 0) { c = scode[1]; cflags = REQ_CASELESS; } else if (c != scode[1]) return 0; break; - } - - code += GET(code, 1); - } -while (*code == OP_ALT); + } + + code += GET(code, 1); + } +while (*code == OP_ALT); *flags = cflags; -return c; -} - - - -/************************************************* +return c; +} + + + +/************************************************* * Add an entry to the name/number table * *************************************************/ @@ -9009,31 +9009,31 @@ cd->names_found++; /************************************************* -* Compile a Regular Expression * -*************************************************/ - -/* This function takes a string and returns a pointer to a block of store -holding a compiled version of the expression. The original API for this -function had no error code return variable; it is retained for backwards -compatibility. The new function is given a new name. - -Arguments: - pattern the regular expression - options various option bits - errorcodeptr pointer to error code variable (pcre_compile2() only) - can be NULL if you don't want a code value - errorptr pointer to pointer to error text - erroroffset ptr offset in pattern where error was detected - tables pointer to character tables or NULL - -Returns: pointer to compiled data block, or NULL on error, - with errorptr and erroroffset set -*/ - +* Compile a Regular Expression * +*************************************************/ + +/* This function takes a string and returns a pointer to a block of store +holding a compiled version of the expression. The original API for this +function had no error code return variable; it is retained for backwards +compatibility. The new function is given a new name. + +Arguments: + pattern the regular expression + options various option bits + errorcodeptr pointer to error code variable (pcre_compile2() only) + can be NULL if you don't want a code value + errorptr pointer to pointer to error text + erroroffset ptr offset in pattern where error was detected + tables pointer to character tables or NULL + +Returns: pointer to compiled data block, or NULL on error, + with errorptr and erroroffset set +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN pcre * PCRE_CALL_CONVENTION -pcre_compile(const char *pattern, int options, const char **errorptr, - int *erroroffset, const unsigned char *tables) +pcre_compile(const char *pattern, int options, const char **errorptr, + int *erroroffset, const unsigned char *tables) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN pcre16 * PCRE_CALL_CONVENTION pcre16_compile(PCRE_SPTR16 pattern, int options, const char **errorptr, @@ -9043,21 +9043,21 @@ PCRE_EXP_DEFN pcre32 * PCRE_CALL_CONVENTION pcre32_compile(PCRE_SPTR32 pattern, int options, const char **errorptr, int *erroroffset, const unsigned char *tables) #endif -{ +{ #if defined COMPILE_PCRE8 -return pcre_compile2(pattern, options, NULL, errorptr, erroroffset, tables); +return pcre_compile2(pattern, options, NULL, errorptr, erroroffset, tables); #elif defined COMPILE_PCRE16 return pcre16_compile2(pattern, options, NULL, errorptr, erroroffset, tables); #elif defined COMPILE_PCRE32 return pcre32_compile2(pattern, options, NULL, errorptr, erroroffset, tables); #endif -} - - +} + + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN pcre * PCRE_CALL_CONVENTION -pcre_compile2(const char *pattern, int options, int *errorcodeptr, - const char **errorptr, int *erroroffset, const unsigned char *tables) +pcre_compile2(const char *pattern, int options, int *errorcodeptr, + const char **errorptr, int *erroroffset, const unsigned char *tables) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN pcre16 * PCRE_CALL_CONVENTION pcre16_compile2(PCRE_SPTR16 pattern, int options, int *errorcodeptr, @@ -9067,97 +9067,97 @@ PCRE_EXP_DEFN pcre32 * PCRE_CALL_CONVENTION pcre32_compile2(PCRE_SPTR32 pattern, int options, int *errorcodeptr, const char **errorptr, int *erroroffset, const unsigned char *tables) #endif -{ +{ REAL_PCRE *re; -int length = 1; /* For final END opcode */ +int length = 1; /* For final END opcode */ pcre_int32 firstcharflags, reqcharflags; pcre_uint32 firstchar, reqchar; pcre_uint32 limit_match = PCRE_UINT32_MAX; pcre_uint32 limit_recursion = PCRE_UINT32_MAX; int newline; -int errorcode = 0; -int skipatstart = 0; +int errorcode = 0; +int skipatstart = 0; BOOL utf; BOOL never_utf = FALSE; -size_t size; +size_t size; pcre_uchar *code; const pcre_uchar *codestart; const pcre_uchar *ptr; -compile_data compile_block; -compile_data *cd = &compile_block; - -/* This space is used for "compiling" into during the first phase, when we are -computing the amount of memory that is needed. Compiled items are thrown away -as soon as possible, so that a fairly large buffer should be sufficient for -this purpose. The same space is used in the second phase for remembering where +compile_data compile_block; +compile_data *cd = &compile_block; + +/* This space is used for "compiling" into during the first phase, when we are +computing the amount of memory that is needed. Compiled items are thrown away +as soon as possible, so that a fairly large buffer should be sufficient for +this purpose. The same space is used in the second phase for remembering where to fill in forward references to subpatterns. That may overflow, in which case new memory is obtained from malloc(). */ - + pcre_uchar cworkspace[COMPILE_WORK_SIZE]; - + /* This vector is used for remembering name groups during the pre-compile. In a similar way to cworkspace, it can be expanded using malloc() if necessary. */ - + named_group named_groups[NAMED_GROUP_LIST_SIZE]; -/* Set this early so that early errors get offset 0. */ - +/* Set this early so that early errors get offset 0. */ + ptr = (const pcre_uchar *)pattern; - -/* We can't pass back an error message if errorptr is NULL; I guess the best we -can do is just return NULL, but we can set a code value if there is a code -pointer. */ - -if (errorptr == NULL) - { - if (errorcodeptr != NULL) *errorcodeptr = 99; - return NULL; - } - -*errorptr = NULL; -if (errorcodeptr != NULL) *errorcodeptr = ERR0; - -/* However, we can give a message for this error */ - -if (erroroffset == NULL) - { - errorcode = ERR16; - goto PCRE_EARLY_ERROR_RETURN2; - } - -*erroroffset = 0; - + +/* We can't pass back an error message if errorptr is NULL; I guess the best we +can do is just return NULL, but we can set a code value if there is a code +pointer. */ + +if (errorptr == NULL) + { + if (errorcodeptr != NULL) *errorcodeptr = 99; + return NULL; + } + +*errorptr = NULL; +if (errorcodeptr != NULL) *errorcodeptr = ERR0; + +/* However, we can give a message for this error */ + +if (erroroffset == NULL) + { + errorcode = ERR16; + goto PCRE_EARLY_ERROR_RETURN2; + } + +*erroroffset = 0; + /* Set up pointers to the individual character tables */ - + if (tables == NULL) tables = PRIV(default_tables); cd->lcc = tables + lcc_offset; cd->fcc = tables + fcc_offset; cd->cbits = tables + cbits_offset; cd->ctypes = tables + ctypes_offset; - + /* Check that all undefined public option bits are zero */ if ((options & ~PUBLIC_COMPILE_OPTIONS) != 0) - { - errorcode = ERR17; - goto PCRE_EARLY_ERROR_RETURN; - } - + { + errorcode = ERR17; + goto PCRE_EARLY_ERROR_RETURN; + } + /* If PCRE_NEVER_UTF is set, remember it. */ - + if ((options & PCRE_NEVER_UTF) != 0) never_utf = TRUE; - -/* Check for global one-time settings at the start of the pattern, and remember -the offset for later. */ - + +/* Check for global one-time settings at the start of the pattern, and remember +the offset for later. */ + cd->external_flags = 0; /* Initialize here for LIMIT_MATCH/RECURSION */ while (ptr[skipatstart] == CHAR_LEFT_PARENTHESIS && ptr[skipatstart+1] == CHAR_ASTERISK) - { - int newnl = 0; - int newbsr = 0; - + { + int newnl = 0; + int newbsr = 0; + /* For completeness and backward compatibility, (*UTFn) is supported in the relevant libraries, but (*UTF) is generic and always supported. Note that PCRE_UTF8 == PCRE_UTF16 == PCRE_UTF32. */ @@ -9223,28 +9223,28 @@ PCRE_UTF8 == PCRE_UTF16 == PCRE_UTF32. */ } if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_CR_RIGHTPAR, 3) == 0) - { skipatstart += 5; newnl = PCRE_NEWLINE_CR; } + { skipatstart += 5; newnl = PCRE_NEWLINE_CR; } else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_LF_RIGHTPAR, 3) == 0) - { skipatstart += 5; newnl = PCRE_NEWLINE_LF; } + { skipatstart += 5; newnl = PCRE_NEWLINE_LF; } else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_CRLF_RIGHTPAR, 5) == 0) - { skipatstart += 7; newnl = PCRE_NEWLINE_CR + PCRE_NEWLINE_LF; } + { skipatstart += 7; newnl = PCRE_NEWLINE_CR + PCRE_NEWLINE_LF; } else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_ANY_RIGHTPAR, 4) == 0) - { skipatstart += 6; newnl = PCRE_NEWLINE_ANY; } + { skipatstart += 6; newnl = PCRE_NEWLINE_ANY; } else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_ANYCRLF_RIGHTPAR, 8) == 0) - { skipatstart += 10; newnl = PCRE_NEWLINE_ANYCRLF; } - + { skipatstart += 10; newnl = PCRE_NEWLINE_ANYCRLF; } + else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_BSR_ANYCRLF_RIGHTPAR, 12) == 0) - { skipatstart += 14; newbsr = PCRE_BSR_ANYCRLF; } + { skipatstart += 14; newbsr = PCRE_BSR_ANYCRLF; } else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_BSR_UNICODE_RIGHTPAR, 12) == 0) - { skipatstart += 14; newbsr = PCRE_BSR_UNICODE; } - - if (newnl != 0) - options = (options & ~PCRE_NEWLINE_BITS) | newnl; - else if (newbsr != 0) - options = (options & ~(PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) | newbsr; - else break; - } - + { skipatstart += 14; newbsr = PCRE_BSR_UNICODE; } + + if (newnl != 0) + options = (options & ~PCRE_NEWLINE_BITS) | newnl; + else if (newbsr != 0) + options = (options & ~(PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) | newbsr; + else break; + } + /* PCRE_UTF(16|32) have the same value as PCRE_UTF8. */ utf = (options & PCRE_UTF8) != 0; if (utf && never_utf) @@ -9289,86 +9289,86 @@ if ((options & PCRE_UCP) != 0) } #endif -/* Check validity of \R options. */ - +/* Check validity of \R options. */ + if ((options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) == (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) - { + { errorcode = ERR56; goto PCRE_EARLY_ERROR_RETURN; - } - -/* Handle different types of newline. The three bits give seven cases. The -current code allows for fixed one- or two-byte sequences, plus "any" and -"anycrlf". */ - -switch (options & PCRE_NEWLINE_BITS) - { - case 0: newline = NEWLINE; break; /* Build-time default */ + } + +/* Handle different types of newline. The three bits give seven cases. The +current code allows for fixed one- or two-byte sequences, plus "any" and +"anycrlf". */ + +switch (options & PCRE_NEWLINE_BITS) + { + case 0: newline = NEWLINE; break; /* Build-time default */ case PCRE_NEWLINE_CR: newline = CHAR_CR; break; case PCRE_NEWLINE_LF: newline = CHAR_NL; break; - case PCRE_NEWLINE_CR+ + case PCRE_NEWLINE_CR+ PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break; - case PCRE_NEWLINE_ANY: newline = -1; break; - case PCRE_NEWLINE_ANYCRLF: newline = -2; break; - default: errorcode = ERR56; goto PCRE_EARLY_ERROR_RETURN; - } - -if (newline == -2) - { - cd->nltype = NLTYPE_ANYCRLF; - } -else if (newline < 0) - { - cd->nltype = NLTYPE_ANY; - } -else - { - cd->nltype = NLTYPE_FIXED; - if (newline > 255) - { - cd->nllen = 2; - cd->nl[0] = (newline >> 8) & 255; - cd->nl[1] = newline & 255; - } - else - { - cd->nllen = 1; - cd->nl[0] = newline; - } - } - -/* Maximum back reference and backref bitmap. The bitmap records up to 31 back -references to help in deciding whether (.*) can be treated as anchored or not. -*/ - -cd->top_backref = 0; -cd->backref_map = 0; - -/* Reflect pattern for debugging output */ - -DPRINTF(("------------------------------------------------------------------\n")); + case PCRE_NEWLINE_ANY: newline = -1; break; + case PCRE_NEWLINE_ANYCRLF: newline = -2; break; + default: errorcode = ERR56; goto PCRE_EARLY_ERROR_RETURN; + } + +if (newline == -2) + { + cd->nltype = NLTYPE_ANYCRLF; + } +else if (newline < 0) + { + cd->nltype = NLTYPE_ANY; + } +else + { + cd->nltype = NLTYPE_FIXED; + if (newline > 255) + { + cd->nllen = 2; + cd->nl[0] = (newline >> 8) & 255; + cd->nl[1] = newline & 255; + } + else + { + cd->nllen = 1; + cd->nl[0] = newline; + } + } + +/* Maximum back reference and backref bitmap. The bitmap records up to 31 back +references to help in deciding whether (.*) can be treated as anchored or not. +*/ + +cd->top_backref = 0; +cd->backref_map = 0; + +/* Reflect pattern for debugging output */ + +DPRINTF(("------------------------------------------------------------------\n")); #ifdef PCRE_DEBUG print_puchar(stdout, (PCRE_PUCHAR)pattern); #endif DPRINTF(("\n")); - -/* Pretend to compile the pattern while actually just accumulating the length -of memory required. This behaviour is triggered by passing a non-NULL final -argument to compile_regex(). We pass a block of workspace (cworkspace) for it -to compile parts of the pattern into; the compiled code is discarded when it is -no longer needed, so hopefully this workspace will never overflow, though there -is a test for its doing so. */ - -cd->bracount = cd->final_bracount = 0; -cd->names_found = 0; -cd->name_entry_size = 0; -cd->name_table = NULL; + +/* Pretend to compile the pattern while actually just accumulating the length +of memory required. This behaviour is triggered by passing a non-NULL final +argument to compile_regex(). We pass a block of workspace (cworkspace) for it +to compile parts of the pattern into; the compiled code is discarded when it is +no longer needed, so hopefully this workspace will never overflow, though there +is a test for its doing so. */ + +cd->bracount = cd->final_bracount = 0; +cd->names_found = 0; +cd->name_entry_size = 0; +cd->name_table = NULL; cd->dupnames = FALSE; cd->dupgroups = FALSE; cd->namedrefcount = 0; -cd->start_code = cworkspace; -cd->hwm = cworkspace; +cd->start_code = cworkspace; +cd->hwm = cworkspace; cd->iscondassert = FALSE; cd->start_workspace = cworkspace; cd->workspace_size = COMPILE_WORK_SIZE; @@ -9376,102 +9376,102 @@ cd->named_groups = named_groups; cd->named_group_list_size = NAMED_GROUP_LIST_SIZE; cd->start_pattern = (const pcre_uchar *)pattern; cd->end_pattern = (const pcre_uchar *)(pattern + STRLEN_UC((const pcre_uchar *)pattern)); -cd->req_varyopt = 0; +cd->req_varyopt = 0; cd->parens_depth = 0; cd->assert_depth = 0; cd->max_lookbehind = 0; -cd->external_options = options; +cd->external_options = options; cd->open_caps = NULL; - -/* Now do the pre-compile. On error, errorcode will be set non-zero, so we -don't need to look at the result of the function here. The initial options have -been put into the cd block so that they can be changed if an option setting is -found within the regex right at the beginning. Bringing initial option settings -outside can help speed up starting point checks. */ - -ptr += skipatstart; -code = cworkspace; -*code = OP_BRA; + +/* Now do the pre-compile. On error, errorcode will be set non-zero, so we +don't need to look at the result of the function here. The initial options have +been put into the cd block so that they can be changed if an option setting is +found within the regex right at the beginning. Bringing initial option settings +outside can help speed up starting point checks. */ + +ptr += skipatstart; +code = cworkspace; +*code = OP_BRA; (void)compile_regex(cd->external_options, &code, &ptr, &errorcode, FALSE, FALSE, 0, 0, &firstchar, &firstcharflags, &reqchar, &reqcharflags, NULL, cd, &length); -if (errorcode != 0) goto PCRE_EARLY_ERROR_RETURN; - -DPRINTF(("end pre-compile: length=%d workspace=%d\n", length, +if (errorcode != 0) goto PCRE_EARLY_ERROR_RETURN; + +DPRINTF(("end pre-compile: length=%d workspace=%d\n", length, (int)(cd->hwm - cworkspace))); - -if (length > MAX_PATTERN_SIZE) - { - errorcode = ERR20; - goto PCRE_EARLY_ERROR_RETURN; - } - + +if (length > MAX_PATTERN_SIZE) + { + errorcode = ERR20; + goto PCRE_EARLY_ERROR_RETURN; + } + /* Compute the size of the data block for storing the compiled pattern. Integer overflow should no longer be possible because nowadays we limit the maximum value of cd->names_found and cd->name_entry_size. */ - + size = sizeof(REAL_PCRE) + (length + cd->names_found * cd->name_entry_size) * sizeof(pcre_uchar); - + /* Get the memory. */ re = (REAL_PCRE *)(PUBL(malloc))(size); -if (re == NULL) - { - errorcode = ERR21; - goto PCRE_EARLY_ERROR_RETURN; - } - -/* Put in the magic number, and save the sizes, initial options, internal -flags, and character table pointer. NULL is used for the default character -tables. The nullpad field is at the end; it's there to help in the case when a -regex compiled on a system with 4-byte pointers is run on another with 8-byte -pointers. */ - -re->magic_number = MAGIC_NUMBER; +if (re == NULL) + { + errorcode = ERR21; + goto PCRE_EARLY_ERROR_RETURN; + } + +/* Put in the magic number, and save the sizes, initial options, internal +flags, and character table pointer. NULL is used for the default character +tables. The nullpad field is at the end; it's there to help in the case when a +regex compiled on a system with 4-byte pointers is run on another with 8-byte +pointers. */ + +re->magic_number = MAGIC_NUMBER; re->size = (int)size; -re->options = cd->external_options; -re->flags = cd->external_flags; +re->options = cd->external_options; +re->flags = cd->external_flags; re->limit_match = limit_match; re->limit_recursion = limit_recursion; re->first_char = 0; re->req_char = 0; re->name_table_offset = sizeof(REAL_PCRE) / sizeof(pcre_uchar); -re->name_entry_size = cd->name_entry_size; -re->name_count = cd->names_found; -re->ref_count = 0; +re->name_entry_size = cd->name_entry_size; +re->name_count = cd->names_found; +re->ref_count = 0; re->tables = (tables == PRIV(default_tables))? NULL : tables; -re->nullpad = NULL; +re->nullpad = NULL; #ifdef COMPILE_PCRE32 re->dummy = 0; #else re->dummy1 = re->dummy2 = re->dummy3 = 0; #endif - -/* The starting points of the name/number translation table and of the code are -passed around in the compile data block. The start/end pattern and initial -options are already set from the pre-compile phase, as is the name_entry_size -field. Reset the bracket count and the names_found field. Also reset the hwm -field; this time it's used for remembering forward references to subpatterns. -*/ - -cd->final_bracount = cd->bracount; /* Save for checking forward references */ + +/* The starting points of the name/number translation table and of the code are +passed around in the compile data block. The start/end pattern and initial +options are already set from the pre-compile phase, as is the name_entry_size +field. Reset the bracket count and the names_found field. Also reset the hwm +field; this time it's used for remembering forward references to subpatterns. +*/ + +cd->final_bracount = cd->bracount; /* Save for checking forward references */ cd->parens_depth = 0; cd->assert_depth = 0; -cd->bracount = 0; +cd->bracount = 0; cd->max_lookbehind = 0; cd->name_table = (pcre_uchar *)re + re->name_table_offset; -codestart = cd->name_table + re->name_entry_size * re->name_count; -cd->start_code = codestart; +codestart = cd->name_table + re->name_entry_size * re->name_count; +cd->start_code = codestart; cd->hwm = (pcre_uchar *)(cd->start_workspace); cd->iscondassert = FALSE; -cd->req_varyopt = 0; -cd->had_accept = FALSE; +cd->req_varyopt = 0; +cd->had_accept = FALSE; cd->had_pruneorskip = FALSE; cd->check_lookbehind = FALSE; cd->open_caps = NULL; - + /* If any named groups were found, create the name/number table from the list created in the first pass. */ @@ -9486,51 +9486,51 @@ if (cd->names_found > 0) (PUBL(free))((void *)cd->named_groups); } -/* Set up a starting, non-extracting bracket, then compile the expression. On -error, errorcode will be set non-zero, so we don't need to look at the result -of the function here. */ - +/* Set up a starting, non-extracting bracket, then compile the expression. On +error, errorcode will be set non-zero, so we don't need to look at the result +of the function here. */ + ptr = (const pcre_uchar *)pattern + skipatstart; code = (pcre_uchar *)codestart; -*code = OP_BRA; +*code = OP_BRA; (void)compile_regex(re->options, &code, &ptr, &errorcode, FALSE, FALSE, 0, 0, &firstchar, &firstcharflags, &reqchar, &reqcharflags, NULL, cd, NULL); -re->top_bracket = cd->bracount; -re->top_backref = cd->top_backref; +re->top_bracket = cd->bracount; +re->top_backref = cd->top_backref; re->max_lookbehind = cd->max_lookbehind; re->flags = cd->external_flags | PCRE_MODE; - + if (cd->had_accept) { reqchar = 0; /* Must disable after (*ACCEPT) */ reqcharflags = REQ_NONE; } - -/* If not reached end of pattern on success, there's an excess bracket. */ - + +/* If not reached end of pattern on success, there's an excess bracket. */ + if (errorcode == 0 && *ptr != CHAR_NULL) errorcode = ERR22; - -/* Fill in the terminating state and check for disastrous overflow, but -if debugging, leave the test till after things are printed out. */ - -*code++ = OP_END; - + +/* Fill in the terminating state and check for disastrous overflow, but +if debugging, leave the test till after things are printed out. */ + +*code++ = OP_END; + #ifndef PCRE_DEBUG -if (code - codestart > length) errorcode = ERR23; -#endif - +if (code - codestart > length) errorcode = ERR23; +#endif + #ifdef SUPPORT_VALGRIND /* If the estimated length exceeds the really used length, mark the extra allocated memory as unaddressable, so that any out-of-bound reads can be detected. */ VALGRIND_MAKE_MEM_NOACCESS(code, (length - (code - codestart)) * sizeof(pcre_uchar)); #endif - + /* Fill in any forward references that are required. There may be repeated references; optimize for them, as searching a large regex takes time. */ if (cd->hwm > cd->start_workspace) - { + { int prev_recno = -1; const pcre_uchar *groupptr = NULL; while (errorcode == 0 && cd->hwm > cd->start_workspace) @@ -9557,8 +9557,8 @@ if (cd->hwm > cd->start_workspace) if (groupptr == NULL) errorcode = ERR53; else PUT(((pcre_uchar *)codestart), offset, (int)(groupptr - codestart)); } - } - + } + /* If the workspace had to be expanded, free the new memory. Set the pointer to NULL to indicate that forward references have been filled in. */ @@ -9566,11 +9566,11 @@ if (cd->workspace_size > COMPILE_WORK_SIZE) (PUBL(free))((void *)cd->start_workspace); cd->start_workspace = NULL; -/* Give an error if there's back reference to a non-existent capturing -subpattern. */ - -if (errorcode == 0 && re->top_backref > re->top_bracket) errorcode = ERR15; - +/* Give an error if there's back reference to a non-existent capturing +subpattern. */ + +if (errorcode == 0 && re->top_backref > re->top_bracket) errorcode = ERR15; + /* Unless disabled, check whether any single character iterators can be auto-possessified. The function overwrites the appropriate opcode values, so the type of the pointer must be cast. NOTE: the intermediate variable "temp" is @@ -9628,39 +9628,39 @@ if (errorcode == 0 && cd->check_lookbehind) } } -/* Failed to compile, or error while post-processing */ - -if (errorcode != 0) - { +/* Failed to compile, or error while post-processing */ + +if (errorcode != 0) + { (PUBL(free))(re); - PCRE_EARLY_ERROR_RETURN: + PCRE_EARLY_ERROR_RETURN: *erroroffset = (int)(ptr - (const pcre_uchar *)pattern); - PCRE_EARLY_ERROR_RETURN2: - *errorptr = find_error_text(errorcode); - if (errorcodeptr != NULL) *errorcodeptr = errorcode; - return NULL; - } - -/* If the anchored option was not passed, set the flag if we can determine that + PCRE_EARLY_ERROR_RETURN2: + *errorptr = find_error_text(errorcode); + if (errorcodeptr != NULL) *errorcodeptr = errorcode; + return NULL; + } + +/* If the anchored option was not passed, set the flag if we can determine that the pattern is anchored by virtue of ^ characters or \A or anything else, such as starting with non-atomic .* when DOTALL is set and there are no occurrences of *PRUNE or *SKIP. - -Otherwise, if we know what the first byte has to be, save it, because that -speeds up unanchored matches no end. If not, see if we can set the -PCRE_STARTLINE flag. This is helpful for multiline matches when all branches + +Otherwise, if we know what the first byte has to be, save it, because that +speeds up unanchored matches no end. If not, see if we can set the +PCRE_STARTLINE flag. This is helpful for multiline matches when all branches start with ^. and also when all branches start with non-atomic .* for non-DOTALL matches when *PRUNE and SKIP are not present. */ - -if ((re->options & PCRE_ANCHORED) == 0) - { + +if ((re->options & PCRE_ANCHORED) == 0) + { if (is_anchored(codestart, 0, cd, 0)) re->options |= PCRE_ANCHORED; - else - { + else + { if (firstcharflags < 0) firstchar = find_firstassertedchar(codestart, &firstcharflags, FALSE); if (firstcharflags >= 0) /* Remove caseless flag for non-caseable chars */ - { + { #if defined COMPILE_PCRE8 re->first_char = firstchar & 0xff; #elif defined COMPILE_PCRE16 @@ -9689,20 +9689,20 @@ if ((re->options & PCRE_ANCHORED) == 0) re->flags |= PCRE_FCH_CASELESS; } - re->flags |= PCRE_FIRSTSET; - } + re->flags |= PCRE_FIRSTSET; + } else if (is_startline(codestart, 0, cd, 0, FALSE)) re->flags |= PCRE_STARTLINE; - } - } - -/* For an anchored pattern, we use the "required byte" only if it follows a -variable length item in the regex. Remove the caseless flag for non-caseable -bytes. */ - + } + } + +/* For an anchored pattern, we use the "required byte" only if it follows a +variable length item in the regex. Remove the caseless flag for non-caseable +bytes. */ + if (reqcharflags >= 0 && ((re->options & PCRE_ANCHORED) == 0 || (reqcharflags & REQ_VARY) != 0)) - { + { #if defined COMPILE_PCRE8 re->req_char = reqchar & 0xff; #elif defined COMPILE_PCRE16 @@ -9730,36 +9730,36 @@ if (reqcharflags >= 0 && re->flags |= PCRE_RCH_CASELESS; } - re->flags |= PCRE_REQCHSET; - } - -/* Print out the compiled data if debugging is enabled. This is never the -case when building a production library. */ - + re->flags |= PCRE_REQCHSET; + } + +/* Print out the compiled data if debugging is enabled. This is never the +case when building a production library. */ + #ifdef PCRE_DEBUG -printf("Length = %d top_bracket = %d top_backref = %d\n", - length, re->top_bracket, re->top_backref); - -printf("Options=%08x\n", re->options); - -if ((re->flags & PCRE_FIRSTSET) != 0) - { +printf("Length = %d top_bracket = %d top_backref = %d\n", + length, re->top_bracket, re->top_backref); + +printf("Options=%08x\n", re->options); + +if ((re->flags & PCRE_FIRSTSET) != 0) + { pcre_uchar ch = re->first_char; const char *caseless = ((re->flags & PCRE_FCH_CASELESS) == 0)? "" : " (caseless)"; if (PRINTABLE(ch)) printf("First char = %c%s\n", ch, caseless); - else printf("First char = \\x%02x%s\n", ch, caseless); - } - -if ((re->flags & PCRE_REQCHSET) != 0) - { + else printf("First char = \\x%02x%s\n", ch, caseless); + } + +if ((re->flags & PCRE_REQCHSET) != 0) + { pcre_uchar ch = re->req_char; const char *caseless = ((re->flags & PCRE_RCH_CASELESS) == 0)? "" : " (caseless)"; if (PRINTABLE(ch)) printf("Req char = %c%s\n", ch, caseless); - else printf("Req char = \\x%02x%s\n", ch, caseless); - } - + else printf("Req char = \\x%02x%s\n", ch, caseless); + } + #if defined COMPILE_PCRE8 pcre_printint((pcre *)re, stdout, TRUE); #elif defined COMPILE_PCRE16 @@ -9767,20 +9767,20 @@ pcre16_printint((pcre *)re, stdout, TRUE); #elif defined COMPILE_PCRE32 pcre32_printint((pcre *)re, stdout, TRUE); #endif - -/* This check is done here in the debugging case so that the code that -was compiled can be seen. */ - -if (code - codestart > length) - { + +/* This check is done here in the debugging case so that the code that +was compiled can be seen. */ + +if (code - codestart > length) + { (PUBL(free))(re); - *errorptr = find_error_text(ERR23); + *errorptr = find_error_text(ERR23); *erroroffset = ptr - (pcre_uchar *)pattern; - if (errorcodeptr != NULL) *errorcodeptr = ERR23; - return NULL; - } + if (errorcodeptr != NULL) *errorcodeptr = ERR23; + return NULL; + } #endif /* PCRE_DEBUG */ - + /* Check for a pattern than can match an empty string, so that this information can be provided to applications. */ @@ -9796,12 +9796,12 @@ do while (*codestart == OP_ALT); #if defined COMPILE_PCRE8 -return (pcre *)re; +return (pcre *)re; #elif defined COMPILE_PCRE16 return (pcre16 *)re; #elif defined COMPILE_PCRE32 return (pcre32 *)re; #endif -} - -/* End of pcre_compile.c */ +} + +/* End of pcre_compile.c */ diff --git a/contrib/libs/pcre/pcre_config.c b/contrib/libs/pcre/pcre_config.c index 3c5364e2f8..6c303244fc 100644 --- a/contrib/libs/pcre/pcre_config.c +++ b/contrib/libs/pcre/pcre_config.c @@ -1,73 +1,73 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2012 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains the external function pcre_config(). */ - - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains the external function pcre_config(). */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - +#endif + /* Keep the original link size. */ static int real_link_size = LINK_SIZE; -#include "pcre_internal.h" - - -/************************************************* -* Return info about what features are configured * -*************************************************/ - -/* This function has an extensible interface so that additional items can be -added compatibly. - -Arguments: - what what information is required - where where to put the information - -Returns: 0 if data returned, negative on error -*/ - +#include "pcre_internal.h" + + +/************************************************* +* Return info about what features are configured * +*************************************************/ + +/* This function has an extensible interface so that additional items can be +added compatibly. + +Arguments: + what what information is required + where where to put the information + +Returns: 0 if data returned, negative on error +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION -pcre_config(int what, void *where) +pcre_config(int what, void *where) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre16_config(int what, void *where) @@ -75,22 +75,22 @@ pcre16_config(int what, void *where) PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre32_config(int what, void *where) #endif -{ -switch (what) - { - case PCRE_CONFIG_UTF8: +{ +switch (what) + { + case PCRE_CONFIG_UTF8: #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 *((int *)where) = 0; return PCRE_ERROR_BADOPTION; #else #if defined SUPPORT_UTF - *((int *)where) = 1; -#else - *((int *)where) = 0; -#endif - break; + *((int *)where) = 1; +#else + *((int *)where) = 0; +#endif + break; #endif - + case PCRE_CONFIG_UTF16: #if defined COMPILE_PCRE8 || defined COMPILE_PCRE32 *((int *)where) = 0; @@ -117,14 +117,14 @@ switch (what) break; #endif - case PCRE_CONFIG_UNICODE_PROPERTIES: -#ifdef SUPPORT_UCP - *((int *)where) = 1; -#else - *((int *)where) = 0; -#endif - break; - + case PCRE_CONFIG_UNICODE_PROPERTIES: +#ifdef SUPPORT_UCP + *((int *)where) = 1; +#else + *((int *)where) = 0; +#endif + break; + case PCRE_CONFIG_JIT: #ifdef SUPPORT_JIT *((int *)where) = 1; @@ -141,50 +141,50 @@ switch (what) #endif break; - case PCRE_CONFIG_NEWLINE: - *((int *)where) = NEWLINE; - break; - - case PCRE_CONFIG_BSR: -#ifdef BSR_ANYCRLF - *((int *)where) = 1; -#else - *((int *)where) = 0; -#endif - break; - - case PCRE_CONFIG_LINK_SIZE: + case PCRE_CONFIG_NEWLINE: + *((int *)where) = NEWLINE; + break; + + case PCRE_CONFIG_BSR: +#ifdef BSR_ANYCRLF + *((int *)where) = 1; +#else + *((int *)where) = 0; +#endif + break; + + case PCRE_CONFIG_LINK_SIZE: *((int *)where) = real_link_size; - break; - - case PCRE_CONFIG_POSIX_MALLOC_THRESHOLD: - *((int *)where) = POSIX_MALLOC_THRESHOLD; - break; - + break; + + case PCRE_CONFIG_POSIX_MALLOC_THRESHOLD: + *((int *)where) = POSIX_MALLOC_THRESHOLD; + break; + case PCRE_CONFIG_PARENS_LIMIT: *((unsigned long int *)where) = PARENS_NEST_LIMIT; break; - case PCRE_CONFIG_MATCH_LIMIT: + case PCRE_CONFIG_MATCH_LIMIT: *((unsigned long int *)where) = MATCH_LIMIT; - break; - - case PCRE_CONFIG_MATCH_LIMIT_RECURSION: + break; + + case PCRE_CONFIG_MATCH_LIMIT_RECURSION: *((unsigned long int *)where) = MATCH_LIMIT_RECURSION; - break; - - case PCRE_CONFIG_STACKRECURSE: -#ifdef NO_RECURSE - *((int *)where) = 0; -#else - *((int *)where) = 1; -#endif - break; - - default: return PCRE_ERROR_BADOPTION; - } - -return 0; -} - -/* End of pcre_config.c */ + break; + + case PCRE_CONFIG_STACKRECURSE: +#ifdef NO_RECURSE + *((int *)where) = 0; +#else + *((int *)where) = 1; +#endif + break; + + default: return PCRE_ERROR_BADOPTION; + } + +return 0; +} + +/* End of pcre_config.c */ diff --git a/contrib/libs/pcre/pcre_config.h b/contrib/libs/pcre/pcre_config.h index 622b2ec59b..ebc9c01fc3 100644 --- a/contrib/libs/pcre/pcre_config.h +++ b/contrib/libs/pcre/pcre_config.h @@ -53,8 +53,8 @@ sure both macros are undefined; an emulation function will then be used. */ /* #undef EBCDIC_NL25 */ /* Define to 1 if you have the `bcopy' function. */ -#define HAVE_BCOPY 1 - +#define HAVE_BCOPY 1 + /* Define to 1 if you have the <bits/type_traits.h> header file. */ /* #undef HAVE_BITS_TYPE_TRAITS_H */ @@ -76,15 +76,15 @@ sure both macros are undefined; an emulation function will then be used. */ /* Define to 1 if you have the <inttypes.h> header file. */ #define HAVE_INTTYPES_H 1 -/* Define to 1 if you have the <limits.h> header file. */ -#define HAVE_LIMITS_H 1 - -/* Define to 1 if the system has the type `long long'. */ +/* Define to 1 if you have the <limits.h> header file. */ +#define HAVE_LIMITS_H 1 + +/* Define to 1 if the system has the type `long long'. */ #define HAVE_LONG_LONG 1 - -/* Define to 1 if you have the `memmove' function. */ -#define HAVE_MEMMOVE 1 - + +/* Define to 1 if you have the `memmove' function. */ +#define HAVE_MEMMOVE 1 + /* Define to 1 if you have the <memory.h> header file. */ #define HAVE_MEMORY_H 1 @@ -100,27 +100,27 @@ sure both macros are undefined; an emulation function will then be used. */ /* Define to 1 if you have the <readline/readline.h> header file. */ /* #undef HAVE_READLINE_READLINE_H */ -/* Define to 1 if you have the <stdint.h> header file. */ -#define HAVE_STDINT_H 1 - -/* Define to 1 if you have the <stdlib.h> header file. */ -#define HAVE_STDLIB_H 1 - -/* Define to 1 if you have the `strerror' function. */ -#define HAVE_STRERROR 1 - +/* Define to 1 if you have the <stdint.h> header file. */ +#define HAVE_STDINT_H 1 + +/* Define to 1 if you have the <stdlib.h> header file. */ +#define HAVE_STDLIB_H 1 + +/* Define to 1 if you have the `strerror' function. */ +#define HAVE_STRERROR 1 + /* Define to 1 if you have the <string> header file. */ #define HAVE_STRING 1 -/* Define to 1 if you have the <strings.h> header file. */ -#define HAVE_STRINGS_H 1 - +/* Define to 1 if you have the <strings.h> header file. */ +#define HAVE_STRINGS_H 1 + /* Define to 1 if you have the <string.h> header file. */ #define HAVE_STRING_H 1 - + /* Define to 1 if you have `strtoimax'. */ /* #undef HAVE_STRTOIMAX */ - + /* Define to 1 if you have `strtoll'. */ /* #undef HAVE_STRTOLL */ @@ -139,9 +139,9 @@ sure both macros are undefined; an emulation function will then be used. */ /* Define to 1 if you have the <unistd.h> header file. */ #define HAVE_UNISTD_H 1 -/* Define to 1 if the system has the type `unsigned long long'. */ +/* Define to 1 if the system has the type `unsigned long long'. */ #define HAVE_UNSIGNED_LONG_LONG 1 - + /* Define to 1 if the compiler supports simple visibility declarations. */ #define HAVE_VISIBILITY 1 @@ -154,44 +154,44 @@ sure both macros are undefined; an emulation function will then be used. */ /* Define to 1 if you have `_strtoi64'. */ /* #undef HAVE__STRTOI64 */ -/* The value of LINK_SIZE determines the number of bytes used to store links - as offsets within the compiled regex. The default is 2, which allows for - compiled patterns up to 64K long. This covers the vast majority of cases. - However, PCRE can also be compiled to use 3 or 4 bytes instead. This allows +/* The value of LINK_SIZE determines the number of bytes used to store links + as offsets within the compiled regex. The default is 2, which allows for + compiled patterns up to 64K long. This covers the vast majority of cases. + However, PCRE can also be compiled to use 3 or 4 bytes instead. This allows for longer patterns in extreme cases. */ -#define LINK_SIZE 2 - +#define LINK_SIZE 2 + /* Define to the sub-directory where libtool stores uninstalled libraries. */ #define LT_OBJDIR ".libs/" -/* The value of MATCH_LIMIT determines the default number of times the - internal match() function can be called during a single execution of - pcre_exec(). There is a runtime interface for setting a different limit. - The limit exists in order to catch runaway regular expressions that take - for ever to determine that they do not match. The default is set very large +/* The value of MATCH_LIMIT determines the default number of times the + internal match() function can be called during a single execution of + pcre_exec(). There is a runtime interface for setting a different limit. + The limit exists in order to catch runaway regular expressions that take + for ever to determine that they do not match. The default is set very large so that it does not accidentally catch legitimate cases. */ -#define MATCH_LIMIT 10000000 - -/* The above limit applies to all calls of match(), whether or not they - increase the recursion depth. In some environments it is desirable to limit - the depth of recursive calls of match() more strictly, in order to restrict - the maximum amount of stack (or heap, if NO_RECURSE is defined) that is - used. The value of MATCH_LIMIT_RECURSION applies only to recursive calls of - match(). To have any useful effect, it must be less than the value of - MATCH_LIMIT. The default is to use the same value as MATCH_LIMIT. There is +#define MATCH_LIMIT 10000000 + +/* The above limit applies to all calls of match(), whether or not they + increase the recursion depth. In some environments it is desirable to limit + the depth of recursive calls of match() more strictly, in order to restrict + the maximum amount of stack (or heap, if NO_RECURSE is defined) that is + used. The value of MATCH_LIMIT_RECURSION applies only to recursive calls of + match(). To have any useful effect, it must be less than the value of + MATCH_LIMIT. The default is to use the same value as MATCH_LIMIT. There is a runtime method for setting a different limit. */ -#define MATCH_LIMIT_RECURSION MATCH_LIMIT - -/* This limit is parameterized just in case anybody ever wants to change it. - Care must be taken if it is increased, because it guards against integer - overflow caused by enormously large patterns. */ -#define MAX_NAME_COUNT 10000 - -/* This limit is parameterized just in case anybody ever wants to change it. - Care must be taken if it is increased, because it guards against integer - overflow caused by enormously large patterns. */ -#define MAX_NAME_SIZE 32 - +#define MATCH_LIMIT_RECURSION MATCH_LIMIT + +/* This limit is parameterized just in case anybody ever wants to change it. + Care must be taken if it is increased, because it guards against integer + overflow caused by enormously large patterns. */ +#define MAX_NAME_COUNT 10000 + +/* This limit is parameterized just in case anybody ever wants to change it. + Care must be taken if it is increased, because it guards against integer + overflow caused by enormously large patterns. */ +#define MAX_NAME_SIZE 32 + /* The value of NEWLINE determines the default newline character sequence. PCRE client programs can override this by selecting other values at run time. In ASCII environments, the value can be 10 (LF), 13 (CR), or 3338 @@ -200,38 +200,38 @@ sure both macros are undefined; an emulation function will then be used. */ 0x25) that are used as the NL line terminator that is equivalent to ASCII LF. In both ASCII and EBCDIC environments the value can also be -1 (ANY), or -2 (ANYCRLF). */ -#define NEWLINE 10 - -/* PCRE uses recursive function calls to handle backtracking while matching. - This can sometimes be a problem on systems that have stacks of limited +#define NEWLINE 10 + +/* PCRE uses recursive function calls to handle backtracking while matching. + This can sometimes be a problem on systems that have stacks of limited size. Define NO_RECURSE to any value to get a version that doesn't use recursion in the match() function; instead it creates its own stack by steam using pcre_recurse_malloc() to obtain memory from the heap. For more detail, see the comments and other stuff just above the match() function. */ -/* #undef NO_RECURSE */ - -/* Name of package */ -#define PACKAGE "pcre" - -/* Define to the address where bug reports for this package should be sent. */ -#define PACKAGE_BUGREPORT "" - -/* Define to the full name of this package. */ -#define PACKAGE_NAME "PCRE" - -/* Define to the full name and version of this package. */ +/* #undef NO_RECURSE */ + +/* Name of package */ +#define PACKAGE "pcre" + +/* Define to the address where bug reports for this package should be sent. */ +#define PACKAGE_BUGREPORT "" + +/* Define to the full name of this package. */ +#define PACKAGE_NAME "PCRE" + +/* Define to the full name and version of this package. */ #define PACKAGE_STRING "PCRE 8.44" - -/* Define to the one symbol short name of this package. */ -#define PACKAGE_TARNAME "pcre" - + +/* Define to the one symbol short name of this package. */ +#define PACKAGE_TARNAME "pcre" + /* Define to the home page for this package. */ #define PACKAGE_URL "" -/* Define to the version of this package. */ +/* Define to the version of this package. */ #define PACKAGE_VERSION "8.44" - + /* The value of PARENS_NEST_LIMIT specifies the maximum depth of nested parentheses (of any kind) in a pattern. This limits the amount of system stack that is used while compiling a pattern. */ @@ -277,27 +277,27 @@ sure both macros are undefined; an emulation function will then be used. */ /* Define to any value if linking statically (TODO: make nice with Libtool) */ /* #undef PCRE_STATIC */ -/* When calling PCRE via the POSIX interface, additional working storage is - required for holding the pointers to capturing substrings because PCRE - requires three integers per substring, whereas the POSIX interface provides - only two. If the number of expected substrings is small, the wrapper - function uses space on the stack, because this is faster than using - malloc() for each call. The threshold above which the stack is no longer +/* When calling PCRE via the POSIX interface, additional working storage is + required for holding the pointers to capturing substrings because PCRE + requires three integers per substring, whereas the POSIX interface provides + only two. If the number of expected substrings is small, the wrapper + function uses space on the stack, because this is faster than using + malloc() for each call. The threshold above which the stack is no longer used is defined by POSIX_MALLOC_THRESHOLD. */ -#define POSIX_MALLOC_THRESHOLD 10 - +#define POSIX_MALLOC_THRESHOLD 10 + /* Define to necessary symbol if this constant uses a non-standard name on your system. */ /* #undef PTHREAD_CREATE_JOINABLE */ -/* Define to 1 if you have the ANSI C header files. */ -#define STDC_HEADERS 1 - +/* Define to 1 if you have the ANSI C header files. */ +#define STDC_HEADERS 1 + #ifdef ARCADIA_PCRE_ENABLE_JIT /* Define to any value to enable support for Just-In-Time compiling. */ #define SUPPORT_JIT /**/ #endif - + /* Define to any value to allow pcregrep to be linked with libbz2, so that it is able to handle .bz2 files. */ /* #undef SUPPORT_LIBBZ2 */ @@ -338,7 +338,7 @@ sure both macros are undefined; an emulation function will then be used. */ /* Define to any value for valgrind support to find invalid memory reads. */ /* #undef SUPPORT_VALGRIND */ -/* Version number of package */ +/* Version number of package */ #define VERSION "8.44" /* Define to empty if `const' does not conform to ANSI C. */ diff --git a/contrib/libs/pcre/pcre_dfa_exec.c b/contrib/libs/pcre/pcre_dfa_exec.c index 81eec05356..649d1b19d9 100644 --- a/contrib/libs/pcre/pcre_dfa_exec.c +++ b/contrib/libs/pcre/pcre_dfa_exec.c @@ -1,49 +1,49 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax and semantics are as close as possible to those of the Perl 5 language (but see below for why this module is different). - - Written by Philip Hazel + + Written by Philip Hazel Copyright (c) 1997-2017 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - -/* This module contains the external function pcre_dfa_exec(), which is an -alternative matching function that uses a sort of DFA algorithm (not a true + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + +/* This module contains the external function pcre_dfa_exec(), which is an +alternative matching function that uses a sort of DFA algorithm (not a true FSM). This is NOT Perl-compatible, but it has advantages in certain -applications. */ - - +applications. */ + + /* NOTE ABOUT PERFORMANCE: A user of this function sent some code that improved the performance of his patterns greatly. I could not use it as it stood, as it was not thread safe, and made assumptions about pattern sizes. Also, it caused @@ -72,61 +72,61 @@ in others, so I abandoned this code. */ -#ifdef HAVE_CONFIG_H +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#define NLBLOCK md /* Block containing newline information */ -#define PSSTART start_subject /* Field containing processed string start */ -#define PSEND end_subject /* Field containing processed string end */ - -#include "pcre_internal.h" - - -/* For use to indent debugging output */ - -#define SP " " - - -/************************************************* -* Code parameters and static tables * -*************************************************/ - -/* These are offsets that are used to turn the OP_TYPESTAR and friends opcodes -into others, under special conditions. A gap of 20 between the blocks should be -enough. The resulting opcodes don't have to be less than 256 because they are -never stored, so we push them well clear of the normal opcodes. */ - -#define OP_PROP_EXTRA 300 -#define OP_EXTUNI_EXTRA 320 -#define OP_ANYNL_EXTRA 340 -#define OP_HSPACE_EXTRA 360 -#define OP_VSPACE_EXTRA 380 - - -/* This table identifies those opcodes that are followed immediately by a +#endif + +#define NLBLOCK md /* Block containing newline information */ +#define PSSTART start_subject /* Field containing processed string start */ +#define PSEND end_subject /* Field containing processed string end */ + +#include "pcre_internal.h" + + +/* For use to indent debugging output */ + +#define SP " " + + +/************************************************* +* Code parameters and static tables * +*************************************************/ + +/* These are offsets that are used to turn the OP_TYPESTAR and friends opcodes +into others, under special conditions. A gap of 20 between the blocks should be +enough. The resulting opcodes don't have to be less than 256 because they are +never stored, so we push them well clear of the normal opcodes. */ + +#define OP_PROP_EXTRA 300 +#define OP_EXTUNI_EXTRA 320 +#define OP_ANYNL_EXTRA 340 +#define OP_HSPACE_EXTRA 360 +#define OP_VSPACE_EXTRA 380 + + +/* This table identifies those opcodes that are followed immediately by a character that is to be tested in some way. This makes it possible to -centralize the loading of these characters. In the case of Type * etc, the -"character" is the opcode for \D, \d, \S, \s, \W, or \w, which will always be a +centralize the loading of these characters. In the case of Type * etc, the +"character" is the opcode for \D, \d, \S, \s, \W, or \w, which will always be a small value. Non-zero values in the table are the offsets from the opcode where the character is to be found. ***NOTE*** If the start of this table is modified, the three tables that follow must also be modified. */ - + static const pcre_uint8 coptable[] = { - 0, /* End */ - 0, 0, 0, 0, 0, /* \A, \G, \K, \B, \b */ - 0, 0, 0, 0, 0, 0, /* \D, \d, \S, \s, \W, \w */ + 0, /* End */ + 0, 0, 0, 0, 0, /* \A, \G, \K, \B, \b */ + 0, 0, 0, 0, 0, 0, /* \D, \d, \S, \s, \W, \w */ 0, 0, 0, /* Any, AllAny, Anybyte */ 0, 0, /* \P, \p */ - 0, 0, 0, 0, 0, /* \R, \H, \h, \V, \v */ + 0, 0, 0, 0, 0, /* \R, \H, \h, \V, \v */ 0, /* \X */ 0, 0, 0, 0, 0, 0, /* \Z, \z, $, $M, ^, ^M */ - 1, /* Char */ + 1, /* Char */ 1, /* Chari */ - 1, /* not */ + 1, /* not */ 1, /* noti */ - /* Positive single-char repeats */ - 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ + /* Positive single-char repeats */ + 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ 1+IMM2_SIZE, 1+IMM2_SIZE, /* upto, minupto */ 1+IMM2_SIZE, /* exact */ 1, 1, 1, 1+IMM2_SIZE, /* *+, ++, ?+, upto+ */ @@ -134,8 +134,8 @@ static const pcre_uint8 coptable[] = { 1+IMM2_SIZE, 1+IMM2_SIZE, /* upto I, minupto I */ 1+IMM2_SIZE, /* exact I */ 1, 1, 1, 1+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ - /* Negative single-char repeats - only for chars < 256 */ - 1, 1, 1, 1, 1, 1, /* NOT *, *?, +, +?, ?, ?? */ + /* Negative single-char repeats - only for chars < 256 */ + 1, 1, 1, 1, 1, 1, /* NOT *, *?, +, +?, ?, ?? */ 1+IMM2_SIZE, 1+IMM2_SIZE, /* NOT upto, minupto */ 1+IMM2_SIZE, /* NOT exact */ 1, 1, 1, 1+IMM2_SIZE, /* NOT *+, ++, ?+, upto+ */ @@ -143,34 +143,34 @@ static const pcre_uint8 coptable[] = { 1+IMM2_SIZE, 1+IMM2_SIZE, /* NOT upto I, minupto I */ 1+IMM2_SIZE, /* NOT exact I */ 1, 1, 1, 1+IMM2_SIZE, /* NOT *+I, ++I, ?+I, upto+I */ - /* Positive type repeats */ - 1, 1, 1, 1, 1, 1, /* Type *, *?, +, +?, ?, ?? */ + /* Positive type repeats */ + 1, 1, 1, 1, 1, 1, /* Type *, *?, +, +?, ?, ?? */ 1+IMM2_SIZE, 1+IMM2_SIZE, /* Type upto, minupto */ 1+IMM2_SIZE, /* Type exact */ 1, 1, 1, 1+IMM2_SIZE, /* Type *+, ++, ?+, upto+ */ - /* Character class & ref repeats */ - 0, 0, 0, 0, 0, 0, /* *, *?, +, +?, ?, ?? */ - 0, 0, /* CRRANGE, CRMINRANGE */ + /* Character class & ref repeats */ + 0, 0, 0, 0, 0, 0, /* *, *?, +, +?, ?, ?? */ + 0, 0, /* CRRANGE, CRMINRANGE */ 0, 0, 0, 0, /* Possessive *+, ++, ?+, CRPOSRANGE */ - 0, /* CLASS */ - 0, /* NCLASS */ - 0, /* XCLASS - variable length */ - 0, /* REF */ + 0, /* CLASS */ + 0, /* NCLASS */ + 0, /* XCLASS - variable length */ + 0, /* REF */ 0, /* REFI */ 0, /* DNREF */ 0, /* DNREFI */ - 0, /* RECURSE */ - 0, /* CALLOUT */ - 0, /* Alt */ - 0, /* Ket */ - 0, /* KetRmax */ - 0, /* KetRmin */ + 0, /* RECURSE */ + 0, /* CALLOUT */ + 0, /* Alt */ + 0, /* Ket */ + 0, /* KetRmax */ + 0, /* KetRmin */ 0, /* KetRpos */ 0, /* Reverse */ - 0, /* Assert */ - 0, /* Assert not */ - 0, /* Assert behind */ - 0, /* Assert behind not */ + 0, /* Assert */ + 0, /* Assert not */ + 0, /* Assert behind */ + 0, /* Assert behind not */ 0, 0, /* ONCE, ONCE_NC */ 0, 0, 0, 0, 0, /* BRA, BRAPOS, CBRA, CBRAPOS, COND */ 0, 0, 0, 0, 0, /* SBRA, SBRAPOS, SCBRA, SCBRAPOS, SCOND */ @@ -238,7 +238,7 @@ static const pcre_uint8 poptable[] = { 0, /* KetRmax */ 0, /* KetRmin */ 0, /* KetRpos */ - 0, /* Reverse */ + 0, /* Reverse */ 0, /* Assert */ 0, /* Assert not */ 0, /* Assert behind */ @@ -248,516 +248,516 @@ static const pcre_uint8 poptable[] = { 0, 0, 0, 0, 0, /* SBRA, SBRAPOS, SCBRA, SCBRAPOS, SCOND */ 0, 0, /* CREF, DNCREF */ 0, 0, /* RREF, DNRREF */ - 0, /* DEF */ + 0, /* DEF */ 0, 0, 0, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ 0, 0, 0, /* MARK, PRUNE, PRUNE_ARG */ 0, 0, 0, 0, /* SKIP, SKIP_ARG, THEN, THEN_ARG */ 0, 0, 0, 0, /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT */ 0, 0 /* CLOSE, SKIPZERO */ -}; - -/* These 2 tables allow for compact code for testing for \D, \d, \S, \s, \W, -and \w */ - +}; + +/* These 2 tables allow for compact code for testing for \D, \d, \S, \s, \W, +and \w */ + static const pcre_uint8 toptable1[] = { - 0, 0, 0, 0, 0, 0, - ctype_digit, ctype_digit, - ctype_space, ctype_space, - ctype_word, ctype_word, + 0, 0, 0, 0, 0, 0, + ctype_digit, ctype_digit, + ctype_space, ctype_space, + ctype_word, ctype_word, 0, 0 /* OP_ANY, OP_ALLANY */ -}; - +}; + static const pcre_uint8 toptable2[] = { - 0, 0, 0, 0, 0, 0, - ctype_digit, 0, - ctype_space, 0, - ctype_word, 0, + 0, 0, 0, 0, 0, 0, + ctype_digit, 0, + ctype_space, 0, + ctype_word, 0, 1, 1 /* OP_ANY, OP_ALLANY */ -}; - - -/* Structure for holding data about a particular state, which is in effect the -current data for an active path through the match tree. It must consist -entirely of ints because the working vector we are passed, and which we put -these structures in, is a vector of ints. */ - -typedef struct stateblock { - int offset; /* Offset to opcode */ - int count; /* Count for repeats */ - int data; /* Some use extra data */ -} stateblock; - +}; + + +/* Structure for holding data about a particular state, which is in effect the +current data for an active path through the match tree. It must consist +entirely of ints because the working vector we are passed, and which we put +these structures in, is a vector of ints. */ + +typedef struct stateblock { + int offset; /* Offset to opcode */ + int count; /* Count for repeats */ + int data; /* Some use extra data */ +} stateblock; + #define INTS_PER_STATEBLOCK (int)(sizeof(stateblock)/sizeof(int)) - - + + #ifdef PCRE_DEBUG -/************************************************* -* Print character string * -*************************************************/ - -/* Character string printing function for debugging. - -Arguments: - p points to string - length number of bytes - f where to print - -Returns: nothing -*/ - -static void +/************************************************* +* Print character string * +*************************************************/ + +/* Character string printing function for debugging. + +Arguments: + p points to string + length number of bytes + f where to print + +Returns: nothing +*/ + +static void pchars(const pcre_uchar *p, int length, FILE *f) -{ +{ pcre_uint32 c; -while (length-- > 0) - { - if (isprint(c = *(p++))) - fprintf(f, "%c", c); - else +while (length-- > 0) + { + if (isprint(c = *(p++))) + fprintf(f, "%c", c); + else fprintf(f, "\\x{%02x}", c); - } -} -#endif - - - -/************************************************* -* Execute a Regular Expression - DFA engine * -*************************************************/ - -/* This internal function applies a compiled pattern to a subject string, -starting at a given point, using a DFA engine. This function is called from the -external one, possibly multiple times if the pattern is not anchored. The -function calls itself recursively for some kinds of subpattern. - -Arguments: - md the match_data block with fixed information - this_start_code the opening bracket of this subexpression's code - current_subject where we currently are in the subject string - start_offset start offset in the subject string - offsets vector to contain the matching string offsets - offsetcount size of same - workspace vector of workspace - wscount size of same - rlevel function call recursion level - + } +} +#endif + + + +/************************************************* +* Execute a Regular Expression - DFA engine * +*************************************************/ + +/* This internal function applies a compiled pattern to a subject string, +starting at a given point, using a DFA engine. This function is called from the +external one, possibly multiple times if the pattern is not anchored. The +function calls itself recursively for some kinds of subpattern. + +Arguments: + md the match_data block with fixed information + this_start_code the opening bracket of this subexpression's code + current_subject where we currently are in the subject string + start_offset start offset in the subject string + offsets vector to contain the matching string offsets + offsetcount size of same + workspace vector of workspace + wscount size of same + rlevel function call recursion level + Returns: > 0 => number of match offset pairs placed in offsets = 0 => offsets overflowed; longest matches are present - -1 => failed to match - < -1 => some kind of unexpected problem - -The following macros are used for adding states to the two state vectors (one -for the current character, one for the following character). */ - -#define ADD_ACTIVE(x,y) \ - if (active_count++ < wscount) \ - { \ - next_active_state->offset = (x); \ - next_active_state->count = (y); \ - next_active_state++; \ - DPRINTF(("%.*sADD_ACTIVE(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \ - } \ - else return PCRE_ERROR_DFA_WSSIZE - -#define ADD_ACTIVE_DATA(x,y,z) \ - if (active_count++ < wscount) \ - { \ - next_active_state->offset = (x); \ - next_active_state->count = (y); \ - next_active_state->data = (z); \ - next_active_state++; \ - DPRINTF(("%.*sADD_ACTIVE_DATA(%d,%d,%d)\n", rlevel*2-2, SP, (x), (y), (z))); \ - } \ - else return PCRE_ERROR_DFA_WSSIZE - -#define ADD_NEW(x,y) \ - if (new_count++ < wscount) \ - { \ - next_new_state->offset = (x); \ - next_new_state->count = (y); \ - next_new_state++; \ - DPRINTF(("%.*sADD_NEW(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \ - } \ - else return PCRE_ERROR_DFA_WSSIZE - -#define ADD_NEW_DATA(x,y,z) \ - if (new_count++ < wscount) \ - { \ - next_new_state->offset = (x); \ - next_new_state->count = (y); \ - next_new_state->data = (z); \ - next_new_state++; \ + -1 => failed to match + < -1 => some kind of unexpected problem + +The following macros are used for adding states to the two state vectors (one +for the current character, one for the following character). */ + +#define ADD_ACTIVE(x,y) \ + if (active_count++ < wscount) \ + { \ + next_active_state->offset = (x); \ + next_active_state->count = (y); \ + next_active_state++; \ + DPRINTF(("%.*sADD_ACTIVE(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \ + } \ + else return PCRE_ERROR_DFA_WSSIZE + +#define ADD_ACTIVE_DATA(x,y,z) \ + if (active_count++ < wscount) \ + { \ + next_active_state->offset = (x); \ + next_active_state->count = (y); \ + next_active_state->data = (z); \ + next_active_state++; \ + DPRINTF(("%.*sADD_ACTIVE_DATA(%d,%d,%d)\n", rlevel*2-2, SP, (x), (y), (z))); \ + } \ + else return PCRE_ERROR_DFA_WSSIZE + +#define ADD_NEW(x,y) \ + if (new_count++ < wscount) \ + { \ + next_new_state->offset = (x); \ + next_new_state->count = (y); \ + next_new_state++; \ + DPRINTF(("%.*sADD_NEW(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \ + } \ + else return PCRE_ERROR_DFA_WSSIZE + +#define ADD_NEW_DATA(x,y,z) \ + if (new_count++ < wscount) \ + { \ + next_new_state->offset = (x); \ + next_new_state->count = (y); \ + next_new_state->data = (z); \ + next_new_state++; \ DPRINTF(("%.*sADD_NEW_DATA(%d,%d,%d) line %d\n", rlevel*2-2, SP, \ (x), (y), (z), __LINE__)); \ - } \ - else return PCRE_ERROR_DFA_WSSIZE - -/* And now, here is the code */ - -static int -internal_dfa_exec( - dfa_match_data *md, + } \ + else return PCRE_ERROR_DFA_WSSIZE + +/* And now, here is the code */ + +static int +internal_dfa_exec( + dfa_match_data *md, const pcre_uchar *this_start_code, const pcre_uchar *current_subject, - int start_offset, - int *offsets, - int offsetcount, - int *workspace, - int wscount, + int start_offset, + int *offsets, + int offsetcount, + int *workspace, + int wscount, int rlevel) -{ -stateblock *active_states, *new_states, *temp_states; -stateblock *next_active_state, *next_new_state; - +{ +stateblock *active_states, *new_states, *temp_states; +stateblock *next_active_state, *next_new_state; + const pcre_uint8 *ctypes, *lcc, *fcc; const pcre_uchar *ptr; const pcre_uchar *end_code, *first_op; - + dfa_recursion_info new_recursive; -int active_count, new_count, match_count; - -/* Some fields in the md block are frequently referenced, so we load them into -independent variables in the hope that this will perform better. */ - +int active_count, new_count, match_count; + +/* Some fields in the md block are frequently referenced, so we load them into +independent variables in the hope that this will perform better. */ + const pcre_uchar *start_subject = md->start_subject; const pcre_uchar *end_subject = md->end_subject; const pcre_uchar *start_code = md->start_code; - + #ifdef SUPPORT_UTF BOOL utf = (md->poptions & PCRE_UTF8) != 0; -#else +#else BOOL utf = FALSE; -#endif - +#endif + BOOL reset_could_continue = FALSE; -rlevel++; -offsetcount &= (-2); - -wscount -= 2; -wscount = (wscount - (wscount % (INTS_PER_STATEBLOCK * 2))) / - (2 * INTS_PER_STATEBLOCK); - -DPRINTF(("\n%.*s---------------------\n" +rlevel++; +offsetcount &= (-2); + +wscount -= 2; +wscount = (wscount - (wscount % (INTS_PER_STATEBLOCK * 2))) / + (2 * INTS_PER_STATEBLOCK); + +DPRINTF(("\n%.*s---------------------\n" "%.*sCall to internal_dfa_exec f=%d\n", rlevel*2-2, SP, rlevel*2-2, SP, rlevel)); - -ctypes = md->tables + ctypes_offset; -lcc = md->tables + lcc_offset; -fcc = md->tables + fcc_offset; - -match_count = PCRE_ERROR_NOMATCH; /* A negative number */ - -active_states = (stateblock *)(workspace + 2); -next_new_state = new_states = active_states + wscount; -new_count = 0; - -first_op = this_start_code + 1 + LINK_SIZE + + +ctypes = md->tables + ctypes_offset; +lcc = md->tables + lcc_offset; +fcc = md->tables + fcc_offset; + +match_count = PCRE_ERROR_NOMATCH; /* A negative number */ + +active_states = (stateblock *)(workspace + 2); +next_new_state = new_states = active_states + wscount; +new_count = 0; + +first_op = this_start_code + 1 + LINK_SIZE + ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA || *this_start_code == OP_CBRAPOS || *this_start_code == OP_SCBRAPOS) ? IMM2_SIZE:0); - -/* The first thing in any (sub) pattern is a bracket of some sort. Push all -the alternative states onto the list, and find out where the end is. This -makes is possible to use this function recursively, when we want to stop at a -matching internal ket rather than at the end. - -If the first opcode in the first alternative is OP_REVERSE, we are dealing with -a backward assertion. In that case, we have to find out the maximum amount to -move back, and set up each alternative appropriately. */ - -if (*first_op == OP_REVERSE) - { - int max_back = 0; - int gone_back; - - end_code = this_start_code; - do - { - int back = GET(end_code, 2+LINK_SIZE); - if (back > max_back) max_back = back; - end_code += GET(end_code, 1); - } - while (*end_code == OP_ALT); - - /* If we can't go back the amount required for the longest lookbehind - pattern, go back as far as we can; some alternatives may still be viable. */ - + +/* The first thing in any (sub) pattern is a bracket of some sort. Push all +the alternative states onto the list, and find out where the end is. This +makes is possible to use this function recursively, when we want to stop at a +matching internal ket rather than at the end. + +If the first opcode in the first alternative is OP_REVERSE, we are dealing with +a backward assertion. In that case, we have to find out the maximum amount to +move back, and set up each alternative appropriately. */ + +if (*first_op == OP_REVERSE) + { + int max_back = 0; + int gone_back; + + end_code = this_start_code; + do + { + int back = GET(end_code, 2+LINK_SIZE); + if (back > max_back) max_back = back; + end_code += GET(end_code, 1); + } + while (*end_code == OP_ALT); + + /* If we can't go back the amount required for the longest lookbehind + pattern, go back as far as we can; some alternatives may still be viable. */ + #ifdef SUPPORT_UTF - /* In character mode we have to step back character by character */ - + /* In character mode we have to step back character by character */ + if (utf) - { - for (gone_back = 0; gone_back < max_back; gone_back++) - { - if (current_subject <= start_subject) break; - current_subject--; + { + for (gone_back = 0; gone_back < max_back; gone_back++) + { + if (current_subject <= start_subject) break; + current_subject--; ACROSSCHAR(current_subject > start_subject, *current_subject, current_subject--); - } - } - else -#endif - - /* In byte-mode we can do this quickly. */ - - { - gone_back = (current_subject - max_back < start_subject)? + } + } + else +#endif + + /* In byte-mode we can do this quickly. */ + + { + gone_back = (current_subject - max_back < start_subject)? (int)(current_subject - start_subject) : max_back; - current_subject -= gone_back; - } - + current_subject -= gone_back; + } + /* Save the earliest consulted character */ if (current_subject < md->start_used_ptr) md->start_used_ptr = current_subject; - /* Now we can process the individual branches. */ - - end_code = this_start_code; - do - { - int back = GET(end_code, 2+LINK_SIZE); - if (back <= gone_back) - { + /* Now we can process the individual branches. */ + + end_code = this_start_code; + do + { + int back = GET(end_code, 2+LINK_SIZE); + if (back <= gone_back) + { int bstate = (int)(end_code - start_code + 2 + 2*LINK_SIZE); - ADD_NEW_DATA(-bstate, 0, gone_back - back); - } - end_code += GET(end_code, 1); - } - while (*end_code == OP_ALT); - } - -/* This is the code for a "normal" subpattern (not a backward assertion). The -start of a whole pattern is always one of these. If we are at the top level, -we may be asked to restart matching from the same point that we reached for a -previous partial match. We still have to scan through the top-level branches to -find the end state. */ - -else - { - end_code = this_start_code; - - /* Restarting */ - - if (rlevel == 1 && (md->moptions & PCRE_DFA_RESTART) != 0) - { - do { end_code += GET(end_code, 1); } while (*end_code == OP_ALT); - new_count = workspace[1]; - if (!workspace[0]) - memcpy(new_states, active_states, new_count * sizeof(stateblock)); - } - - /* Not restarting */ - - else - { - int length = 1 + LINK_SIZE + + ADD_NEW_DATA(-bstate, 0, gone_back - back); + } + end_code += GET(end_code, 1); + } + while (*end_code == OP_ALT); + } + +/* This is the code for a "normal" subpattern (not a backward assertion). The +start of a whole pattern is always one of these. If we are at the top level, +we may be asked to restart matching from the same point that we reached for a +previous partial match. We still have to scan through the top-level branches to +find the end state. */ + +else + { + end_code = this_start_code; + + /* Restarting */ + + if (rlevel == 1 && (md->moptions & PCRE_DFA_RESTART) != 0) + { + do { end_code += GET(end_code, 1); } while (*end_code == OP_ALT); + new_count = workspace[1]; + if (!workspace[0]) + memcpy(new_states, active_states, new_count * sizeof(stateblock)); + } + + /* Not restarting */ + + else + { + int length = 1 + LINK_SIZE + ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA || *this_start_code == OP_CBRAPOS || *this_start_code == OP_SCBRAPOS) ? IMM2_SIZE:0); - do - { + do + { ADD_NEW((int)(end_code - start_code + length), 0); - end_code += GET(end_code, 1); - length = 1 + LINK_SIZE; - } - while (*end_code == OP_ALT); - } - } - -workspace[0] = 0; /* Bit indicating which vector is current */ - + end_code += GET(end_code, 1); + length = 1 + LINK_SIZE; + } + while (*end_code == OP_ALT); + } + } + +workspace[0] = 0; /* Bit indicating which vector is current */ + DPRINTF(("%.*sEnd state = %d\n", rlevel*2-2, SP, (int)(end_code - start_code))); - -/* Loop for scanning the subject */ - -ptr = current_subject; -for (;;) - { - int i, j; - int clen, dlen; + +/* Loop for scanning the subject */ + +ptr = current_subject; +for (;;) + { + int i, j; + int clen, dlen; pcre_uint32 c, d; int forced_fail = 0; BOOL partial_newline = FALSE; BOOL could_continue = reset_could_continue; reset_could_continue = FALSE; - - /* Make the new state list into the active state list and empty the - new state list. */ - - temp_states = active_states; - active_states = new_states; - new_states = temp_states; - active_count = new_count; - new_count = 0; - - workspace[0] ^= 1; /* Remember for the restarting feature */ - workspace[1] = active_count; - + + /* Make the new state list into the active state list and empty the + new state list. */ + + temp_states = active_states; + active_states = new_states; + new_states = temp_states; + active_count = new_count; + new_count = 0; + + workspace[0] ^= 1; /* Remember for the restarting feature */ + workspace[1] = active_count; + #ifdef PCRE_DEBUG - printf("%.*sNext character: rest of subject = \"", rlevel*2-2, SP); + printf("%.*sNext character: rest of subject = \"", rlevel*2-2, SP); pchars(ptr, STRLEN_UC(ptr), stdout); - printf("\"\n"); - - printf("%.*sActive states: ", rlevel*2-2, SP); - for (i = 0; i < active_count; i++) - printf("%d/%d ", active_states[i].offset, active_states[i].count); - printf("\n"); -#endif - - /* Set the pointers for adding new states */ - - next_active_state = active_states + active_count; - next_new_state = new_states; - - /* Load the current character from the subject outside the loop, as many - different states may want to look at it, and we assume that at least one - will. */ - - if (ptr < end_subject) - { + printf("\"\n"); + + printf("%.*sActive states: ", rlevel*2-2, SP); + for (i = 0; i < active_count; i++) + printf("%d/%d ", active_states[i].offset, active_states[i].count); + printf("\n"); +#endif + + /* Set the pointers for adding new states */ + + next_active_state = active_states + active_count; + next_new_state = new_states; + + /* Load the current character from the subject outside the loop, as many + different states may want to look at it, and we assume that at least one + will. */ + + if (ptr < end_subject) + { clen = 1; /* Number of data items in the character */ #ifdef SUPPORT_UTF GETCHARLENTEST(c, ptr, clen); #else - c = *ptr; + c = *ptr; #endif /* SUPPORT_UTF */ - } - else - { - clen = 0; /* This indicates the end of the subject */ - c = NOTACHAR; /* This value should never actually be used */ - } - - /* Scan up the active states and act on each one. The result of an action - may be to add more states to the currently active list (e.g. on hitting a - parenthesis) or it may be to put states on the new list, for considering - when we move the character pointer on. */ - - for (i = 0; i < active_count; i++) - { - stateblock *current_state = active_states + i; + } + else + { + clen = 0; /* This indicates the end of the subject */ + c = NOTACHAR; /* This value should never actually be used */ + } + + /* Scan up the active states and act on each one. The result of an action + may be to add more states to the currently active list (e.g. on hitting a + parenthesis) or it may be to put states on the new list, for considering + when we move the character pointer on. */ + + for (i = 0; i < active_count; i++) + { + stateblock *current_state = active_states + i; BOOL caseless = FALSE; const pcre_uchar *code; - int state_offset = current_state->offset; + int state_offset = current_state->offset; int codevalue, rrc; int count; - + #ifdef PCRE_DEBUG - printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset); - if (clen == 0) printf("EOL\n"); - else if (c > 32 && c < 127) printf("'%c'\n", c); - else printf("0x%02x\n", c); -#endif - - /* A negative offset is a special case meaning "hold off going to this - (negated) state until the number of characters in the data field have + printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset); + if (clen == 0) printf("EOL\n"); + else if (c > 32 && c < 127) printf("'%c'\n", c); + else printf("0x%02x\n", c); +#endif + + /* A negative offset is a special case meaning "hold off going to this + (negated) state until the number of characters in the data field have been skipped". If the could_continue flag was passed over from a previous state, arrange for it to passed on. */ - - if (state_offset < 0) - { - if (current_state->data > 0) - { - DPRINTF(("%.*sSkipping this character\n", rlevel*2-2, SP)); - ADD_NEW_DATA(state_offset, current_state->count, - current_state->data - 1); + + if (state_offset < 0) + { + if (current_state->data > 0) + { + DPRINTF(("%.*sSkipping this character\n", rlevel*2-2, SP)); + ADD_NEW_DATA(state_offset, current_state->count, + current_state->data - 1); if (could_continue) reset_could_continue = TRUE; - continue; - } - else - { - current_state->offset = state_offset = -state_offset; - } - } - + continue; + } + else + { + current_state->offset = state_offset = -state_offset; + } + } + /* Check for a duplicate state with the same count, and skip if found. See the note at the head of this module about the possibility of improving performance here. */ - - for (j = 0; j < i; j++) - { - if (active_states[j].offset == state_offset && - active_states[j].count == current_state->count) - { - DPRINTF(("%.*sDuplicate state: skipped\n", rlevel*2-2, SP)); - goto NEXT_ACTIVE_STATE; - } - } - - /* The state offset is the offset to the opcode */ - - code = start_code + state_offset; - codevalue = *code; - + + for (j = 0; j < i; j++) + { + if (active_states[j].offset == state_offset && + active_states[j].count == current_state->count) + { + DPRINTF(("%.*sDuplicate state: skipped\n", rlevel*2-2, SP)); + goto NEXT_ACTIVE_STATE; + } + } + + /* The state offset is the offset to the opcode */ + + code = start_code + state_offset; + codevalue = *code; + /* If this opcode inspects a character, but we are at the end of the subject, remember the fact for use when testing for a partial match. */ if (clen == 0 && poptable[codevalue] != 0) could_continue = TRUE; - /* If this opcode is followed by an inline character, load it. It is - tempting to test for the presence of a subject character here, but that - is wrong, because sometimes zero repetitions of the subject are - permitted. - - We also use this mechanism for opcodes such as OP_TYPEPLUS that take an + /* If this opcode is followed by an inline character, load it. It is + tempting to test for the presence of a subject character here, but that + is wrong, because sometimes zero repetitions of the subject are + permitted. + + We also use this mechanism for opcodes such as OP_TYPEPLUS that take an argument that is not a data character - but is always one byte long because the values are small. We have to take special action to deal with \P, \p, \H, \h, \V, \v and \X in this case. To keep the other cases fast, convert these ones to new opcodes. */ - - if (coptable[codevalue] > 0) - { - dlen = 1; + + if (coptable[codevalue] > 0) + { + dlen = 1; #ifdef SUPPORT_UTF if (utf) { GETCHARLEN(d, (code + coptable[codevalue]), dlen); } else #endif /* SUPPORT_UTF */ - d = code[coptable[codevalue]]; - if (codevalue >= OP_TYPESTAR) - { - switch(d) - { - case OP_ANYBYTE: return PCRE_ERROR_DFA_UITEM; - case OP_NOTPROP: - case OP_PROP: codevalue += OP_PROP_EXTRA; break; - case OP_ANYNL: codevalue += OP_ANYNL_EXTRA; break; - case OP_EXTUNI: codevalue += OP_EXTUNI_EXTRA; break; - case OP_NOT_HSPACE: - case OP_HSPACE: codevalue += OP_HSPACE_EXTRA; break; - case OP_NOT_VSPACE: - case OP_VSPACE: codevalue += OP_VSPACE_EXTRA; break; - default: break; - } - } - } - else - { - dlen = 0; /* Not strictly necessary, but compilers moan */ - d = NOTACHAR; /* if these variables are not set. */ - } - - - /* Now process the individual opcodes */ - - switch (codevalue) - { + d = code[coptable[codevalue]]; + if (codevalue >= OP_TYPESTAR) + { + switch(d) + { + case OP_ANYBYTE: return PCRE_ERROR_DFA_UITEM; + case OP_NOTPROP: + case OP_PROP: codevalue += OP_PROP_EXTRA; break; + case OP_ANYNL: codevalue += OP_ANYNL_EXTRA; break; + case OP_EXTUNI: codevalue += OP_EXTUNI_EXTRA; break; + case OP_NOT_HSPACE: + case OP_HSPACE: codevalue += OP_HSPACE_EXTRA; break; + case OP_NOT_VSPACE: + case OP_VSPACE: codevalue += OP_VSPACE_EXTRA; break; + default: break; + } + } + } + else + { + dlen = 0; /* Not strictly necessary, but compilers moan */ + d = NOTACHAR; /* if these variables are not set. */ + } + + + /* Now process the individual opcodes */ + + switch (codevalue) + { /* ========================================================================== */ /* These cases are never obeyed. This is a fudge that causes a compile- time error if the vectors coptable or poptable, which are indexed by opcode, are not the correct length. It seems to be the only way to do such a check at compile time, as the sizeof() operator does not work in the C preprocessor. */ - + case OP_TABLE_LENGTH: case OP_TABLE_LENGTH + ((sizeof(coptable) == OP_TABLE_LENGTH) && (sizeof(poptable) == OP_TABLE_LENGTH)): break; -/* ========================================================================== */ - /* Reached a closing bracket. If not at the end of the pattern, carry +/* ========================================================================== */ + /* Reached a closing bracket. If not at the end of the pattern, carry on with the next opcode. For repeating opcodes, also add the repeat state. Note that KETRPOS will always be encountered at the end of the subpattern, because the possessive subpattern repeats are always handled @@ -766,27 +766,27 @@ for (;;) At the end of the (sub)pattern, unless we have an empty string and PCRE_NOTEMPTY is set, or PCRE_NOTEMPTY_ATSTART is set and we are at the start of the subject, save the match data, shifting up all previous - matches so we always have the longest first. */ - - case OP_KET: - case OP_KETRMIN: - case OP_KETRMAX: + matches so we always have the longest first. */ + + case OP_KET: + case OP_KETRMIN: + case OP_KETRMAX: case OP_KETRPOS: - if (code != end_code) - { - ADD_ACTIVE(state_offset + 1 + LINK_SIZE, 0); - if (codevalue != OP_KET) - { - ADD_ACTIVE(state_offset - GET(code, 1), 0); - } - } + if (code != end_code) + { + ADD_ACTIVE(state_offset + 1 + LINK_SIZE, 0); + if (codevalue != OP_KET) + { + ADD_ACTIVE(state_offset - GET(code, 1), 0); + } + } else - { + { if (ptr > current_subject || ((md->moptions & PCRE_NOTEMPTY) == 0 && ((md->moptions & PCRE_NOTEMPTY_ATSTART) == 0 || current_subject > start_subject + md->start_offset))) - { + { if (match_count < 0) match_count = (offsetcount >= 2)? 1 : 0; else if (match_count > 0 && ++match_count * 2 > offsetcount) match_count = 0; @@ -806,53 +806,53 @@ for (;;) match_count, rlevel*2-2, SP)); return match_count; } - } - } - break; - -/* ========================================================================== */ - /* These opcodes add to the current list of states without looking - at the current character. */ - - /*-----------------------------------------------------------------*/ - case OP_ALT: - do { code += GET(code, 1); } while (*code == OP_ALT); + } + } + break; + +/* ========================================================================== */ + /* These opcodes add to the current list of states without looking + at the current character. */ + + /*-----------------------------------------------------------------*/ + case OP_ALT: + do { code += GET(code, 1); } while (*code == OP_ALT); ADD_ACTIVE((int)(code - start_code), 0); - break; - - /*-----------------------------------------------------------------*/ - case OP_BRA: - case OP_SBRA: - do - { + break; + + /*-----------------------------------------------------------------*/ + case OP_BRA: + case OP_SBRA: + do + { ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0); - code += GET(code, 1); - } - while (*code == OP_ALT); - break; - - /*-----------------------------------------------------------------*/ - case OP_CBRA: - case OP_SCBRA: + code += GET(code, 1); + } + while (*code == OP_ALT); + break; + + /*-----------------------------------------------------------------*/ + case OP_CBRA: + case OP_SCBRA: ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE + IMM2_SIZE), 0); - code += GET(code, 1); - while (*code == OP_ALT) - { + code += GET(code, 1); + while (*code == OP_ALT) + { ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0); - code += GET(code, 1); - } - break; - - /*-----------------------------------------------------------------*/ - case OP_BRAZERO: - case OP_BRAMINZERO: - ADD_ACTIVE(state_offset + 1, 0); - code += 1 + GET(code, 2); - while (*code == OP_ALT) code += GET(code, 1); + code += GET(code, 1); + } + break; + + /*-----------------------------------------------------------------*/ + case OP_BRAZERO: + case OP_BRAMINZERO: + ADD_ACTIVE(state_offset + 1, 0); + code += 1 + GET(code, 2); + while (*code == OP_ALT) code += GET(code, 1); ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0); - break; - - /*-----------------------------------------------------------------*/ + break; + + /*-----------------------------------------------------------------*/ case OP_SKIPZERO: code += 1 + GET(code, 2); while (*code == OP_ALT) code += GET(code, 1); @@ -860,19 +860,19 @@ for (;;) break; /*-----------------------------------------------------------------*/ - case OP_CIRC: + case OP_CIRC: if (ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0) - { ADD_ACTIVE(state_offset + 1, 0); } - break; - - /*-----------------------------------------------------------------*/ + { ADD_ACTIVE(state_offset + 1, 0); } + break; + + /*-----------------------------------------------------------------*/ case OP_CIRCM: if ((ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0) || (ptr != end_subject && WAS_NEWLINE(ptr))) { ADD_ACTIVE(state_offset + 1, 0); } - break; - - /*-----------------------------------------------------------------*/ + break; + + /*-----------------------------------------------------------------*/ case OP_EOD: if (ptr >= end_subject) { @@ -880,27 +880,27 @@ for (;;) could_continue = TRUE; else { ADD_ACTIVE(state_offset + 1, 0); } } - break; - - /*-----------------------------------------------------------------*/ - case OP_SOD: - if (ptr == start_subject) { ADD_ACTIVE(state_offset + 1, 0); } - break; - - /*-----------------------------------------------------------------*/ - case OP_SOM: - if (ptr == start_subject + start_offset) { ADD_ACTIVE(state_offset + 1, 0); } - break; - - -/* ========================================================================== */ - /* These opcodes inspect the next subject character, and sometimes - the previous one as well, but do not have an argument. The variable - clen contains the length of the current character and is zero if we are - at the end of the subject. */ - - /*-----------------------------------------------------------------*/ - case OP_ANY: + break; + + /*-----------------------------------------------------------------*/ + case OP_SOD: + if (ptr == start_subject) { ADD_ACTIVE(state_offset + 1, 0); } + break; + + /*-----------------------------------------------------------------*/ + case OP_SOM: + if (ptr == start_subject + start_offset) { ADD_ACTIVE(state_offset + 1, 0); } + break; + + +/* ========================================================================== */ + /* These opcodes inspect the next subject character, and sometimes + the previous one as well, but do not have an argument. The variable + clen contains the length of the current character and is zero if we are + at the end of the subject. */ + + /*-----------------------------------------------------------------*/ + case OP_ANY: if (clen > 0 && !IS_NEWLINE(ptr)) { if (ptr + 1 >= md->end_subject && @@ -921,28 +921,28 @@ for (;;) /*-----------------------------------------------------------------*/ case OP_ALLANY: if (clen > 0) - { ADD_NEW(state_offset + 1, 0); } - break; - - /*-----------------------------------------------------------------*/ - case OP_EODN: + { ADD_NEW(state_offset + 1, 0); } + break; + + /*-----------------------------------------------------------------*/ + case OP_EODN: if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0) could_continue = TRUE; else if (clen == 0 || (IS_NEWLINE(ptr) && ptr == end_subject - md->nllen)) - { ADD_ACTIVE(state_offset + 1, 0); } - break; - - /*-----------------------------------------------------------------*/ - case OP_DOLL: - if ((md->moptions & PCRE_NOTEOL) == 0) - { + { ADD_ACTIVE(state_offset + 1, 0); } + break; + + /*-----------------------------------------------------------------*/ + case OP_DOLL: + if ((md->moptions & PCRE_NOTEOL) == 0) + { if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0) could_continue = TRUE; else if (clen == 0 || ((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr) && (ptr == end_subject - md->nllen) - )) - { ADD_ACTIVE(state_offset + 1, 0); } + )) + { ADD_ACTIVE(state_offset + 1, 0); } else if (ptr + 1 >= md->end_subject && (md->moptions & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) != 0 && NLBLOCK->nltype == NLTYPE_FIXED && @@ -956,7 +956,7 @@ for (;;) } else could_continue = partial_newline = TRUE; } - } + } break; /*-----------------------------------------------------------------*/ @@ -983,42 +983,42 @@ for (;;) } } else if (IS_NEWLINE(ptr)) - { ADD_ACTIVE(state_offset + 1, 0); } - break; - - /*-----------------------------------------------------------------*/ - - case OP_DIGIT: - case OP_WHITESPACE: - case OP_WORDCHAR: - if (clen > 0 && c < 256 && - ((ctypes[c] & toptable1[codevalue]) ^ toptable2[codevalue]) != 0) - { ADD_NEW(state_offset + 1, 0); } - break; - - /*-----------------------------------------------------------------*/ - case OP_NOT_DIGIT: - case OP_NOT_WHITESPACE: - case OP_NOT_WORDCHAR: - if (clen > 0 && (c >= 256 || - ((ctypes[c] & toptable1[codevalue]) ^ toptable2[codevalue]) != 0)) - { ADD_NEW(state_offset + 1, 0); } - break; - - /*-----------------------------------------------------------------*/ - case OP_WORD_BOUNDARY: - case OP_NOT_WORD_BOUNDARY: - { - int left_word, right_word; - - if (ptr > start_subject) - { + { ADD_ACTIVE(state_offset + 1, 0); } + break; + + /*-----------------------------------------------------------------*/ + + case OP_DIGIT: + case OP_WHITESPACE: + case OP_WORDCHAR: + if (clen > 0 && c < 256 && + ((ctypes[c] & toptable1[codevalue]) ^ toptable2[codevalue]) != 0) + { ADD_NEW(state_offset + 1, 0); } + break; + + /*-----------------------------------------------------------------*/ + case OP_NOT_DIGIT: + case OP_NOT_WHITESPACE: + case OP_NOT_WORDCHAR: + if (clen > 0 && (c >= 256 || + ((ctypes[c] & toptable1[codevalue]) ^ toptable2[codevalue]) != 0)) + { ADD_NEW(state_offset + 1, 0); } + break; + + /*-----------------------------------------------------------------*/ + case OP_WORD_BOUNDARY: + case OP_NOT_WORD_BOUNDARY: + { + int left_word, right_word; + + if (ptr > start_subject) + { const pcre_uchar *temp = ptr - 1; if (temp < md->start_used_ptr) md->start_used_ptr = temp; #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf) { BACKCHAR(temp); } -#endif - GETCHARTEST(d, temp); +#endif + GETCHARTEST(d, temp); #ifdef SUPPORT_UCP if ((md->poptions & PCRE_UCP) != 0) { @@ -1030,10 +1030,10 @@ for (;;) } else #endif - left_word = d < 256 && (ctypes[d] & ctype_word) != 0; - } + left_word = d < 256 && (ctypes[d] & ctype_word) != 0; + } else left_word = FALSE; - + if (clen > 0) { #ifdef SUPPORT_UCP @@ -1050,49 +1050,49 @@ for (;;) right_word = c < 256 && (ctypes[c] & ctype_word) != 0; } else right_word = FALSE; - - if ((left_word == right_word) == (codevalue == OP_NOT_WORD_BOUNDARY)) - { ADD_ACTIVE(state_offset + 1, 0); } - } - break; - - - /*-----------------------------------------------------------------*/ - /* Check the next character by Unicode property. We will get here only - if the support is in the binary; otherwise a compile-time error occurs. - */ - -#ifdef SUPPORT_UCP - case OP_PROP: - case OP_NOTPROP: - if (clen > 0) - { - BOOL OK; + + if ((left_word == right_word) == (codevalue == OP_NOT_WORD_BOUNDARY)) + { ADD_ACTIVE(state_offset + 1, 0); } + } + break; + + + /*-----------------------------------------------------------------*/ + /* Check the next character by Unicode property. We will get here only + if the support is in the binary; otherwise a compile-time error occurs. + */ + +#ifdef SUPPORT_UCP + case OP_PROP: + case OP_NOTPROP: + if (clen > 0) + { + BOOL OK; const pcre_uint32 *cp; const ucd_record * prop = GET_UCD(c); - switch(code[1]) - { - case PT_ANY: - OK = TRUE; - break; - - case PT_LAMP: + switch(code[1]) + { + case PT_ANY: + OK = TRUE; + break; + + case PT_LAMP: OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; - break; - - case PT_GC: + break; + + case PT_GC: OK = PRIV(ucp_gentype)[prop->chartype] == code[2]; - break; - - case PT_PC: + break; + + case PT_PC: OK = prop->chartype == code[2]; - break; - - case PT_SC: + break; + + case PT_SC: OK = prop->script == code[2]; - break; - + break; + /* These are specials for combination cases. */ case PT_ALNUM: @@ -1140,33 +1140,33 @@ for (;;) c >= 0xe000; break; - /* Should never occur, but keep compilers from grumbling. */ - - default: - OK = codevalue != OP_PROP; - break; - } - - if (OK == (codevalue == OP_PROP)) { ADD_NEW(state_offset + 3, 0); } - } - break; -#endif - - - -/* ========================================================================== */ - /* These opcodes likewise inspect the subject character, but have an - argument that is not a data character. It is one of these opcodes: + /* Should never occur, but keep compilers from grumbling. */ + + default: + OK = codevalue != OP_PROP; + break; + } + + if (OK == (codevalue == OP_PROP)) { ADD_NEW(state_offset + 3, 0); } + } + break; +#endif + + + +/* ========================================================================== */ + /* These opcodes likewise inspect the subject character, but have an + argument that is not a data character. It is one of these opcodes: OP_ANY, OP_ALLANY, OP_DIGIT, OP_NOT_DIGIT, OP_WHITESPACE, OP_NOT_SPACE, OP_WORDCHAR, OP_NOT_WORDCHAR. The value is loaded into d. */ - - case OP_TYPEPLUS: - case OP_TYPEMINPLUS: - case OP_TYPEPOSPLUS: - count = current_state->count; /* Already matched */ - if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } - if (clen > 0) - { + + case OP_TYPEPLUS: + case OP_TYPEMINPLUS: + case OP_TYPEPOSPLUS: + count = current_state->count; /* Already matched */ + if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } + if (clen > 0) + { if (d == OP_ANY && ptr + 1 >= md->end_subject && (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && NLBLOCK->nltype == NLTYPE_FIXED && @@ -1176,28 +1176,28 @@ for (;;) could_continue = partial_newline = TRUE; } else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || - (c < 256 && + (c < 256 && (d != OP_ANY || !IS_NEWLINE(ptr)) && - ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) - { - if (count > 0 && codevalue == OP_TYPEPOSPLUS) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - count++; - ADD_NEW(state_offset, count); - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_TYPEQUERY: - case OP_TYPEMINQUERY: - case OP_TYPEPOSQUERY: - ADD_ACTIVE(state_offset + 2, 0); - if (clen > 0) - { + ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) + { + if (count > 0 && codevalue == OP_TYPEPOSPLUS) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + count++; + ADD_NEW(state_offset, count); + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_TYPEQUERY: + case OP_TYPEMINQUERY: + case OP_TYPEPOSQUERY: + ADD_ACTIVE(state_offset + 2, 0); + if (clen > 0) + { if (d == OP_ANY && ptr + 1 >= md->end_subject && (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && NLBLOCK->nltype == NLTYPE_FIXED && @@ -1207,27 +1207,27 @@ for (;;) could_continue = partial_newline = TRUE; } else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || - (c < 256 && + (c < 256 && (d != OP_ANY || !IS_NEWLINE(ptr)) && - ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) - { - if (codevalue == OP_TYPEPOSQUERY) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - ADD_NEW(state_offset + 2, 0); - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_TYPESTAR: - case OP_TYPEMINSTAR: - case OP_TYPEPOSSTAR: - ADD_ACTIVE(state_offset + 2, 0); - if (clen > 0) - { + ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) + { + if (codevalue == OP_TYPEPOSQUERY) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + ADD_NEW(state_offset + 2, 0); + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_TYPESTAR: + case OP_TYPEMINSTAR: + case OP_TYPEPOSSTAR: + ADD_ACTIVE(state_offset + 2, 0); + if (clen > 0) + { if (d == OP_ANY && ptr + 1 >= md->end_subject && (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && NLBLOCK->nltype == NLTYPE_FIXED && @@ -1237,25 +1237,25 @@ for (;;) could_continue = partial_newline = TRUE; } else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || - (c < 256 && + (c < 256 && (d != OP_ANY || !IS_NEWLINE(ptr)) && - ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) - { - if (codevalue == OP_TYPEPOSSTAR) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - ADD_NEW(state_offset, 0); - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_TYPEEXACT: - count = current_state->count; /* Number already matched */ - if (clen > 0) - { + ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) + { + if (codevalue == OP_TYPEPOSSTAR) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + ADD_NEW(state_offset, 0); + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_TYPEEXACT: + count = current_state->count; /* Number already matched */ + if (clen > 0) + { if (d == OP_ANY && ptr + 1 >= md->end_subject && (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && NLBLOCK->nltype == NLTYPE_FIXED && @@ -1265,26 +1265,26 @@ for (;;) could_continue = partial_newline = TRUE; } else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || - (c < 256 && + (c < 256 && (d != OP_ANY || !IS_NEWLINE(ptr)) && - ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) - { + ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) + { if (++count >= (int)GET2(code, 1)) { ADD_NEW(state_offset + 1 + IMM2_SIZE + 1, 0); } - else - { ADD_NEW(state_offset, count); } - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_TYPEUPTO: - case OP_TYPEMINUPTO: - case OP_TYPEPOSUPTO: + else + { ADD_NEW(state_offset, count); } + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_TYPEUPTO: + case OP_TYPEMINUPTO: + case OP_TYPEPOSUPTO: ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); - count = current_state->count; /* Number already matched */ - if (clen > 0) - { + count = current_state->count; /* Number already matched */ + if (clen > 0) + { if (d == OP_ANY && ptr + 1 >= md->end_subject && (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && NLBLOCK->nltype == NLTYPE_FIXED && @@ -1294,63 +1294,63 @@ for (;;) could_continue = partial_newline = TRUE; } else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || - (c < 256 && + (c < 256 && (d != OP_ANY || !IS_NEWLINE(ptr)) && - ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) - { - if (codevalue == OP_TYPEPOSUPTO) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) + { + if (codevalue == OP_TYPEPOSUPTO) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } if (++count >= (int)GET2(code, 1)) { ADD_NEW(state_offset + 2 + IMM2_SIZE, 0); } - else - { ADD_NEW(state_offset, count); } - } - } - break; - -/* ========================================================================== */ - /* These are virtual opcodes that are used when something like - OP_TYPEPLUS has OP_PROP, OP_NOTPROP, OP_ANYNL, or OP_EXTUNI as its - argument. It keeps the code above fast for the other cases. The argument - is in the d variable. */ - -#ifdef SUPPORT_UCP - case OP_PROP_EXTRA + OP_TYPEPLUS: - case OP_PROP_EXTRA + OP_TYPEMINPLUS: - case OP_PROP_EXTRA + OP_TYPEPOSPLUS: - count = current_state->count; /* Already matched */ - if (count > 0) { ADD_ACTIVE(state_offset + 4, 0); } - if (clen > 0) - { - BOOL OK; + else + { ADD_NEW(state_offset, count); } + } + } + break; + +/* ========================================================================== */ + /* These are virtual opcodes that are used when something like + OP_TYPEPLUS has OP_PROP, OP_NOTPROP, OP_ANYNL, or OP_EXTUNI as its + argument. It keeps the code above fast for the other cases. The argument + is in the d variable. */ + +#ifdef SUPPORT_UCP + case OP_PROP_EXTRA + OP_TYPEPLUS: + case OP_PROP_EXTRA + OP_TYPEMINPLUS: + case OP_PROP_EXTRA + OP_TYPEPOSPLUS: + count = current_state->count; /* Already matched */ + if (count > 0) { ADD_ACTIVE(state_offset + 4, 0); } + if (clen > 0) + { + BOOL OK; const pcre_uint32 *cp; const ucd_record * prop = GET_UCD(c); - switch(code[2]) - { - case PT_ANY: - OK = TRUE; - break; - - case PT_LAMP: + switch(code[2]) + { + case PT_ANY: + OK = TRUE; + break; + + case PT_LAMP: OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; - break; - - case PT_GC: + break; + + case PT_GC: OK = PRIV(ucp_gentype)[prop->chartype] == code[3]; - break; - - case PT_PC: + break; + + case PT_PC: OK = prop->chartype == code[3]; - break; - - case PT_SC: + break; + + case PT_SC: OK = prop->script == code[3]; - break; - + break; + /* These are specials for combination cases. */ case PT_ALNUM: @@ -1398,211 +1398,211 @@ for (;;) c >= 0xe000; break; - /* Should never occur, but keep compilers from grumbling. */ - - default: - OK = codevalue != OP_PROP; - break; - } - - if (OK == (d == OP_PROP)) - { - if (count > 0 && codevalue == OP_PROP_EXTRA + OP_TYPEPOSPLUS) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - count++; - ADD_NEW(state_offset, count); - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_EXTUNI_EXTRA + OP_TYPEPLUS: - case OP_EXTUNI_EXTRA + OP_TYPEMINPLUS: - case OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS: - count = current_state->count; /* Already matched */ - if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } + /* Should never occur, but keep compilers from grumbling. */ + + default: + OK = codevalue != OP_PROP; + break; + } + + if (OK == (d == OP_PROP)) + { + if (count > 0 && codevalue == OP_PROP_EXTRA + OP_TYPEPOSPLUS) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + count++; + ADD_NEW(state_offset, count); + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_EXTUNI_EXTRA + OP_TYPEPLUS: + case OP_EXTUNI_EXTRA + OP_TYPEMINPLUS: + case OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS: + count = current_state->count; /* Already matched */ + if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } if (clen > 0) - { + { int lgb, rgb; const pcre_uchar *nptr = ptr + clen; - int ncount = 0; - if (count > 0 && codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + int ncount = 0; + if (count > 0 && codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } lgb = UCD_GRAPHBREAK(c); - while (nptr < end_subject) - { + while (nptr < end_subject) + { dlen = 1; if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); } rgb = UCD_GRAPHBREAK(d); if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; - ncount++; + ncount++; lgb = rgb; nptr += dlen; - } - count++; - ADD_NEW_DATA(-state_offset, count, ncount); - } - break; -#endif - - /*-----------------------------------------------------------------*/ - case OP_ANYNL_EXTRA + OP_TYPEPLUS: - case OP_ANYNL_EXTRA + OP_TYPEMINPLUS: - case OP_ANYNL_EXTRA + OP_TYPEPOSPLUS: - count = current_state->count; /* Already matched */ - if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } - if (clen > 0) - { - int ncount = 0; - switch (c) - { + } + count++; + ADD_NEW_DATA(-state_offset, count, ncount); + } + break; +#endif + + /*-----------------------------------------------------------------*/ + case OP_ANYNL_EXTRA + OP_TYPEPLUS: + case OP_ANYNL_EXTRA + OP_TYPEMINPLUS: + case OP_ANYNL_EXTRA + OP_TYPEPOSPLUS: + count = current_state->count; /* Already matched */ + if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } + if (clen > 0) + { + int ncount = 0; + switch (c) + { case CHAR_VT: case CHAR_FF: case CHAR_NEL: #ifndef EBCDIC - case 0x2028: - case 0x2029: + case 0x2028: + case 0x2029: #endif /* Not EBCDIC */ - if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; - goto ANYNL01; - + if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; + goto ANYNL01; + case CHAR_CR: if (ptr + 1 < end_subject && UCHAR21TEST(ptr + 1) == CHAR_LF) ncount = 1; - /* Fall through */ - - ANYNL01: + /* Fall through */ + + ANYNL01: case CHAR_LF: - if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - count++; - ADD_NEW_DATA(-state_offset, count, ncount); - break; - - default: - break; - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_VSPACE_EXTRA + OP_TYPEPLUS: - case OP_VSPACE_EXTRA + OP_TYPEMINPLUS: - case OP_VSPACE_EXTRA + OP_TYPEPOSPLUS: - count = current_state->count; /* Already matched */ - if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } - if (clen > 0) - { - BOOL OK; - switch (c) - { + if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + count++; + ADD_NEW_DATA(-state_offset, count, ncount); + break; + + default: + break; + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_VSPACE_EXTRA + OP_TYPEPLUS: + case OP_VSPACE_EXTRA + OP_TYPEMINPLUS: + case OP_VSPACE_EXTRA + OP_TYPEPOSPLUS: + count = current_state->count; /* Already matched */ + if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } + if (clen > 0) + { + BOOL OK; + switch (c) + { VSPACE_CASES: - OK = TRUE; - break; - - default: - OK = FALSE; - break; - } - - if (OK == (d == OP_VSPACE)) - { - if (count > 0 && codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSPLUS) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - count++; - ADD_NEW_DATA(-state_offset, count, 0); - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_HSPACE_EXTRA + OP_TYPEPLUS: - case OP_HSPACE_EXTRA + OP_TYPEMINPLUS: - case OP_HSPACE_EXTRA + OP_TYPEPOSPLUS: - count = current_state->count; /* Already matched */ - if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } - if (clen > 0) - { - BOOL OK; - switch (c) - { + OK = TRUE; + break; + + default: + OK = FALSE; + break; + } + + if (OK == (d == OP_VSPACE)) + { + if (count > 0 && codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSPLUS) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + count++; + ADD_NEW_DATA(-state_offset, count, 0); + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_HSPACE_EXTRA + OP_TYPEPLUS: + case OP_HSPACE_EXTRA + OP_TYPEMINPLUS: + case OP_HSPACE_EXTRA + OP_TYPEPOSPLUS: + count = current_state->count; /* Already matched */ + if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } + if (clen > 0) + { + BOOL OK; + switch (c) + { HSPACE_CASES: - OK = TRUE; - break; - - default: - OK = FALSE; - break; - } - - if (OK == (d == OP_HSPACE)) - { - if (count > 0 && codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSPLUS) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - count++; - ADD_NEW_DATA(-state_offset, count, 0); - } - } - break; - - /*-----------------------------------------------------------------*/ -#ifdef SUPPORT_UCP - case OP_PROP_EXTRA + OP_TYPEQUERY: - case OP_PROP_EXTRA + OP_TYPEMINQUERY: - case OP_PROP_EXTRA + OP_TYPEPOSQUERY: - count = 4; - goto QS1; - - case OP_PROP_EXTRA + OP_TYPESTAR: - case OP_PROP_EXTRA + OP_TYPEMINSTAR: - case OP_PROP_EXTRA + OP_TYPEPOSSTAR: - count = 0; - - QS1: - - ADD_ACTIVE(state_offset + 4, 0); - if (clen > 0) - { - BOOL OK; + OK = TRUE; + break; + + default: + OK = FALSE; + break; + } + + if (OK == (d == OP_HSPACE)) + { + if (count > 0 && codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSPLUS) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + count++; + ADD_NEW_DATA(-state_offset, count, 0); + } + } + break; + + /*-----------------------------------------------------------------*/ +#ifdef SUPPORT_UCP + case OP_PROP_EXTRA + OP_TYPEQUERY: + case OP_PROP_EXTRA + OP_TYPEMINQUERY: + case OP_PROP_EXTRA + OP_TYPEPOSQUERY: + count = 4; + goto QS1; + + case OP_PROP_EXTRA + OP_TYPESTAR: + case OP_PROP_EXTRA + OP_TYPEMINSTAR: + case OP_PROP_EXTRA + OP_TYPEPOSSTAR: + count = 0; + + QS1: + + ADD_ACTIVE(state_offset + 4, 0); + if (clen > 0) + { + BOOL OK; const pcre_uint32 *cp; const ucd_record * prop = GET_UCD(c); - switch(code[2]) - { - case PT_ANY: - OK = TRUE; - break; - - case PT_LAMP: + switch(code[2]) + { + case PT_ANY: + OK = TRUE; + break; + + case PT_LAMP: OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; - break; - - case PT_GC: + break; + + case PT_GC: OK = PRIV(ucp_gentype)[prop->chartype] == code[3]; - break; - - case PT_PC: + break; + + case PT_PC: OK = prop->chartype == code[3]; - break; - - case PT_SC: + break; + + case PT_SC: OK = prop->script == code[3]; - break; - + break; + /* These are specials for combination cases. */ case PT_ALNUM: @@ -1650,236 +1650,236 @@ for (;;) c >= 0xe000; break; - /* Should never occur, but keep compilers from grumbling. */ - - default: - OK = codevalue != OP_PROP; - break; - } - - if (OK == (d == OP_PROP)) - { - if (codevalue == OP_PROP_EXTRA + OP_TYPEPOSSTAR || - codevalue == OP_PROP_EXTRA + OP_TYPEPOSQUERY) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - ADD_NEW(state_offset + count, 0); - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_EXTUNI_EXTRA + OP_TYPEQUERY: - case OP_EXTUNI_EXTRA + OP_TYPEMINQUERY: - case OP_EXTUNI_EXTRA + OP_TYPEPOSQUERY: - count = 2; - goto QS2; - - case OP_EXTUNI_EXTRA + OP_TYPESTAR: - case OP_EXTUNI_EXTRA + OP_TYPEMINSTAR: - case OP_EXTUNI_EXTRA + OP_TYPEPOSSTAR: - count = 0; - - QS2: - - ADD_ACTIVE(state_offset + 2, 0); + /* Should never occur, but keep compilers from grumbling. */ + + default: + OK = codevalue != OP_PROP; + break; + } + + if (OK == (d == OP_PROP)) + { + if (codevalue == OP_PROP_EXTRA + OP_TYPEPOSSTAR || + codevalue == OP_PROP_EXTRA + OP_TYPEPOSQUERY) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + ADD_NEW(state_offset + count, 0); + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_EXTUNI_EXTRA + OP_TYPEQUERY: + case OP_EXTUNI_EXTRA + OP_TYPEMINQUERY: + case OP_EXTUNI_EXTRA + OP_TYPEPOSQUERY: + count = 2; + goto QS2; + + case OP_EXTUNI_EXTRA + OP_TYPESTAR: + case OP_EXTUNI_EXTRA + OP_TYPEMINSTAR: + case OP_EXTUNI_EXTRA + OP_TYPEPOSSTAR: + count = 0; + + QS2: + + ADD_ACTIVE(state_offset + 2, 0); if (clen > 0) - { + { int lgb, rgb; const pcre_uchar *nptr = ptr + clen; - int ncount = 0; - if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSSTAR || - codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSQUERY) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + int ncount = 0; + if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSSTAR || + codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSQUERY) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } lgb = UCD_GRAPHBREAK(c); - while (nptr < end_subject) - { + while (nptr < end_subject) + { dlen = 1; if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); } rgb = UCD_GRAPHBREAK(d); if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; - ncount++; + ncount++; lgb = rgb; nptr += dlen; - } - ADD_NEW_DATA(-(state_offset + count), 0, ncount); - } - break; -#endif - - /*-----------------------------------------------------------------*/ - case OP_ANYNL_EXTRA + OP_TYPEQUERY: - case OP_ANYNL_EXTRA + OP_TYPEMINQUERY: - case OP_ANYNL_EXTRA + OP_TYPEPOSQUERY: - count = 2; - goto QS3; - - case OP_ANYNL_EXTRA + OP_TYPESTAR: - case OP_ANYNL_EXTRA + OP_TYPEMINSTAR: - case OP_ANYNL_EXTRA + OP_TYPEPOSSTAR: - count = 0; - - QS3: - ADD_ACTIVE(state_offset + 2, 0); - if (clen > 0) - { - int ncount = 0; - switch (c) - { + } + ADD_NEW_DATA(-(state_offset + count), 0, ncount); + } + break; +#endif + + /*-----------------------------------------------------------------*/ + case OP_ANYNL_EXTRA + OP_TYPEQUERY: + case OP_ANYNL_EXTRA + OP_TYPEMINQUERY: + case OP_ANYNL_EXTRA + OP_TYPEPOSQUERY: + count = 2; + goto QS3; + + case OP_ANYNL_EXTRA + OP_TYPESTAR: + case OP_ANYNL_EXTRA + OP_TYPEMINSTAR: + case OP_ANYNL_EXTRA + OP_TYPEPOSSTAR: + count = 0; + + QS3: + ADD_ACTIVE(state_offset + 2, 0); + if (clen > 0) + { + int ncount = 0; + switch (c) + { case CHAR_VT: case CHAR_FF: case CHAR_NEL: #ifndef EBCDIC - case 0x2028: - case 0x2029: + case 0x2028: + case 0x2029: #endif /* Not EBCDIC */ - if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; - goto ANYNL02; - + if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; + goto ANYNL02; + case CHAR_CR: if (ptr + 1 < end_subject && UCHAR21TEST(ptr + 1) == CHAR_LF) ncount = 1; - /* Fall through */ - - ANYNL02: + /* Fall through */ + + ANYNL02: case CHAR_LF: - if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR || - codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR || + codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } ADD_NEW_DATA(-(state_offset + (int)count), 0, ncount); - break; - - default: - break; - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_VSPACE_EXTRA + OP_TYPEQUERY: - case OP_VSPACE_EXTRA + OP_TYPEMINQUERY: - case OP_VSPACE_EXTRA + OP_TYPEPOSQUERY: - count = 2; - goto QS4; - - case OP_VSPACE_EXTRA + OP_TYPESTAR: - case OP_VSPACE_EXTRA + OP_TYPEMINSTAR: - case OP_VSPACE_EXTRA + OP_TYPEPOSSTAR: - count = 0; - - QS4: - ADD_ACTIVE(state_offset + 2, 0); - if (clen > 0) - { - BOOL OK; - switch (c) - { + break; + + default: + break; + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_VSPACE_EXTRA + OP_TYPEQUERY: + case OP_VSPACE_EXTRA + OP_TYPEMINQUERY: + case OP_VSPACE_EXTRA + OP_TYPEPOSQUERY: + count = 2; + goto QS4; + + case OP_VSPACE_EXTRA + OP_TYPESTAR: + case OP_VSPACE_EXTRA + OP_TYPEMINSTAR: + case OP_VSPACE_EXTRA + OP_TYPEPOSSTAR: + count = 0; + + QS4: + ADD_ACTIVE(state_offset + 2, 0); + if (clen > 0) + { + BOOL OK; + switch (c) + { VSPACE_CASES: - OK = TRUE; - break; - - default: - OK = FALSE; - break; - } - if (OK == (d == OP_VSPACE)) - { - if (codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSSTAR || - codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSQUERY) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + OK = TRUE; + break; + + default: + OK = FALSE; + break; + } + if (OK == (d == OP_VSPACE)) + { + if (codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSSTAR || + codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSQUERY) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } ADD_NEW_DATA(-(state_offset + (int)count), 0, 0); - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_HSPACE_EXTRA + OP_TYPEQUERY: - case OP_HSPACE_EXTRA + OP_TYPEMINQUERY: - case OP_HSPACE_EXTRA + OP_TYPEPOSQUERY: - count = 2; - goto QS5; - - case OP_HSPACE_EXTRA + OP_TYPESTAR: - case OP_HSPACE_EXTRA + OP_TYPEMINSTAR: - case OP_HSPACE_EXTRA + OP_TYPEPOSSTAR: - count = 0; - - QS5: - ADD_ACTIVE(state_offset + 2, 0); - if (clen > 0) - { - BOOL OK; - switch (c) - { + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_HSPACE_EXTRA + OP_TYPEQUERY: + case OP_HSPACE_EXTRA + OP_TYPEMINQUERY: + case OP_HSPACE_EXTRA + OP_TYPEPOSQUERY: + count = 2; + goto QS5; + + case OP_HSPACE_EXTRA + OP_TYPESTAR: + case OP_HSPACE_EXTRA + OP_TYPEMINSTAR: + case OP_HSPACE_EXTRA + OP_TYPEPOSSTAR: + count = 0; + + QS5: + ADD_ACTIVE(state_offset + 2, 0); + if (clen > 0) + { + BOOL OK; + switch (c) + { HSPACE_CASES: - OK = TRUE; - break; - - default: - OK = FALSE; - break; - } - - if (OK == (d == OP_HSPACE)) - { - if (codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSSTAR || - codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSQUERY) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + OK = TRUE; + break; + + default: + OK = FALSE; + break; + } + + if (OK == (d == OP_HSPACE)) + { + if (codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSSTAR || + codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSQUERY) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } ADD_NEW_DATA(-(state_offset + (int)count), 0, 0); - } - } - break; - - /*-----------------------------------------------------------------*/ -#ifdef SUPPORT_UCP - case OP_PROP_EXTRA + OP_TYPEEXACT: - case OP_PROP_EXTRA + OP_TYPEUPTO: - case OP_PROP_EXTRA + OP_TYPEMINUPTO: - case OP_PROP_EXTRA + OP_TYPEPOSUPTO: - if (codevalue != OP_PROP_EXTRA + OP_TYPEEXACT) + } + } + break; + + /*-----------------------------------------------------------------*/ +#ifdef SUPPORT_UCP + case OP_PROP_EXTRA + OP_TYPEEXACT: + case OP_PROP_EXTRA + OP_TYPEUPTO: + case OP_PROP_EXTRA + OP_TYPEMINUPTO: + case OP_PROP_EXTRA + OP_TYPEPOSUPTO: + if (codevalue != OP_PROP_EXTRA + OP_TYPEEXACT) { ADD_ACTIVE(state_offset + 1 + IMM2_SIZE + 3, 0); } - count = current_state->count; /* Number already matched */ - if (clen > 0) - { - BOOL OK; + count = current_state->count; /* Number already matched */ + if (clen > 0) + { + BOOL OK; const pcre_uint32 *cp; const ucd_record * prop = GET_UCD(c); switch(code[1 + IMM2_SIZE + 1]) - { - case PT_ANY: - OK = TRUE; - break; - - case PT_LAMP: + { + case PT_ANY: + OK = TRUE; + break; + + case PT_LAMP: OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; - break; - - case PT_GC: + break; + + case PT_GC: OK = PRIV(ucp_gentype)[prop->chartype] == code[1 + IMM2_SIZE + 2]; - break; - - case PT_PC: + break; + + case PT_PC: OK = prop->chartype == code[1 + IMM2_SIZE + 2]; - break; - - case PT_SC: + break; + + case PT_SC: OK = prop->script == code[1 + IMM2_SIZE + 2]; - break; - + break; + /* These are specials for combination cases. */ case PT_ALNUM: @@ -1927,357 +1927,357 @@ for (;;) c >= 0xe000; break; - /* Should never occur, but keep compilers from grumbling. */ - - default: - OK = codevalue != OP_PROP; - break; - } - - if (OK == (d == OP_PROP)) - { - if (codevalue == OP_PROP_EXTRA + OP_TYPEPOSUPTO) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + /* Should never occur, but keep compilers from grumbling. */ + + default: + OK = codevalue != OP_PROP; + break; + } + + if (OK == (d == OP_PROP)) + { + if (codevalue == OP_PROP_EXTRA + OP_TYPEPOSUPTO) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } if (++count >= (int)GET2(code, 1)) { ADD_NEW(state_offset + 1 + IMM2_SIZE + 3, 0); } - else - { ADD_NEW(state_offset, count); } - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_EXTUNI_EXTRA + OP_TYPEEXACT: - case OP_EXTUNI_EXTRA + OP_TYPEUPTO: - case OP_EXTUNI_EXTRA + OP_TYPEMINUPTO: - case OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO: - if (codevalue != OP_EXTUNI_EXTRA + OP_TYPEEXACT) + else + { ADD_NEW(state_offset, count); } + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_EXTUNI_EXTRA + OP_TYPEEXACT: + case OP_EXTUNI_EXTRA + OP_TYPEUPTO: + case OP_EXTUNI_EXTRA + OP_TYPEMINUPTO: + case OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO: + if (codevalue != OP_EXTUNI_EXTRA + OP_TYPEEXACT) { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); } - count = current_state->count; /* Number already matched */ + count = current_state->count; /* Number already matched */ if (clen > 0) - { + { int lgb, rgb; const pcre_uchar *nptr = ptr + clen; - int ncount = 0; - if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + int ncount = 0; + if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } lgb = UCD_GRAPHBREAK(c); - while (nptr < end_subject) - { + while (nptr < end_subject) + { dlen = 1; if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); } rgb = UCD_GRAPHBREAK(d); if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; - ncount++; + ncount++; lgb = rgb; nptr += dlen; - } + } if (nptr >= end_subject && (md->moptions & PCRE_PARTIAL_HARD) != 0) reset_could_continue = TRUE; if (++count >= (int)GET2(code, 1)) { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, ncount); } - else - { ADD_NEW_DATA(-state_offset, count, ncount); } - } - break; -#endif - - /*-----------------------------------------------------------------*/ - case OP_ANYNL_EXTRA + OP_TYPEEXACT: - case OP_ANYNL_EXTRA + OP_TYPEUPTO: - case OP_ANYNL_EXTRA + OP_TYPEMINUPTO: - case OP_ANYNL_EXTRA + OP_TYPEPOSUPTO: - if (codevalue != OP_ANYNL_EXTRA + OP_TYPEEXACT) + else + { ADD_NEW_DATA(-state_offset, count, ncount); } + } + break; +#endif + + /*-----------------------------------------------------------------*/ + case OP_ANYNL_EXTRA + OP_TYPEEXACT: + case OP_ANYNL_EXTRA + OP_TYPEUPTO: + case OP_ANYNL_EXTRA + OP_TYPEMINUPTO: + case OP_ANYNL_EXTRA + OP_TYPEPOSUPTO: + if (codevalue != OP_ANYNL_EXTRA + OP_TYPEEXACT) { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); } - count = current_state->count; /* Number already matched */ - if (clen > 0) - { - int ncount = 0; - switch (c) - { + count = current_state->count; /* Number already matched */ + if (clen > 0) + { + int ncount = 0; + switch (c) + { case CHAR_VT: case CHAR_FF: case CHAR_NEL: #ifndef EBCDIC - case 0x2028: - case 0x2029: + case 0x2028: + case 0x2029: #endif /* Not EBCDIC */ - if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; - goto ANYNL03; - + if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; + goto ANYNL03; + case CHAR_CR: if (ptr + 1 < end_subject && UCHAR21TEST(ptr + 1) == CHAR_LF) ncount = 1; - /* Fall through */ - - ANYNL03: + /* Fall through */ + + ANYNL03: case CHAR_LF: - if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } if (++count >= (int)GET2(code, 1)) { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, ncount); } - else - { ADD_NEW_DATA(-state_offset, count, ncount); } - break; - - default: - break; - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_VSPACE_EXTRA + OP_TYPEEXACT: - case OP_VSPACE_EXTRA + OP_TYPEUPTO: - case OP_VSPACE_EXTRA + OP_TYPEMINUPTO: - case OP_VSPACE_EXTRA + OP_TYPEPOSUPTO: - if (codevalue != OP_VSPACE_EXTRA + OP_TYPEEXACT) + else + { ADD_NEW_DATA(-state_offset, count, ncount); } + break; + + default: + break; + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_VSPACE_EXTRA + OP_TYPEEXACT: + case OP_VSPACE_EXTRA + OP_TYPEUPTO: + case OP_VSPACE_EXTRA + OP_TYPEMINUPTO: + case OP_VSPACE_EXTRA + OP_TYPEPOSUPTO: + if (codevalue != OP_VSPACE_EXTRA + OP_TYPEEXACT) { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); } - count = current_state->count; /* Number already matched */ - if (clen > 0) - { - BOOL OK; - switch (c) - { + count = current_state->count; /* Number already matched */ + if (clen > 0) + { + BOOL OK; + switch (c) + { VSPACE_CASES: - OK = TRUE; - break; - - default: - OK = FALSE; - } - - if (OK == (d == OP_VSPACE)) - { - if (codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSUPTO) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + OK = TRUE; + break; + + default: + OK = FALSE; + } + + if (OK == (d == OP_VSPACE)) + { + if (codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSUPTO) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } if (++count >= (int)GET2(code, 1)) { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, 0); } - else - { ADD_NEW_DATA(-state_offset, count, 0); } - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_HSPACE_EXTRA + OP_TYPEEXACT: - case OP_HSPACE_EXTRA + OP_TYPEUPTO: - case OP_HSPACE_EXTRA + OP_TYPEMINUPTO: - case OP_HSPACE_EXTRA + OP_TYPEPOSUPTO: - if (codevalue != OP_HSPACE_EXTRA + OP_TYPEEXACT) + else + { ADD_NEW_DATA(-state_offset, count, 0); } + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_HSPACE_EXTRA + OP_TYPEEXACT: + case OP_HSPACE_EXTRA + OP_TYPEUPTO: + case OP_HSPACE_EXTRA + OP_TYPEMINUPTO: + case OP_HSPACE_EXTRA + OP_TYPEPOSUPTO: + if (codevalue != OP_HSPACE_EXTRA + OP_TYPEEXACT) { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); } - count = current_state->count; /* Number already matched */ - if (clen > 0) - { - BOOL OK; - switch (c) - { + count = current_state->count; /* Number already matched */ + if (clen > 0) + { + BOOL OK; + switch (c) + { HSPACE_CASES: - OK = TRUE; - break; - - default: - OK = FALSE; - break; - } - - if (OK == (d == OP_HSPACE)) - { - if (codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSUPTO) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + OK = TRUE; + break; + + default: + OK = FALSE; + break; + } + + if (OK == (d == OP_HSPACE)) + { + if (codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSUPTO) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } if (++count >= (int)GET2(code, 1)) { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, 0); } - else - { ADD_NEW_DATA(-state_offset, count, 0); } - } - } - break; - -/* ========================================================================== */ - /* These opcodes are followed by a character that is usually compared - to the current subject character; it is loaded into d. We still get - here even if there is no subject character, because in some cases zero - repetitions are permitted. */ - - /*-----------------------------------------------------------------*/ - case OP_CHAR: - if (clen > 0 && c == d) { ADD_NEW(state_offset + dlen + 1, 0); } - break; - - /*-----------------------------------------------------------------*/ + else + { ADD_NEW_DATA(-state_offset, count, 0); } + } + } + break; + +/* ========================================================================== */ + /* These opcodes are followed by a character that is usually compared + to the current subject character; it is loaded into d. We still get + here even if there is no subject character, because in some cases zero + repetitions are permitted. */ + + /*-----------------------------------------------------------------*/ + case OP_CHAR: + if (clen > 0 && c == d) { ADD_NEW(state_offset + dlen + 1, 0); } + break; + + /*-----------------------------------------------------------------*/ case OP_CHARI: - if (clen == 0) break; - + if (clen == 0) break; + #ifdef SUPPORT_UTF if (utf) - { - if (c == d) { ADD_NEW(state_offset + dlen + 1, 0); } else - { - unsigned int othercase; + { + if (c == d) { ADD_NEW(state_offset + dlen + 1, 0); } else + { + unsigned int othercase; if (c < 128) othercase = fcc[c]; else /* If we have Unicode property support, we can use it to test the other case of the character. */ -#ifdef SUPPORT_UCP +#ifdef SUPPORT_UCP othercase = UCD_OTHERCASE(c); -#else +#else othercase = NOTACHAR; -#endif - - if (d == othercase) { ADD_NEW(state_offset + dlen + 1, 0); } - } - } - else +#endif + + if (d == othercase) { ADD_NEW(state_offset + dlen + 1, 0); } + } + } + else #endif /* SUPPORT_UTF */ /* Not UTF mode */ - { + { if (TABLE_GET(c, lcc, c) == TABLE_GET(d, lcc, d)) { ADD_NEW(state_offset + 2, 0); } - } - break; - - -#ifdef SUPPORT_UCP - /*-----------------------------------------------------------------*/ - /* This is a tricky one because it can match more than one character. - Find out how many characters to skip, and then set up a negative state - to wait for them to pass before continuing. */ - - case OP_EXTUNI: + } + break; + + +#ifdef SUPPORT_UCP + /*-----------------------------------------------------------------*/ + /* This is a tricky one because it can match more than one character. + Find out how many characters to skip, and then set up a negative state + to wait for them to pass before continuing. */ + + case OP_EXTUNI: if (clen > 0) - { + { int lgb, rgb; const pcre_uchar *nptr = ptr + clen; - int ncount = 0; + int ncount = 0; lgb = UCD_GRAPHBREAK(c); - while (nptr < end_subject) - { + while (nptr < end_subject) + { dlen = 1; if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); } rgb = UCD_GRAPHBREAK(d); if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; - ncount++; + ncount++; lgb = rgb; nptr += dlen; - } + } if (nptr >= end_subject && (md->moptions & PCRE_PARTIAL_HARD) != 0) reset_could_continue = TRUE; - ADD_NEW_DATA(-(state_offset + 1), 0, ncount); - } - break; -#endif - - /*-----------------------------------------------------------------*/ - /* This is a tricky like EXTUNI because it too can match more than one - character (when CR is followed by LF). In this case, set up a negative - state to wait for one character to pass before continuing. */ - - case OP_ANYNL: - if (clen > 0) switch(c) - { + ADD_NEW_DATA(-(state_offset + 1), 0, ncount); + } + break; +#endif + + /*-----------------------------------------------------------------*/ + /* This is a tricky like EXTUNI because it too can match more than one + character (when CR is followed by LF). In this case, set up a negative + state to wait for one character to pass before continuing. */ + + case OP_ANYNL: + if (clen > 0) switch(c) + { case CHAR_VT: case CHAR_FF: case CHAR_NEL: #ifndef EBCDIC - case 0x2028: - case 0x2029: + case 0x2028: + case 0x2029: #endif /* Not EBCDIC */ - if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; - + if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; + case CHAR_LF: - ADD_NEW(state_offset + 1, 0); - break; - + ADD_NEW(state_offset + 1, 0); + break; + case CHAR_CR: if (ptr + 1 >= end_subject) - { + { ADD_NEW(state_offset + 1, 0); if ((md->moptions & PCRE_PARTIAL_HARD) != 0) reset_could_continue = TRUE; } else if (UCHAR21TEST(ptr + 1) == CHAR_LF) { - ADD_NEW_DATA(-(state_offset + 1), 0, 1); - } - else - { - ADD_NEW(state_offset + 1, 0); - } - break; - } - break; - - /*-----------------------------------------------------------------*/ - case OP_NOT_VSPACE: - if (clen > 0) switch(c) - { + ADD_NEW_DATA(-(state_offset + 1), 0, 1); + } + else + { + ADD_NEW(state_offset + 1, 0); + } + break; + } + break; + + /*-----------------------------------------------------------------*/ + case OP_NOT_VSPACE: + if (clen > 0) switch(c) + { VSPACE_CASES: - break; - - default: - ADD_NEW(state_offset + 1, 0); - break; - } - break; - - /*-----------------------------------------------------------------*/ - case OP_VSPACE: - if (clen > 0) switch(c) - { + break; + + default: + ADD_NEW(state_offset + 1, 0); + break; + } + break; + + /*-----------------------------------------------------------------*/ + case OP_VSPACE: + if (clen > 0) switch(c) + { VSPACE_CASES: - ADD_NEW(state_offset + 1, 0); - break; - + ADD_NEW(state_offset + 1, 0); + break; + default: break; - } - break; - - /*-----------------------------------------------------------------*/ - case OP_NOT_HSPACE: - if (clen > 0) switch(c) - { + } + break; + + /*-----------------------------------------------------------------*/ + case OP_NOT_HSPACE: + if (clen > 0) switch(c) + { HSPACE_CASES: - break; - - default: - ADD_NEW(state_offset + 1, 0); - break; - } - break; - - /*-----------------------------------------------------------------*/ - case OP_HSPACE: - if (clen > 0) switch(c) - { + break; + + default: + ADD_NEW(state_offset + 1, 0); + break; + } + break; + + /*-----------------------------------------------------------------*/ + case OP_HSPACE: + if (clen > 0) switch(c) + { HSPACE_CASES: - ADD_NEW(state_offset + 1, 0); - break; + ADD_NEW(state_offset + 1, 0); + break; default: break; - } - break; - - /*-----------------------------------------------------------------*/ + } + break; + + /*-----------------------------------------------------------------*/ /* Match a negated single character casefully. */ - - case OP_NOT: + + case OP_NOT: if (clen > 0 && c != d) { ADD_NEW(state_offset + dlen + 1, 0); } break; @@ -2285,8 +2285,8 @@ for (;;) /* Match a negated single character caselessly. */ case OP_NOTI: - if (clen > 0) - { + if (clen > 0) + { pcre_uint32 otherd; #ifdef SUPPORT_UTF if (utf && d >= 128) @@ -2302,10 +2302,10 @@ for (;;) otherd = TABLE_GET(d, fcc, d); if (c != d && c != otherd) { ADD_NEW(state_offset + dlen + 1, 0); } - } - break; - - /*-----------------------------------------------------------------*/ + } + break; + + /*-----------------------------------------------------------------*/ case OP_PLUSI: case OP_MINPLUSI: case OP_POSPLUSI: @@ -2316,45 +2316,45 @@ for (;;) codevalue -= OP_STARI - OP_STAR; /* Fall through */ - case OP_PLUS: - case OP_MINPLUS: - case OP_POSPLUS: - case OP_NOTPLUS: - case OP_NOTMINPLUS: - case OP_NOTPOSPLUS: - count = current_state->count; /* Already matched */ - if (count > 0) { ADD_ACTIVE(state_offset + dlen + 1, 0); } - if (clen > 0) - { + case OP_PLUS: + case OP_MINPLUS: + case OP_POSPLUS: + case OP_NOTPLUS: + case OP_NOTMINPLUS: + case OP_NOTPOSPLUS: + count = current_state->count; /* Already matched */ + if (count > 0) { ADD_ACTIVE(state_offset + dlen + 1, 0); } + if (clen > 0) + { pcre_uint32 otherd = NOTACHAR; if (caseless) - { + { #ifdef SUPPORT_UTF if (utf && d >= 128) - { -#ifdef SUPPORT_UCP + { +#ifdef SUPPORT_UCP otherd = UCD_OTHERCASE(d); -#endif /* SUPPORT_UCP */ - } - else +#endif /* SUPPORT_UCP */ + } + else #endif /* SUPPORT_UTF */ otherd = TABLE_GET(d, fcc, d); - } - if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) - { - if (count > 0 && - (codevalue == OP_POSPLUS || codevalue == OP_NOTPOSPLUS)) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - count++; - ADD_NEW(state_offset, count); - } - } - break; - - /*-----------------------------------------------------------------*/ + } + if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) + { + if (count > 0 && + (codevalue == OP_POSPLUS || codevalue == OP_NOTPOSPLUS)) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + count++; + ADD_NEW(state_offset, count); + } + } + break; + + /*-----------------------------------------------------------------*/ case OP_QUERYI: case OP_MINQUERYI: case OP_POSQUERYI: @@ -2364,42 +2364,42 @@ for (;;) caseless = TRUE; codevalue -= OP_STARI - OP_STAR; /* Fall through */ - case OP_QUERY: - case OP_MINQUERY: - case OP_POSQUERY: - case OP_NOTQUERY: - case OP_NOTMINQUERY: - case OP_NOTPOSQUERY: - ADD_ACTIVE(state_offset + dlen + 1, 0); - if (clen > 0) - { + case OP_QUERY: + case OP_MINQUERY: + case OP_POSQUERY: + case OP_NOTQUERY: + case OP_NOTMINQUERY: + case OP_NOTPOSQUERY: + ADD_ACTIVE(state_offset + dlen + 1, 0); + if (clen > 0) + { pcre_uint32 otherd = NOTACHAR; if (caseless) - { + { #ifdef SUPPORT_UTF if (utf && d >= 128) - { -#ifdef SUPPORT_UCP + { +#ifdef SUPPORT_UCP otherd = UCD_OTHERCASE(d); -#endif /* SUPPORT_UCP */ - } - else +#endif /* SUPPORT_UCP */ + } + else #endif /* SUPPORT_UTF */ otherd = TABLE_GET(d, fcc, d); - } - if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) - { - if (codevalue == OP_POSQUERY || codevalue == OP_NOTPOSQUERY) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - ADD_NEW(state_offset + dlen + 1, 0); - } - } - break; - - /*-----------------------------------------------------------------*/ + } + if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) + { + if (codevalue == OP_POSQUERY || codevalue == OP_NOTPOSQUERY) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + ADD_NEW(state_offset + dlen + 1, 0); + } + } + break; + + /*-----------------------------------------------------------------*/ case OP_STARI: case OP_MINSTARI: case OP_POSSTARI: @@ -2409,77 +2409,77 @@ for (;;) caseless = TRUE; codevalue -= OP_STARI - OP_STAR; /* Fall through */ - case OP_STAR: - case OP_MINSTAR: - case OP_POSSTAR: - case OP_NOTSTAR: - case OP_NOTMINSTAR: - case OP_NOTPOSSTAR: - ADD_ACTIVE(state_offset + dlen + 1, 0); - if (clen > 0) - { + case OP_STAR: + case OP_MINSTAR: + case OP_POSSTAR: + case OP_NOTSTAR: + case OP_NOTMINSTAR: + case OP_NOTPOSSTAR: + ADD_ACTIVE(state_offset + dlen + 1, 0); + if (clen > 0) + { pcre_uint32 otherd = NOTACHAR; if (caseless) - { + { #ifdef SUPPORT_UTF if (utf && d >= 128) - { -#ifdef SUPPORT_UCP + { +#ifdef SUPPORT_UCP otherd = UCD_OTHERCASE(d); -#endif /* SUPPORT_UCP */ - } - else +#endif /* SUPPORT_UCP */ + } + else #endif /* SUPPORT_UTF */ otherd = TABLE_GET(d, fcc, d); - } - if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) - { - if (codevalue == OP_POSSTAR || codevalue == OP_NOTPOSSTAR) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } - ADD_NEW(state_offset, 0); - } - } - break; - - /*-----------------------------------------------------------------*/ + } + if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) + { + if (codevalue == OP_POSSTAR || codevalue == OP_NOTPOSSTAR) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } + ADD_NEW(state_offset, 0); + } + } + break; + + /*-----------------------------------------------------------------*/ case OP_EXACTI: case OP_NOTEXACTI: caseless = TRUE; codevalue -= OP_STARI - OP_STAR; /* Fall through */ - case OP_EXACT: - case OP_NOTEXACT: - count = current_state->count; /* Number already matched */ - if (clen > 0) - { + case OP_EXACT: + case OP_NOTEXACT: + count = current_state->count; /* Number already matched */ + if (clen > 0) + { pcre_uint32 otherd = NOTACHAR; if (caseless) - { + { #ifdef SUPPORT_UTF if (utf && d >= 128) - { -#ifdef SUPPORT_UCP + { +#ifdef SUPPORT_UCP otherd = UCD_OTHERCASE(d); -#endif /* SUPPORT_UCP */ - } - else +#endif /* SUPPORT_UCP */ + } + else #endif /* SUPPORT_UTF */ otherd = TABLE_GET(d, fcc, d); - } - if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) - { + } + if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) + { if (++count >= (int)GET2(code, 1)) { ADD_NEW(state_offset + dlen + 1 + IMM2_SIZE, 0); } - else - { ADD_NEW(state_offset, count); } - } - } - break; - - /*-----------------------------------------------------------------*/ + else + { ADD_NEW(state_offset, count); } + } + } + break; + + /*-----------------------------------------------------------------*/ case OP_UPTOI: case OP_MINUPTOI: case OP_POSUPTOI: @@ -2489,92 +2489,92 @@ for (;;) caseless = TRUE; codevalue -= OP_STARI - OP_STAR; /* Fall through */ - case OP_UPTO: - case OP_MINUPTO: - case OP_POSUPTO: - case OP_NOTUPTO: - case OP_NOTMINUPTO: - case OP_NOTPOSUPTO: + case OP_UPTO: + case OP_MINUPTO: + case OP_POSUPTO: + case OP_NOTUPTO: + case OP_NOTMINUPTO: + case OP_NOTPOSUPTO: ADD_ACTIVE(state_offset + dlen + 1 + IMM2_SIZE, 0); - count = current_state->count; /* Number already matched */ - if (clen > 0) - { + count = current_state->count; /* Number already matched */ + if (clen > 0) + { pcre_uint32 otherd = NOTACHAR; if (caseless) - { + { #ifdef SUPPORT_UTF if (utf && d >= 128) - { -#ifdef SUPPORT_UCP + { +#ifdef SUPPORT_UCP otherd = UCD_OTHERCASE(d); -#endif /* SUPPORT_UCP */ - } - else +#endif /* SUPPORT_UCP */ + } + else #endif /* SUPPORT_UTF */ otherd = TABLE_GET(d, fcc, d); - } - if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) - { - if (codevalue == OP_POSUPTO || codevalue == OP_NOTPOSUPTO) - { - active_count--; /* Remove non-match possibility */ - next_active_state--; - } + } + if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) + { + if (codevalue == OP_POSUPTO || codevalue == OP_NOTPOSUPTO) + { + active_count--; /* Remove non-match possibility */ + next_active_state--; + } if (++count >= (int)GET2(code, 1)) { ADD_NEW(state_offset + dlen + 1 + IMM2_SIZE, 0); } - else - { ADD_NEW(state_offset, count); } - } - } - break; - - -/* ========================================================================== */ - /* These are the class-handling opcodes */ - - case OP_CLASS: - case OP_NCLASS: - case OP_XCLASS: - { - BOOL isinclass = FALSE; - int next_state_offset; + else + { ADD_NEW(state_offset, count); } + } + } + break; + + +/* ========================================================================== */ + /* These are the class-handling opcodes */ + + case OP_CLASS: + case OP_NCLASS: + case OP_XCLASS: + { + BOOL isinclass = FALSE; + int next_state_offset; const pcre_uchar *ecode; - - /* For a simple class, there is always just a 32-byte table, and we - can set isinclass from it. */ - - if (codevalue != OP_XCLASS) - { + + /* For a simple class, there is always just a 32-byte table, and we + can set isinclass from it. */ + + if (codevalue != OP_XCLASS) + { ecode = code + 1 + (32 / sizeof(pcre_uchar)); - if (clen > 0) - { - isinclass = (c > 255)? (codevalue == OP_NCLASS) : + if (clen > 0) + { + isinclass = (c > 255)? (codevalue == OP_NCLASS) : ((((pcre_uint8 *)(code + 1))[c/8] & (1 << (c&7))) != 0); - } - } - - /* An extended class may have a table or a list of single characters, - ranges, or both, and it may be positive or negative. There's a - function that sorts all this out. */ - - else - { - ecode = code + GET(code, 1); + } + } + + /* An extended class may have a table or a list of single characters, + ranges, or both, and it may be positive or negative. There's a + function that sorts all this out. */ + + else + { + ecode = code + GET(code, 1); if (clen > 0) isinclass = PRIV(xclass)(c, code + 1 + LINK_SIZE, utf); - } - - /* At this point, isinclass is set for all kinds of class, and ecode - points to the byte after the end of the class. If there is a - quantifier, this is where it will be. */ - + } + + /* At this point, isinclass is set for all kinds of class, and ecode + points to the byte after the end of the class. If there is a + quantifier, this is where it will be. */ + next_state_offset = (int)(ecode - start_code); - - switch (*ecode) - { - case OP_CRSTAR: - case OP_CRMINSTAR: + + switch (*ecode) + { + case OP_CRSTAR: + case OP_CRMINSTAR: case OP_CRPOSSTAR: - ADD_ACTIVE(next_state_offset + 1, 0); + ADD_ACTIVE(next_state_offset + 1, 0); if (isinclass) { if (*ecode == OP_CRPOSSTAR) @@ -2584,13 +2584,13 @@ for (;;) } ADD_NEW(state_offset, 0); } - break; - - case OP_CRPLUS: - case OP_CRMINPLUS: + break; + + case OP_CRPLUS: + case OP_CRMINPLUS: case OP_CRPOSPLUS: - count = current_state->count; /* Already matched */ - if (count > 0) { ADD_ACTIVE(next_state_offset + 1, 0); } + count = current_state->count; /* Already matched */ + if (count > 0) { ADD_ACTIVE(next_state_offset + 1, 0); } if (isinclass) { if (count > 0 && *ecode == OP_CRPOSPLUS) @@ -2601,12 +2601,12 @@ for (;;) count++; ADD_NEW(state_offset, count); } - break; - - case OP_CRQUERY: - case OP_CRMINQUERY: + break; + + case OP_CRQUERY: + case OP_CRMINQUERY: case OP_CRPOSQUERY: - ADD_ACTIVE(next_state_offset + 1, 0); + ADD_ACTIVE(next_state_offset + 1, 0); if (isinclass) { if (*ecode == OP_CRPOSQUERY) @@ -2616,90 +2616,90 @@ for (;;) } ADD_NEW(next_state_offset + 1, 0); } - break; - - case OP_CRRANGE: - case OP_CRMINRANGE: + break; + + case OP_CRRANGE: + case OP_CRMINRANGE: case OP_CRPOSRANGE: - count = current_state->count; /* Already matched */ + count = current_state->count; /* Already matched */ if (count >= (int)GET2(ecode, 1)) { ADD_ACTIVE(next_state_offset + 1 + 2 * IMM2_SIZE, 0); } - if (isinclass) - { + if (isinclass) + { int max = (int)GET2(ecode, 1 + IMM2_SIZE); if (*ecode == OP_CRPOSRANGE && count >= (int)GET2(ecode, 1)) { active_count--; /* Remove non-match possibility */ next_active_state--; } - if (++count >= max && max != 0) /* Max 0 => no limit */ + if (++count >= max && max != 0) /* Max 0 => no limit */ { ADD_NEW(next_state_offset + 1 + 2 * IMM2_SIZE, 0); } - else - { ADD_NEW(state_offset, count); } - } - break; - - default: - if (isinclass) { ADD_NEW(next_state_offset, 0); } - break; - } - } - break; - -/* ========================================================================== */ - /* These are the opcodes for fancy brackets of various kinds. We have + else + { ADD_NEW(state_offset, count); } + } + break; + + default: + if (isinclass) { ADD_NEW(next_state_offset, 0); } + break; + } + } + break; + +/* ========================================================================== */ + /* These are the opcodes for fancy brackets of various kinds. We have to use recursion in order to handle them. The "always failing" assertion (?!) is optimised to OP_FAIL when compiling, so we have to support that, though the other "backtracking verbs" are not supported. */ - + case OP_FAIL: forced_fail++; /* Count FAILs for multiple states */ break; - case OP_ASSERT: - case OP_ASSERT_NOT: - case OP_ASSERTBACK: - case OP_ASSERTBACK_NOT: - { - int rc; - int local_offsets[2]; - int local_workspace[1000]; + case OP_ASSERT: + case OP_ASSERT_NOT: + case OP_ASSERTBACK: + case OP_ASSERTBACK_NOT: + { + int rc; + int local_offsets[2]; + int local_workspace[1000]; const pcre_uchar *endasscode = code + GET(code, 1); - - while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1); - - rc = internal_dfa_exec( - md, /* static match data */ - code, /* this subexpression's code */ - ptr, /* where we currently are */ + + while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1); + + rc = internal_dfa_exec( + md, /* static match data */ + code, /* this subexpression's code */ + ptr, /* where we currently are */ (int)(ptr - start_subject), /* start offset */ - local_offsets, /* offset vector */ - sizeof(local_offsets)/sizeof(int), /* size of same */ - local_workspace, /* workspace vector */ - sizeof(local_workspace)/sizeof(int), /* size of same */ + local_offsets, /* offset vector */ + sizeof(local_offsets)/sizeof(int), /* size of same */ + local_workspace, /* workspace vector */ + sizeof(local_workspace)/sizeof(int), /* size of same */ rlevel); /* function recursion level */ - + if (rc == PCRE_ERROR_DFA_UITEM) return rc; - if ((rc >= 0) == (codevalue == OP_ASSERT || codevalue == OP_ASSERTBACK)) + if ((rc >= 0) == (codevalue == OP_ASSERT || codevalue == OP_ASSERTBACK)) { ADD_ACTIVE((int)(endasscode + LINK_SIZE + 1 - start_code), 0); } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_COND: - case OP_SCOND: - { - int local_offsets[1000]; - int local_workspace[1000]; + } + break; + + /*-----------------------------------------------------------------*/ + case OP_COND: + case OP_SCOND: + { + int local_offsets[1000]; + int local_workspace[1000]; int codelink = GET(code, 1); int condcode; - + /* Because of the way auto-callout works during compile, a callout item is inserted between OP_COND and an assertion condition. This does not happen for the other conditions. */ - + if (code[LINK_SIZE+1] == OP_CALLOUT) - { + { rrc = 0; if (PUBL(callout) != NULL) { @@ -2727,8 +2727,8 @@ for (;;) } if (rrc > 0) break; /* Fail this thread */ code += PRIV(OP_lengths)[OP_CALLOUT]; /* Skip callout data */ - } - + } + condcode = code[LINK_SIZE+1]; /* Back reference conditions and duplicate named recursion conditions @@ -2744,63 +2744,63 @@ for (;;) if (condcode == OP_DEF || condcode == OP_FAIL) { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); } - /* The only supported version of OP_RREF is for the value RREF_ANY, - which means "test if in any recursion". We can't test for specifically - recursed groups. */ - - else if (condcode == OP_RREF) - { + /* The only supported version of OP_RREF is for the value RREF_ANY, + which means "test if in any recursion". We can't test for specifically + recursed groups. */ + + else if (condcode == OP_RREF) + { int value = GET2(code, LINK_SIZE + 2); - if (value != RREF_ANY) return PCRE_ERROR_DFA_UCOND; + if (value != RREF_ANY) return PCRE_ERROR_DFA_UCOND; if (md->recursive != NULL) { ADD_ACTIVE(state_offset + LINK_SIZE + 2 + IMM2_SIZE, 0); } else { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); } - } - - /* Otherwise, the condition is an assertion */ - - else - { - int rc; + } + + /* Otherwise, the condition is an assertion */ + + else + { + int rc; const pcre_uchar *asscode = code + LINK_SIZE + 1; const pcre_uchar *endasscode = asscode + GET(asscode, 1); - - while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1); - - rc = internal_dfa_exec( - md, /* fixed match data */ - asscode, /* this subexpression's code */ - ptr, /* where we currently are */ + + while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1); + + rc = internal_dfa_exec( + md, /* fixed match data */ + asscode, /* this subexpression's code */ + ptr, /* where we currently are */ (int)(ptr - start_subject), /* start offset */ - local_offsets, /* offset vector */ - sizeof(local_offsets)/sizeof(int), /* size of same */ - local_workspace, /* workspace vector */ - sizeof(local_workspace)/sizeof(int), /* size of same */ + local_offsets, /* offset vector */ + sizeof(local_offsets)/sizeof(int), /* size of same */ + local_workspace, /* workspace vector */ + sizeof(local_workspace)/sizeof(int), /* size of same */ rlevel); /* function recursion level */ - + if (rc == PCRE_ERROR_DFA_UITEM) return rc; - if ((rc >= 0) == - (condcode == OP_ASSERT || condcode == OP_ASSERTBACK)) + if ((rc >= 0) == + (condcode == OP_ASSERT || condcode == OP_ASSERTBACK)) { ADD_ACTIVE((int)(endasscode + LINK_SIZE + 1 - start_code), 0); } - else + else { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); } - } - } - break; - - /*-----------------------------------------------------------------*/ - case OP_RECURSE: - { + } + } + break; + + /*-----------------------------------------------------------------*/ + case OP_RECURSE: + { dfa_recursion_info *ri; - int local_offsets[1000]; - int local_workspace[1000]; + int local_offsets[1000]; + int local_workspace[1000]; const pcre_uchar *callpat = start_code + GET(code, 1); int recno = (callpat == md->start_code)? 0 : GET2(callpat, 1 + LINK_SIZE); - int rc; - + int rc; + DPRINTF(("%.*sStarting regex recursion\n", rlevel*2-2, SP)); - + /* Check for repeating a recursion without advancing the subject pointer. This should catch convoluted mutual recursions. (Some simple cases are caught at compile time.) */ @@ -2817,35 +2817,35 @@ for (;;) new_recursive.prevrec = md->recursive; md->recursive = &new_recursive; - rc = internal_dfa_exec( - md, /* fixed match data */ + rc = internal_dfa_exec( + md, /* fixed match data */ callpat, /* this subexpression's code */ - ptr, /* where we currently are */ + ptr, /* where we currently are */ (int)(ptr - start_subject), /* start offset */ - local_offsets, /* offset vector */ - sizeof(local_offsets)/sizeof(int), /* size of same */ - local_workspace, /* workspace vector */ - sizeof(local_workspace)/sizeof(int), /* size of same */ + local_offsets, /* offset vector */ + sizeof(local_offsets)/sizeof(int), /* size of same */ + local_workspace, /* workspace vector */ + sizeof(local_workspace)/sizeof(int), /* size of same */ rlevel); /* function recursion level */ - + md->recursive = new_recursive.prevrec; /* Done this recursion */ - + DPRINTF(("%.*sReturn from regex recursion: rc=%d\n", rlevel*2-2, SP, rc)); - /* Ran out of internal offsets */ - - if (rc == 0) return PCRE_ERROR_DFA_RECURSE; - - /* For each successful matched substring, set up the next state with a - count of characters to skip before trying it. Note that the count is in - characters, not bytes. */ - - if (rc > 0) - { - for (rc = rc*2 - 2; rc >= 0; rc -= 2) - { - int charcount = local_offsets[rc+1] - local_offsets[rc]; + /* Ran out of internal offsets */ + + if (rc == 0) return PCRE_ERROR_DFA_RECURSE; + + /* For each successful matched substring, set up the next state with a + count of characters to skip before trying it. Note that the count is in + characters, not bytes. */ + + if (rc > 0) + { + for (rc = rc*2 - 2; rc >= 0; rc -= 2) + { + int charcount = local_offsets[rc+1] - local_offsets[rc]; #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf) { @@ -2854,21 +2854,21 @@ for (;;) while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--; } #endif - if (charcount > 0) - { - ADD_NEW_DATA(-(state_offset + LINK_SIZE + 1), 0, (charcount - 1)); - } - else - { - ADD_ACTIVE(state_offset + LINK_SIZE + 1, 0); - } - } - } - else if (rc != PCRE_ERROR_NOMATCH) return rc; - } - break; - - /*-----------------------------------------------------------------*/ + if (charcount > 0) + { + ADD_NEW_DATA(-(state_offset + LINK_SIZE + 1), 0, (charcount - 1)); + } + else + { + ADD_ACTIVE(state_offset + LINK_SIZE + 1, 0); + } + } + } + else if (rc != PCRE_ERROR_NOMATCH) return rc; + } + break; + + /*-----------------------------------------------------------------*/ case OP_BRAPOS: case OP_SBRAPOS: case OP_CBRAPOS: @@ -2960,78 +2960,78 @@ for (;;) break; /*-----------------------------------------------------------------*/ - case OP_ONCE: + case OP_ONCE: case OP_ONCE_NC: - { - int local_offsets[2]; - int local_workspace[1000]; - - int rc = internal_dfa_exec( - md, /* fixed match data */ - code, /* this subexpression's code */ - ptr, /* where we currently are */ + { + int local_offsets[2]; + int local_workspace[1000]; + + int rc = internal_dfa_exec( + md, /* fixed match data */ + code, /* this subexpression's code */ + ptr, /* where we currently are */ (int)(ptr - start_subject), /* start offset */ - local_offsets, /* offset vector */ - sizeof(local_offsets)/sizeof(int), /* size of same */ - local_workspace, /* workspace vector */ - sizeof(local_workspace)/sizeof(int), /* size of same */ + local_offsets, /* offset vector */ + sizeof(local_offsets)/sizeof(int), /* size of same */ + local_workspace, /* workspace vector */ + sizeof(local_workspace)/sizeof(int), /* size of same */ rlevel); /* function recursion level */ - - if (rc >= 0) - { + + if (rc >= 0) + { const pcre_uchar *end_subpattern = code; - int charcount = local_offsets[1] - local_offsets[0]; - int next_state_offset, repeat_state_offset; - - do { end_subpattern += GET(end_subpattern, 1); } - while (*end_subpattern == OP_ALT); + int charcount = local_offsets[1] - local_offsets[0]; + int next_state_offset, repeat_state_offset; + + do { end_subpattern += GET(end_subpattern, 1); } + while (*end_subpattern == OP_ALT); next_state_offset = (int)(end_subpattern - start_code + LINK_SIZE + 1); - - /* If the end of this subpattern is KETRMAX or KETRMIN, we must - arrange for the repeat state also to be added to the relevant list. - Calculate the offset, or set -1 for no repeat. */ - - repeat_state_offset = (*end_subpattern == OP_KETRMAX || - *end_subpattern == OP_KETRMIN)? + + /* If the end of this subpattern is KETRMAX or KETRMIN, we must + arrange for the repeat state also to be added to the relevant list. + Calculate the offset, or set -1 for no repeat. */ + + repeat_state_offset = (*end_subpattern == OP_KETRMAX || + *end_subpattern == OP_KETRMIN)? (int)(end_subpattern - start_code - GET(end_subpattern, 1)) : -1; - - /* If we have matched an empty string, add the next state at the - current character pointer. This is important so that the duplicate - checking kicks in, which is what breaks infinite loops that match an - empty string. */ - - if (charcount == 0) - { - ADD_ACTIVE(next_state_offset, 0); - } - - /* Optimization: if there are no more active states, and there - are no new states yet set up, then skip over the subject string - right here, to save looping. Otherwise, set up the new state to swing + + /* If we have matched an empty string, add the next state at the + current character pointer. This is important so that the duplicate + checking kicks in, which is what breaks infinite loops that match an + empty string. */ + + if (charcount == 0) + { + ADD_ACTIVE(next_state_offset, 0); + } + + /* Optimization: if there are no more active states, and there + are no new states yet set up, then skip over the subject string + right here, to save looping. Otherwise, set up the new state to swing into action when the end of the matched substring is reached. */ - - else if (i + 1 >= active_count && new_count == 0) - { - ptr += charcount; - clen = 0; - ADD_NEW(next_state_offset, 0); - - /* If we are adding a repeat state at the new character position, - we must fudge things so that it is the only current state. - Otherwise, it might be a duplicate of one we processed before, and - that would cause it to be skipped. */ - - if (repeat_state_offset >= 0) - { - next_active_state = active_states; - active_count = 0; - i = -1; - ADD_ACTIVE(repeat_state_offset, 0); - } - } - else - { + + else if (i + 1 >= active_count && new_count == 0) + { + ptr += charcount; + clen = 0; + ADD_NEW(next_state_offset, 0); + + /* If we are adding a repeat state at the new character position, + we must fudge things so that it is the only current state. + Otherwise, it might be a duplicate of one we processed before, and + that would cause it to be skipped. */ + + if (repeat_state_offset >= 0) + { + next_active_state = active_states; + active_count = 0; + i = -1; + ADD_ACTIVE(repeat_state_offset, 0); + } + } + else + { #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 if (utf) { @@ -3040,29 +3040,29 @@ for (;;) while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--; } #endif - ADD_NEW_DATA(-next_state_offset, 0, (charcount - 1)); - if (repeat_state_offset >= 0) - { ADD_NEW_DATA(-repeat_state_offset, 0, (charcount - 1)); } - } - } - else if (rc != PCRE_ERROR_NOMATCH) return rc; - } - break; - - -/* ========================================================================== */ - /* Handle callouts */ - - case OP_CALLOUT: + ADD_NEW_DATA(-next_state_offset, 0, (charcount - 1)); + if (repeat_state_offset >= 0) + { ADD_NEW_DATA(-repeat_state_offset, 0, (charcount - 1)); } + } + } + else if (rc != PCRE_ERROR_NOMATCH) return rc; + } + break; + + +/* ========================================================================== */ + /* Handle callouts */ + + case OP_CALLOUT: rrc = 0; if (PUBL(callout) != NULL) - { + { PUBL(callout_block) cb; - cb.version = 1; /* Version 1 of the callout block */ - cb.callout_number = code[1]; - cb.offset_vector = offsets; + cb.version = 1; /* Version 1 of the callout block */ + cb.callout_number = code[1]; + cb.offset_vector = offsets; #if defined COMPILE_PCRE8 - cb.subject = (PCRE_SPTR)start_subject; + cb.subject = (PCRE_SPTR)start_subject; #elif defined COMPILE_PCRE16 cb.subject = (PCRE_SPTR16)start_subject; #elif defined COMPILE_PCRE32 @@ -3071,33 +3071,33 @@ for (;;) cb.subject_length = (int)(end_subject - start_subject); cb.start_match = (int)(current_subject - start_subject); cb.current_position = (int)(ptr - start_subject); - cb.pattern_position = GET(code, 2); - cb.next_item_length = GET(code, 2 + LINK_SIZE); - cb.capture_top = 1; - cb.capture_last = -1; - cb.callout_data = md->callout_data; + cb.pattern_position = GET(code, 2); + cb.next_item_length = GET(code, 2 + LINK_SIZE); + cb.capture_top = 1; + cb.capture_last = -1; + cb.callout_data = md->callout_data; cb.mark = NULL; /* No (*MARK) support */ if ((rrc = (*PUBL(callout))(&cb)) < 0) return rrc; /* Abandon */ - } + } if (rrc == 0) { ADD_ACTIVE(state_offset + PRIV(OP_lengths)[OP_CALLOUT], 0); } - break; - - -/* ========================================================================== */ - default: /* Unsupported opcode */ - return PCRE_ERROR_DFA_UITEM; - } - - NEXT_ACTIVE_STATE: continue; - - } /* End of loop scanning active states */ - - /* We have finished the processing at the current subject character. If no - new states have been set for the next character, we have found all the - matches that we are going to find. If we are at the top level and partial + break; + + +/* ========================================================================== */ + default: /* Unsupported opcode */ + return PCRE_ERROR_DFA_UITEM; + } + + NEXT_ACTIVE_STATE: continue; + + } /* End of loop scanning active states */ + + /* We have finished the processing at the current subject character. If no + new states have been set for the next character, we have found all the + matches that we are going to find. If we are at the top level and partial matching has been requested, check for appropriate conditions. - + The "forced_ fail" variable counts the number of (*F) encountered for the character. If it is equal to the original active_count (saved in workspace[1]) it means that (*F) was found on every active state. In this @@ -3106,8 +3106,8 @@ for (;;) The "could_continue" variable is true if a state could have continued but for the fact that the end of the subject was reached. */ - if (new_count <= 0) - { + if (new_count <= 0) + { if (rlevel == 1 && /* Top level, and */ could_continue && /* Some could go on, and */ forced_fail != workspace[1] && /* Not all forced fail & */ @@ -3124,61 +3124,61 @@ for (;;) ptr > md->start_used_ptr) /* Inspected non-empty string */ ) ) - match_count = PCRE_ERROR_PARTIAL; - DPRINTF(("%.*sEnd of internal_dfa_exec %d: returning %d\n" - "%.*s---------------------\n\n", rlevel*2-2, SP, rlevel, match_count, - rlevel*2-2, SP)); - break; /* In effect, "return", but see the comment below */ - } - - /* One or more states are active for the next character. */ - - ptr += clen; /* Advance to next subject character */ - } /* Loop to move along the subject string */ - -/* Control gets here from "break" a few lines above. We do it this way because -if we use "return" above, we have compiler trouble. Some compilers warn if -there's nothing here because they think the function doesn't return a value. On -the other hand, if we put a dummy statement here, some more clever compilers -complain that it can't be reached. Sigh. */ - -return match_count; -} - - - - -/************************************************* -* Execute a Regular Expression - DFA engine * -*************************************************/ - -/* This external function applies a compiled re to a subject string using a DFA -engine. This function calls the internal function multiple times if the pattern -is not anchored. - -Arguments: - argument_re points to the compiled expression - extra_data points to extra data or is NULL - subject points to the subject string - length length of subject string (may contain binary zeros) - start_offset where to start in the subject string - options option bits - offsets vector of match offsets - offsetcount size of same - workspace workspace vector - wscount size of same - -Returns: > 0 => number of match offset pairs placed in offsets - = 0 => offsets overflowed; longest matches are present - -1 => failed to match - < -1 => some kind of unexpected problem -*/ - + match_count = PCRE_ERROR_PARTIAL; + DPRINTF(("%.*sEnd of internal_dfa_exec %d: returning %d\n" + "%.*s---------------------\n\n", rlevel*2-2, SP, rlevel, match_count, + rlevel*2-2, SP)); + break; /* In effect, "return", but see the comment below */ + } + + /* One or more states are active for the next character. */ + + ptr += clen; /* Advance to next subject character */ + } /* Loop to move along the subject string */ + +/* Control gets here from "break" a few lines above. We do it this way because +if we use "return" above, we have compiler trouble. Some compilers warn if +there's nothing here because they think the function doesn't return a value. On +the other hand, if we put a dummy statement here, some more clever compilers +complain that it can't be reached. Sigh. */ + +return match_count; +} + + + + +/************************************************* +* Execute a Regular Expression - DFA engine * +*************************************************/ + +/* This external function applies a compiled re to a subject string using a DFA +engine. This function calls the internal function multiple times if the pattern +is not anchored. + +Arguments: + argument_re points to the compiled expression + extra_data points to extra data or is NULL + subject points to the subject string + length length of subject string (may contain binary zeros) + start_offset where to start in the subject string + options option bits + offsets vector of match offsets + offsetcount size of same + workspace workspace vector + wscount size of same + +Returns: > 0 => number of match offset pairs placed in offsets + = 0 => offsets overflowed; longest matches are present + -1 => failed to match + < -1 => some kind of unexpected problem +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION -pcre_dfa_exec(const pcre *argument_re, const pcre_extra *extra_data, - const char *subject, int length, int start_offset, int options, int *offsets, - int offsetcount, int *workspace, int wscount) +pcre_dfa_exec(const pcre *argument_re, const pcre_extra *extra_data, + const char *subject, int length, int start_offset, int options, int *offsets, + int offsetcount, int *workspace, int wscount) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre16_dfa_exec(const pcre16 *argument_re, const pcre16_extra *extra_data, @@ -3190,14 +3190,14 @@ pcre32_dfa_exec(const pcre32 *argument_re, const pcre32_extra *extra_data, PCRE_SPTR32 subject, int length, int start_offset, int options, int *offsets, int offsetcount, int *workspace, int wscount) #endif -{ +{ REAL_PCRE *re = (REAL_PCRE *)argument_re; -dfa_match_data match_block; -dfa_match_data *md = &match_block; +dfa_match_data match_block; +dfa_match_data *md = &match_block; BOOL utf, anchored, startline, firstline; const pcre_uchar *current_subject, *end_subject; -const pcre_study_data *study = NULL; - +const pcre_study_data *study = NULL; + const pcre_uchar *req_char_ptr; const pcre_uint8 *start_bits = NULL; BOOL has_first_char = FALSE; @@ -3206,23 +3206,23 @@ pcre_uchar first_char = 0; pcre_uchar first_char2 = 0; pcre_uchar req_char = 0; pcre_uchar req_char2 = 0; -int newline; - -/* Plausibility checks */ - -if ((options & ~PUBLIC_DFA_EXEC_OPTIONS) != 0) return PCRE_ERROR_BADOPTION; -if (re == NULL || subject == NULL || workspace == NULL || - (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL; -if (offsetcount < 0) return PCRE_ERROR_BADCOUNT; -if (wscount < 20) return PCRE_ERROR_DFA_WSSIZE; +int newline; + +/* Plausibility checks */ + +if ((options & ~PUBLIC_DFA_EXEC_OPTIONS) != 0) return PCRE_ERROR_BADOPTION; +if (re == NULL || subject == NULL || workspace == NULL || + (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL; +if (offsetcount < 0) return PCRE_ERROR_BADCOUNT; +if (wscount < 20) return PCRE_ERROR_DFA_WSSIZE; if (length < 0) return PCRE_ERROR_BADLENGTH; if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET; - + /* Check that the first field in the block is the magic number. If it is not, return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which means that the pattern is likely compiled with different endianness. */ - + if (re->magic_number != MAGIC_NUMBER) return re->magic_number == REVERSED_MAGIC_NUMBER? PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC; @@ -3240,116 +3240,116 @@ if ((options & PCRE_DFA_RESTART) != 0) /* Set up study, callout, and table data */ -md->tables = re->tables; -md->callout_data = NULL; - -if (extra_data != NULL) - { +md->tables = re->tables; +md->callout_data = NULL; + +if (extra_data != NULL) + { unsigned long int flags = extra_data->flags; - if ((flags & PCRE_EXTRA_STUDY_DATA) != 0) - study = (const pcre_study_data *)extra_data->study_data; - if ((flags & PCRE_EXTRA_MATCH_LIMIT) != 0) return PCRE_ERROR_DFA_UMLIMIT; - if ((flags & PCRE_EXTRA_MATCH_LIMIT_RECURSION) != 0) - return PCRE_ERROR_DFA_UMLIMIT; - if ((flags & PCRE_EXTRA_CALLOUT_DATA) != 0) - md->callout_data = extra_data->callout_data; - if ((flags & PCRE_EXTRA_TABLES) != 0) - md->tables = extra_data->tables; - } - -/* Set some local values */ - + if ((flags & PCRE_EXTRA_STUDY_DATA) != 0) + study = (const pcre_study_data *)extra_data->study_data; + if ((flags & PCRE_EXTRA_MATCH_LIMIT) != 0) return PCRE_ERROR_DFA_UMLIMIT; + if ((flags & PCRE_EXTRA_MATCH_LIMIT_RECURSION) != 0) + return PCRE_ERROR_DFA_UMLIMIT; + if ((flags & PCRE_EXTRA_CALLOUT_DATA) != 0) + md->callout_data = extra_data->callout_data; + if ((flags & PCRE_EXTRA_TABLES) != 0) + md->tables = extra_data->tables; + } + +/* Set some local values */ + current_subject = (const pcre_uchar *)subject + start_offset; end_subject = (const pcre_uchar *)subject + length; req_char_ptr = current_subject - 1; - + #ifdef SUPPORT_UTF /* PCRE_UTF(16|32) have the same value as PCRE_UTF8. */ utf = (re->options & PCRE_UTF8) != 0; -#else +#else utf = FALSE; -#endif - -anchored = (options & (PCRE_ANCHORED|PCRE_DFA_RESTART)) != 0 || - (re->options & PCRE_ANCHORED) != 0; - -/* The remaining fixed data for passing around. */ - +#endif + +anchored = (options & (PCRE_ANCHORED|PCRE_DFA_RESTART)) != 0 || + (re->options & PCRE_ANCHORED) != 0; + +/* The remaining fixed data for passing around. */ + md->start_code = (const pcre_uchar *)argument_re + - re->name_table_offset + re->name_count * re->name_entry_size; + re->name_table_offset + re->name_count * re->name_entry_size; md->start_subject = (const pcre_uchar *)subject; -md->end_subject = end_subject; +md->end_subject = end_subject; md->start_offset = start_offset; -md->moptions = options; -md->poptions = re->options; - -/* If the BSR option is not set at match time, copy what was set -at compile time. */ - -if ((md->moptions & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) == 0) - { - if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0) - md->moptions |= re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE); -#ifdef BSR_ANYCRLF - else md->moptions |= PCRE_BSR_ANYCRLF; -#endif - } - -/* Handle different types of newline. The three bits give eight cases. If -nothing is set at run time, whatever was used at compile time applies. */ - -switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) & - PCRE_NEWLINE_BITS) - { - case 0: newline = NEWLINE; break; /* Compile-time default */ +md->moptions = options; +md->poptions = re->options; + +/* If the BSR option is not set at match time, copy what was set +at compile time. */ + +if ((md->moptions & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) == 0) + { + if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0) + md->moptions |= re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE); +#ifdef BSR_ANYCRLF + else md->moptions |= PCRE_BSR_ANYCRLF; +#endif + } + +/* Handle different types of newline. The three bits give eight cases. If +nothing is set at run time, whatever was used at compile time applies. */ + +switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) & + PCRE_NEWLINE_BITS) + { + case 0: newline = NEWLINE; break; /* Compile-time default */ case PCRE_NEWLINE_CR: newline = CHAR_CR; break; case PCRE_NEWLINE_LF: newline = CHAR_NL; break; - case PCRE_NEWLINE_CR+ + case PCRE_NEWLINE_CR+ PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break; - case PCRE_NEWLINE_ANY: newline = -1; break; - case PCRE_NEWLINE_ANYCRLF: newline = -2; break; - default: return PCRE_ERROR_BADNEWLINE; - } - -if (newline == -2) - { - md->nltype = NLTYPE_ANYCRLF; - } -else if (newline < 0) - { - md->nltype = NLTYPE_ANY; - } -else - { - md->nltype = NLTYPE_FIXED; - if (newline > 255) - { - md->nllen = 2; - md->nl[0] = (newline >> 8) & 255; - md->nl[1] = newline & 255; - } - else - { - md->nllen = 1; - md->nl[0] = newline; - } - } - -/* Check a UTF-8 string if required. Unfortunately there's no way of passing -back the character offset. */ - + case PCRE_NEWLINE_ANY: newline = -1; break; + case PCRE_NEWLINE_ANYCRLF: newline = -2; break; + default: return PCRE_ERROR_BADNEWLINE; + } + +if (newline == -2) + { + md->nltype = NLTYPE_ANYCRLF; + } +else if (newline < 0) + { + md->nltype = NLTYPE_ANY; + } +else + { + md->nltype = NLTYPE_FIXED; + if (newline > 255) + { + md->nllen = 2; + md->nl[0] = (newline >> 8) & 255; + md->nl[1] = newline & 255; + } + else + { + md->nllen = 1; + md->nl[0] = newline; + } + } + +/* Check a UTF-8 string if required. Unfortunately there's no way of passing +back the character offset. */ + #ifdef SUPPORT_UTF if (utf && (options & PCRE_NO_UTF8_CHECK) == 0) - { + { int erroroffset; int errorcode = PRIV(valid_utf)((pcre_uchar *)subject, length, &erroroffset); if (errorcode != 0) - { + { if (offsetcount >= 2) - { + { offsets[0] = erroroffset; offsets[1] = errorcode; - } + } #if defined COMPILE_PCRE8 return (errorcode <= PCRE_UTF8_ERR5 && (options & PCRE_PARTIAL_HARD) != 0) ? PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8; @@ -3359,37 +3359,37 @@ if (utf && (options & PCRE_NO_UTF8_CHECK) == 0) #elif defined COMPILE_PCRE32 return PCRE_ERROR_BADUTF32; #endif - } + } #if defined COMPILE_PCRE8 || defined COMPILE_PCRE16 if (start_offset > 0 && start_offset < length && NOT_FIRSTCHAR(((PCRE_PUCHAR)subject)[start_offset])) return PCRE_ERROR_BADUTF8_OFFSET; #endif - } -#endif - -/* If the exec call supplied NULL for tables, use the inbuilt ones. This -is a feature that makes it possible to save compiled regex and re-use them -in other programs later. */ - + } +#endif + +/* If the exec call supplied NULL for tables, use the inbuilt ones. This +is a feature that makes it possible to save compiled regex and re-use them +in other programs later. */ + if (md->tables == NULL) md->tables = PRIV(default_tables); - + /* The "must be at the start of a line" flags are used in a loop when finding where to start. */ - -startline = (re->flags & PCRE_STARTLINE) != 0; -firstline = (re->options & PCRE_FIRSTLINE) != 0; - -/* Set up the first character to match, if available. The first_byte value is -never set for an anchored regular expression, but the anchoring may be forced -at run time, so we have to test for anchoring. The first char may be unset for -an unanchored pattern, of course. If there's no first char and the pattern was -studied, there may be a bitmap of possible first characters. */ - -if (!anchored) - { - if ((re->flags & PCRE_FIRSTSET) != 0) - { + +startline = (re->flags & PCRE_STARTLINE) != 0; +firstline = (re->options & PCRE_FIRSTLINE) != 0; + +/* Set up the first character to match, if available. The first_byte value is +never set for an anchored regular expression, but the anchoring may be forced +at run time, so we have to test for anchoring. The first char may be unset for +an unanchored pattern, of course. If there's no first char and the pattern was +studied, there may be a bitmap of possible first characters. */ + +if (!anchored) + { + if ((re->flags & PCRE_FIRSTSET) != 0) + { has_first_char = TRUE; first_char = first_char2 = (pcre_uchar)(re->first_char); if ((re->flags & PCRE_FCH_CASELESS) != 0) @@ -3400,20 +3400,20 @@ if (!anchored) first_char2 = UCD_OTHERCASE(first_char); #endif } - } - else - { + } + else + { if (!startline && study != NULL && (study->flags & PCRE_STUDY_MAPPED) != 0) - start_bits = study->start_bits; - } - } - -/* For anchored or unanchored matches, there may be a "last known required -character" set. */ - -if ((re->flags & PCRE_REQCHSET) != 0) - { + start_bits = study->start_bits; + } + } + +/* For anchored or unanchored matches, there may be a "last known required +character" set. */ + +if ((re->flags & PCRE_REQCHSET) != 0) + { has_req_char = TRUE; req_char = req_char2 = (pcre_uchar)(re->req_char); if ((re->flags & PCRE_RCH_CASELESS) != 0) @@ -3424,27 +3424,27 @@ if ((re->flags & PCRE_REQCHSET) != 0) req_char2 = UCD_OTHERCASE(req_char); #endif } - } - -/* Call the main matching function, looping for a non-anchored regex after a + } + +/* Call the main matching function, looping for a non-anchored regex after a failed match. If not restarting, perform certain optimizations at the start of a match. */ - -for (;;) - { - int rc; - - if ((options & PCRE_DFA_RESTART) == 0) - { + +for (;;) + { + int rc; + + if ((options & PCRE_DFA_RESTART) == 0) + { const pcre_uchar *save_end_subject = end_subject; - + /* If firstline is TRUE, the start of the match is constrained to the first line of a multiline string. Implement this by temporarily adjusting end_subject so that we stop scanning at a newline. If the match fails at the newline, later code breaks this loop. */ - - if (firstline) - { + + if (firstline) + { PCRE_PUCHAR t = current_subject; #ifdef SUPPORT_UTF if (utf) @@ -3457,10 +3457,10 @@ for (;;) } else #endif - while (t < md->end_subject && !IS_NEWLINE(t)) t++; - end_subject = t; - } - + while (t < md->end_subject && !IS_NEWLINE(t)) t++; + end_subject = t; + } + /* There are some optimizations that avoid running the match if a known starting point is not found. However, there is an option that disables these, for testing and for ensuring that all callouts do actually occur. @@ -3468,9 +3468,9 @@ for (;;) match-time options. */ if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0) - { + { /* Advance to a known first pcre_uchar (i.e. data item) */ - + if (has_first_char) { if (first_char != first_char2) @@ -3485,11 +3485,11 @@ for (;;) UCHAR21TEST(current_subject) != first_char) current_subject++; } - + /* Or to just after a linebreak for a multiline match if possible */ else if (startline) - { + { if (current_subject > md->start_subject + start_offset) { #ifdef SUPPORT_UTF @@ -3507,23 +3507,23 @@ for (;;) #endif while (current_subject < end_subject && !WAS_NEWLINE(current_subject)) current_subject++; - + /* If we have just passed a CR and the newline option is ANY or ANYCRLF, and we are now at a LF, advance the match position by one more character. */ - + if (UCHAR21TEST(current_subject - 1) == CHAR_CR && (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) && current_subject < end_subject && UCHAR21TEST(current_subject) == CHAR_NL) current_subject++; } - } - + } + /* Advance to a non-unique first pcre_uchar after study */ - + else if (start_bits != NULL) - { + { while (current_subject < end_subject) { register pcre_uint32 c = UCHAR21TEST(current_subject); @@ -3533,17 +3533,17 @@ for (;;) if ((start_bits[c/8] & (1 << (c&7))) != 0) break; current_subject++; } - } - } - - /* Restore fudged end_subject */ - - end_subject = save_end_subject; - + } + } + + /* Restore fudged end_subject */ + + end_subject = save_end_subject; + /* The following two optimizations are disabled for partial matching or if disabling is explicitly requested (and of course, by the test above, this code is not obeyed when restarting after a partial match). */ - + if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0 && (options & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) == 0) { @@ -3552,11 +3552,11 @@ for (;;) pattern. Although the value is, strictly, in characters, we treat it as in pcre_uchar units to avoid spending too much time in this optimization. */ - + if (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0 && (pcre_uint32)(end_subject - current_subject) < study->minlength) return PCRE_ERROR_NOMATCH; - + /* If req_char is set, we know that that pcre_uchar must appear in the subject for the match to succeed. If the first pcre_uchar is set, req_char must be later in the subject; otherwise the test starts at the @@ -3564,21 +3564,21 @@ for (;;) with nested unlimited repeats that aren't going to match. Writing separate code for cased/caseless versions makes it go faster, as does using an autoincrement and backing off on a match. - + HOWEVER: when the subject string is very, very long, searching to its end can take a long time, and give bad performance on quite ordinary patterns. This showed up when somebody was matching /^C/ on a 32-megabyte string... so we don't do this when the string is sufficiently long. */ - + if (has_req_char && end_subject - current_subject < REQ_BYTE_MAX) - { + { register PCRE_PUCHAR p = current_subject + (has_first_char? 1:0); /* We don't need to repeat the search if we haven't yet reached the place we found it at last time. */ if (p > req_char_ptr) - { + { if (req_char != req_char2) { while (p < end_subject) @@ -3594,41 +3594,41 @@ for (;;) if (UCHAR21INCTEST(p) == req_char) { p--; break; } } } - + /* If we can't find the required pcre_uchar, break the matching loop, which will cause a return or PCRE_ERROR_NOMATCH. */ - + if (p >= end_subject) break; - + /* If we have found the required pcre_uchar, save the point where we found it, so that we don't search again next time round the loop if the start hasn't passed this point yet. */ - + req_char_ptr = p; } } - } + } } /* End of optimizations that are done when not restarting */ - - /* OK, now we can do the business */ - + + /* OK, now we can do the business */ + md->start_used_ptr = current_subject; md->recursive = NULL; - rc = internal_dfa_exec( - md, /* fixed match data */ - md->start_code, /* this subexpression's code */ - current_subject, /* where we currently are */ - start_offset, /* start offset in subject */ - offsets, /* offset vector */ - offsetcount, /* size of same */ - workspace, /* workspace vector */ - wscount, /* size of same */ + rc = internal_dfa_exec( + md, /* fixed match data */ + md->start_code, /* this subexpression's code */ + current_subject, /* where we currently are */ + start_offset, /* start offset in subject */ + offsets, /* offset vector */ + offsetcount, /* size of same */ + workspace, /* workspace vector */ + wscount, /* size of same */ 0); /* function recurse level */ - - /* Anything other than "no match" means we are done, always; otherwise, carry - on only if not anchored. */ - + + /* Anything other than "no match" means we are done, always; otherwise, carry + on only if not anchored. */ + if (rc != PCRE_ERROR_NOMATCH || anchored) { if (rc == PCRE_ERROR_PARTIAL && offsetcount >= 2) @@ -3640,37 +3640,37 @@ for (;;) } return rc; } - - /* Advance to the next subject character unless we are at the end of a line - and firstline is set. */ - - if (firstline && IS_NEWLINE(current_subject)) break; - current_subject++; + + /* Advance to the next subject character unless we are at the end of a line + and firstline is set. */ + + if (firstline && IS_NEWLINE(current_subject)) break; + current_subject++; #ifdef SUPPORT_UTF if (utf) - { + { ACROSSCHAR(current_subject < end_subject, *current_subject, current_subject++); - } + } #endif - if (current_subject > end_subject) break; - - /* If we have just passed a CR and we are now at a LF, and the pattern does - not contain any explicit matches for \r or \n, and the newline option is CRLF - or ANY or ANYCRLF, advance the match position by one more character. */ - + if (current_subject > end_subject) break; + + /* If we have just passed a CR and we are now at a LF, and the pattern does + not contain any explicit matches for \r or \n, and the newline option is CRLF + or ANY or ANYCRLF, advance the match position by one more character. */ + if (UCHAR21TEST(current_subject - 1) == CHAR_CR && - current_subject < end_subject && + current_subject < end_subject && UCHAR21TEST(current_subject) == CHAR_NL && - (re->flags & PCRE_HASCRORLF) == 0 && - (md->nltype == NLTYPE_ANY || - md->nltype == NLTYPE_ANYCRLF || - md->nllen == 2)) - current_subject++; - - } /* "Bumpalong" loop */ - -return PCRE_ERROR_NOMATCH; -} - -/* End of pcre_dfa_exec.c */ + (re->flags & PCRE_HASCRORLF) == 0 && + (md->nltype == NLTYPE_ANY || + md->nltype == NLTYPE_ANYCRLF || + md->nllen == 2)) + current_subject++; + + } /* "Bumpalong" loop */ + +return PCRE_ERROR_NOMATCH; +} + +/* End of pcre_dfa_exec.c */ diff --git a/contrib/libs/pcre/pcre_exec.c b/contrib/libs/pcre/pcre_exec.c index 4b5cb73fea..9d023d74e9 100644 --- a/contrib/libs/pcre/pcre_exec.c +++ b/contrib/libs/pcre/pcre_exec.c @@ -1,61 +1,61 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2018 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - -/* This module contains pcre_exec(), the externally visible function that does -pattern matching using an NFA algorithm, trying to mimic Perl as closely as -possible. There are also some static supporting functions. */ - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + +/* This module contains pcre_exec(), the externally visible function that does +pattern matching using an NFA algorithm, trying to mimic Perl as closely as +possible. There are also some static supporting functions. */ + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#define NLBLOCK md /* Block containing newline information */ -#define PSSTART start_subject /* Field containing processed string start */ -#define PSEND end_subject /* Field containing processed string end */ - -#include "pcre_internal.h" - -/* Undefine some potentially clashing cpp symbols */ - -#undef min -#undef max - +#endif + +#define NLBLOCK md /* Block containing newline information */ +#define PSSTART start_subject /* Field containing processed string start */ +#define PSEND end_subject /* Field containing processed string end */ + +#include "pcre_internal.h" + +/* Undefine some potentially clashing cpp symbols */ + +#undef min +#undef max + /* The md->capture_last field uses the lower 16 bits for the last captured substring (which can never be greater than 65535) and a bit in the top half to mean "capture vector overflowed". This odd way of doing things was @@ -65,11 +65,11 @@ interface, and doing it this way saved on (a) another variable, which would have increased the stack frame size (a big NO-NO in PCRE) and (b) another separate set of save/restore instructions. The following defines are used in implementing this. */ - + #define CAPLMASK 0x0000ffff /* The bits used for last_capture */ #define OVFLMASK 0xffff0000 /* The bits used for the overflow flag */ #define OVFLBIT 0x00010000 /* The bit that is set for overflow */ - + /* Values for setting in md->match_function_type to indicate two special types of call to match(). We do it this way to save on using another stack variable, as stack usage is to be discouraged. */ @@ -77,15 +77,15 @@ as stack usage is to be discouraged. */ #define MATCH_CONDASSERT 1 /* Called to check a condition assertion */ #define MATCH_CBEGROUP 2 /* Could-be-empty unlimited repeat group */ -/* Non-error returns from the match() function. Error returns are externally -defined PCRE_ERROR_xxx codes, which are all negative. */ - -#define MATCH_MATCH 1 -#define MATCH_NOMATCH 0 - -/* Special internal returns from the match() function. Make them sufficiently -negative to avoid the external error codes. */ - +/* Non-error returns from the match() function. Error returns are externally +defined PCRE_ERROR_xxx codes, which are all negative. */ + +#define MATCH_MATCH 1 +#define MATCH_NOMATCH 0 + +/* Special internal returns from the match() function. Make them sufficiently +negative to avoid the external error codes. */ + #define MATCH_ACCEPT (-999) #define MATCH_KETRPOS (-998) #define MATCH_ONCE (-997) @@ -98,103 +98,103 @@ for any one of them can use a range. */ #define MATCH_THEN (-992) #define MATCH_BACKTRACK_MAX MATCH_THEN #define MATCH_BACKTRACK_MIN MATCH_COMMIT - -/* Maximum number of ints of offset to save on the stack for recursive calls. -If the offset vector is bigger, malloc is used. This should be a multiple of 3, -because the offset vector is always a multiple of 3 long. */ - -#define REC_STACK_SAVE_MAX 30 - -/* Min and max values for the common repeats; for the maxima, 0 => infinity */ - + +/* Maximum number of ints of offset to save on the stack for recursive calls. +If the offset vector is bigger, malloc is used. This should be a multiple of 3, +because the offset vector is always a multiple of 3 long. */ + +#define REC_STACK_SAVE_MAX 30 + +/* Min and max values for the common repeats; for the maxima, 0 => infinity */ + static const char rep_min[] = { 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, }; static const char rep_max[] = { 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 1, }; - + #ifdef PCRE_DEBUG -/************************************************* -* Debugging function to print chars * -*************************************************/ - -/* Print a sequence of chars in printable format, stopping at the end of the -subject if the requested. - -Arguments: - p points to characters - length number to print - is_subject TRUE if printing from within md->start_subject - md pointer to matching data block, if is_subject is TRUE - -Returns: nothing -*/ - -static void +/************************************************* +* Debugging function to print chars * +*************************************************/ + +/* Print a sequence of chars in printable format, stopping at the end of the +subject if the requested. + +Arguments: + p points to characters + length number to print + is_subject TRUE if printing from within md->start_subject + md pointer to matching data block, if is_subject is TRUE + +Returns: nothing +*/ + +static void pchars(const pcre_uchar *p, int length, BOOL is_subject, match_data *md) -{ +{ pcre_uint32 c; BOOL utf = md->utf; -if (is_subject && length > md->end_subject - p) length = md->end_subject - p; -while (length-- > 0) +if (is_subject && length > md->end_subject - p) length = md->end_subject - p; +while (length-- > 0) if (isprint(c = UCHAR21INCTEST(p))) printf("%c", (char)c); else printf("\\x{%02x}", c); -} -#endif - - - -/************************************************* -* Match a back-reference * -*************************************************/ - +} +#endif + + + +/************************************************* +* Match a back-reference * +*************************************************/ + /* Normally, if a back reference hasn't been set, the length that is passed is negative, so the match always fails. However, in JavaScript compatibility mode, the length passed is zero. Note that in caseless UTF-8 mode, the number of subject bytes matched may be different to the number of reference bytes. - -Arguments: - offset index into the offset vector + +Arguments: + offset index into the offset vector eptr pointer into the subject length length of reference to be matched (number of bytes) - md points to match data block + md points to match data block caseless TRUE if caseless - + Returns: >= 0 the number of subject bytes matched -1 no match -2 partial match; always given if at end subject -*/ - +*/ + static int match_ref(int offset, register PCRE_PUCHAR eptr, int length, match_data *md, BOOL caseless) -{ +{ PCRE_PUCHAR eptr_start = eptr; register PCRE_PUCHAR p = md->start_subject + md->offset_vector[offset]; #if defined SUPPORT_UTF && defined SUPPORT_UCP BOOL utf = md->utf; #endif - + #ifdef PCRE_DEBUG -if (eptr >= md->end_subject) - printf("matching subject <null>"); -else - { - printf("matching subject "); - pchars(eptr, length, TRUE, md); - } -printf(" against backref "); -pchars(p, length, FALSE, md); -printf("\n"); -#endif - +if (eptr >= md->end_subject) + printf("matching subject <null>"); +else + { + printf("matching subject "); + pchars(eptr, length, TRUE, md); + } +printf(" against backref "); +pchars(p, length, FALSE, md); +printf("\n"); +#endif + /* Always fail if reference not set (and not JavaScript compatible - in that case the length is passed as zero). */ - + if (length < 0) return -1; - + /* Separate the caseless case for speed. In UTF-8 mode we can only do this properly if Unicode properties are supported. Otherwise, we can check only ASCII characters. */ - + if (caseless) - { + { #if defined SUPPORT_UTF && defined SUPPORT_UCP if (utf) { @@ -251,103 +251,103 @@ are in UTF-8 mode. */ else { - while (length-- > 0) + while (length-- > 0) { if (eptr >= md->end_subject) return -2; /* Partial match */ if (UCHAR21INCTEST(p) != UCHAR21INCTEST(eptr)) return -1; } - } - + } + return (int)(eptr - eptr_start); -} - - - -/*************************************************************************** -**************************************************************************** - RECURSION IN THE match() FUNCTION - -The match() function is highly recursive, though not every recursive call -increases the recursive depth. Nevertheless, some regular expressions can cause -it to recurse to a great depth. I was writing for Unix, so I just let it call -itself recursively. This uses the stack for saving everything that has to be -saved for a recursive call. On Unix, the stack can be large, and this works -fine. - -It turns out that on some non-Unix-like systems there are problems with -programs that use a lot of stack. (This despite the fact that every last chip -has oodles of memory these days, and techniques for extending the stack have -been known for decades.) So.... - -There is a fudge, triggered by defining NO_RECURSE, which avoids recursive -calls by keeping local variables that need to be preserved in blocks of memory -obtained from malloc() instead instead of on the stack. Macros are used to -achieve this so that the actual code doesn't look very different to what it -always used to. - -The original heap-recursive code used longjmp(). However, it seems that this -can be very slow on some operating systems. Following a suggestion from Stan -Switzer, the use of longjmp() has been abolished, at the cost of having to -provide a unique number for each call to RMATCH. There is no way of generating -a sequence of numbers at compile time in C. I have given them names, to make -them stand out more clearly. - -Crude tests on x86 Linux show a small speedup of around 5-8%. However, on -FreeBSD, avoiding longjmp() more than halves the time taken to run the standard -tests. Furthermore, not using longjmp() means that local dynamic variables -don't have indeterminate values; this has meant that the frame size can be -reduced because the result can be "passed back" by straight setting of the -variable instead of being passed in the frame. -**************************************************************************** -***************************************************************************/ - -/* Numbers for RMATCH calls. When this list is changed, the code at HEAP_RETURN -below must be updated in sync. */ - -enum { RM1=1, RM2, RM3, RM4, RM5, RM6, RM7, RM8, RM9, RM10, - RM11, RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20, - RM21, RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30, - RM31, RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40, - RM41, RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50, +} + + + +/*************************************************************************** +**************************************************************************** + RECURSION IN THE match() FUNCTION + +The match() function is highly recursive, though not every recursive call +increases the recursive depth. Nevertheless, some regular expressions can cause +it to recurse to a great depth. I was writing for Unix, so I just let it call +itself recursively. This uses the stack for saving everything that has to be +saved for a recursive call. On Unix, the stack can be large, and this works +fine. + +It turns out that on some non-Unix-like systems there are problems with +programs that use a lot of stack. (This despite the fact that every last chip +has oodles of memory these days, and techniques for extending the stack have +been known for decades.) So.... + +There is a fudge, triggered by defining NO_RECURSE, which avoids recursive +calls by keeping local variables that need to be preserved in blocks of memory +obtained from malloc() instead instead of on the stack. Macros are used to +achieve this so that the actual code doesn't look very different to what it +always used to. + +The original heap-recursive code used longjmp(). However, it seems that this +can be very slow on some operating systems. Following a suggestion from Stan +Switzer, the use of longjmp() has been abolished, at the cost of having to +provide a unique number for each call to RMATCH. There is no way of generating +a sequence of numbers at compile time in C. I have given them names, to make +them stand out more clearly. + +Crude tests on x86 Linux show a small speedup of around 5-8%. However, on +FreeBSD, avoiding longjmp() more than halves the time taken to run the standard +tests. Furthermore, not using longjmp() means that local dynamic variables +don't have indeterminate values; this has meant that the frame size can be +reduced because the result can be "passed back" by straight setting of the +variable instead of being passed in the frame. +**************************************************************************** +***************************************************************************/ + +/* Numbers for RMATCH calls. When this list is changed, the code at HEAP_RETURN +below must be updated in sync. */ + +enum { RM1=1, RM2, RM3, RM4, RM5, RM6, RM7, RM8, RM9, RM10, + RM11, RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20, + RM21, RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30, + RM31, RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40, + RM41, RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50, RM51, RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60, RM61, RM62, RM63, RM64, RM65, RM66, RM67 }; - -/* These versions of the macros use the stack, as normal. There are debugging -versions and production versions. Note that the "rw" argument of RMATCH isn't + +/* These versions of the macros use the stack, as normal. There are debugging +versions and production versions. Note that the "rw" argument of RMATCH isn't actually used in this definition. */ - -#ifndef NO_RECURSE -#define REGISTER register - + +#ifndef NO_RECURSE +#define REGISTER register + #ifdef PCRE_DEBUG #define RMATCH(ra,rb,rc,rd,re,rw) \ - { \ - printf("match() called in line %d\n", __LINE__); \ + { \ + printf("match() called in line %d\n", __LINE__); \ rrc = match(ra,rb,mstart,rc,rd,re,rdepth+1); \ - printf("to line %d\n", __LINE__); \ - } -#define RRETURN(ra) \ - { \ + printf("to line %d\n", __LINE__); \ + } +#define RRETURN(ra) \ + { \ printf("match() returned %d from line %d\n", ra, __LINE__); \ - return ra; \ - } -#else + return ra; \ + } +#else #define RMATCH(ra,rb,rc,rd,re,rw) \ rrc = match(ra,rb,mstart,rc,rd,re,rdepth+1) -#define RRETURN(ra) return ra -#endif - -#else - - -/* These versions of the macros manage a private stack on the heap. Note that -the "rd" argument of RMATCH isn't actually used in this definition. It's the md -argument of match(), which never changes. */ - -#define REGISTER - +#define RRETURN(ra) return ra +#endif + +#else + + +/* These versions of the macros manage a private stack on the heap. Note that +the "rd" argument of RMATCH isn't actually used in this definition. It's the md +argument of match(), which never changes. */ + +#define REGISTER + #define RMATCH(ra,rb,rc,rd,re,rw)\ - {\ + {\ heapframe *newframe = frame->Xnextframe;\ if (newframe == NULL)\ {\ @@ -357,50 +357,50 @@ argument of match(), which never changes. */ frame->Xnextframe = newframe;\ }\ frame->Xwhere = rw;\ - newframe->Xeptr = ra;\ - newframe->Xecode = rb;\ - newframe->Xmstart = mstart;\ - newframe->Xoffset_top = rc;\ + newframe->Xeptr = ra;\ + newframe->Xecode = rb;\ + newframe->Xmstart = mstart;\ + newframe->Xoffset_top = rc;\ newframe->Xeptrb = re;\ - newframe->Xrdepth = frame->Xrdepth + 1;\ - newframe->Xprevframe = frame;\ - frame = newframe;\ - DPRINTF(("restarting from line %d\n", __LINE__));\ - goto HEAP_RECURSE;\ - L_##rw:\ - DPRINTF(("jumped back to line %d\n", __LINE__));\ - } - -#define RRETURN(ra)\ - {\ + newframe->Xrdepth = frame->Xrdepth + 1;\ + newframe->Xprevframe = frame;\ + frame = newframe;\ + DPRINTF(("restarting from line %d\n", __LINE__));\ + goto HEAP_RECURSE;\ + L_##rw:\ + DPRINTF(("jumped back to line %d\n", __LINE__));\ + } + +#define RRETURN(ra)\ + {\ heapframe *oldframe = frame;\ frame = oldframe->Xprevframe;\ - if (frame != NULL)\ - {\ - rrc = ra;\ - goto HEAP_RETURN;\ - }\ - return ra;\ - } - - -/* Structure for remembering the local variables in a private frame */ - -typedef struct heapframe { - struct heapframe *Xprevframe; + if (frame != NULL)\ + {\ + rrc = ra;\ + goto HEAP_RETURN;\ + }\ + return ra;\ + } + + +/* Structure for remembering the local variables in a private frame */ + +typedef struct heapframe { + struct heapframe *Xprevframe; struct heapframe *Xnextframe; - - /* Function arguments that may change */ - + + /* Function arguments that may change */ + PCRE_PUCHAR Xeptr; const pcre_uchar *Xecode; PCRE_PUCHAR Xmstart; - int Xoffset_top; - eptrblock *Xeptrb; - unsigned int Xrdepth; - - /* Function local variables */ - + int Xoffset_top; + eptrblock *Xeptrb; + unsigned int Xrdepth; + + /* Function local variables */ + PCRE_PUCHAR Xcallpat; #ifdef SUPPORT_UTF PCRE_PUCHAR Xcharptr; @@ -410,59 +410,59 @@ typedef struct heapframe { PCRE_PUCHAR Xpp; PCRE_PUCHAR Xprev; PCRE_PUCHAR Xsaved_eptr; - - recursion_info Xnew_recursive; - - BOOL Xcur_is_word; - BOOL Xcondition; - BOOL Xprev_is_word; - -#ifdef SUPPORT_UCP - int Xprop_type; + + recursion_info Xnew_recursive; + + BOOL Xcur_is_word; + BOOL Xcondition; + BOOL Xprev_is_word; + +#ifdef SUPPORT_UCP + int Xprop_type; unsigned int Xprop_value; - int Xprop_fail_result; - int Xoclength; + int Xprop_fail_result; + int Xoclength; pcre_uchar Xocchars[6]; -#endif - +#endif + int Xcodelink; - int Xctype; - unsigned int Xfc; - int Xfi; - int Xlength; - int Xmax; - int Xmin; + int Xctype; + unsigned int Xfc; + int Xfi; + int Xlength; + int Xmax; + int Xmin; unsigned int Xnumber; - int Xoffset; + int Xoffset; unsigned int Xop; pcre_int32 Xsave_capture_last; - int Xsave_offset1, Xsave_offset2, Xsave_offset3; - int Xstacksave[REC_STACK_SAVE_MAX]; - - eptrblock Xnewptrb; - - /* Where to jump back to */ - - int Xwhere; - -} heapframe; - -#endif - - -/*************************************************************************** -***************************************************************************/ - - - -/************************************************* -* Match from current position * -*************************************************/ - -/* This function is called recursively in many circumstances. Whenever it -returns a negative (error) response, the outer incarnation must also return the + int Xsave_offset1, Xsave_offset2, Xsave_offset3; + int Xstacksave[REC_STACK_SAVE_MAX]; + + eptrblock Xnewptrb; + + /* Where to jump back to */ + + int Xwhere; + +} heapframe; + +#endif + + +/*************************************************************************** +***************************************************************************/ + + + +/************************************************* +* Match from current position * +*************************************************/ + +/* This function is called recursively in many circumstances. Whenever it +returns a negative (error) response, the outer incarnation must also return the same response. */ - + /* These macros pack up tests that are used for partial matching, and which appear several times in the code. We set the "hit end" flag if the pointer is at the end of the subject and also past the start of the subject (i.e. @@ -488,29 +488,29 @@ the subject. */ /* Performance note: It might be tempting to extract commonly used fields from the md structure (e.g. utf, end_subject) into individual variables to improve -performance. Tests using gcc on a SPARC disproved this; in the first case, it -made performance worse. - -Arguments: - eptr pointer to current character in subject - ecode pointer to current position in compiled code - mstart pointer to the current match start position (can be modified - by encountering \K) - offset_top current top pointer - md pointer to "static" info for the match - eptrb pointer to chain of blocks containing eptr at start of - brackets - for testing for empty matches - rdepth the recursion depth - -Returns: MATCH_MATCH if matched ) these values are >= 0 - MATCH_NOMATCH if failed to match ) +performance. Tests using gcc on a SPARC disproved this; in the first case, it +made performance worse. + +Arguments: + eptr pointer to current character in subject + ecode pointer to current position in compiled code + mstart pointer to the current match start position (can be modified + by encountering \K) + offset_top current top pointer + md pointer to "static" info for the match + eptrb pointer to chain of blocks containing eptr at start of + brackets - for testing for empty matches + rdepth the recursion depth + +Returns: MATCH_MATCH if matched ) these values are >= 0 + MATCH_NOMATCH if failed to match ) a negative MATCH_xxx value for PRUNE, SKIP, etc - a negative PCRE_ERROR_xxx value if aborted by an error condition - (e.g. stopped by repeated call or recursion limit) -*/ - + a negative PCRE_ERROR_xxx value if aborted by an error condition + (e.g. stopped by repeated call or recursion limit) +*/ + #ifdef __GNUC__ -static int +static int match(REGISTER PCRE_PUCHAR eptr, REGISTER const pcre_uchar *ecode, PCRE_PUCHAR mstart, int offset_top, match_data *md, eptrblock *eptrb, unsigned int rdepth) __attribute__((noinline,noclone)); @@ -519,104 +519,104 @@ static int match(REGISTER PCRE_PUCHAR eptr, REGISTER const pcre_uchar *ecode, PCRE_PUCHAR mstart, int offset_top, match_data *md, eptrblock *eptrb, unsigned int rdepth) -{ -/* These variables do not need to be preserved over recursion in this function, -so they can be ordinary variables in all cases. Mark some of them with -"register" because they are used a lot in loops. */ - -register int rrc; /* Returns from recursive calls */ -register int i; /* Used for loops not involving calls to RMATCH() */ +{ +/* These variables do not need to be preserved over recursion in this function, +so they can be ordinary variables in all cases. Mark some of them with +"register" because they are used a lot in loops. */ + +register int rrc; /* Returns from recursive calls */ +register int i; /* Used for loops not involving calls to RMATCH() */ register pcre_uint32 c; /* Character values not kept over RMATCH() calls */ register BOOL utf; /* Local copy of UTF flag for speed */ - -BOOL minimize, possessive; /* Quantifier options */ + +BOOL minimize, possessive; /* Quantifier options */ BOOL caseless; int condcode; - -/* When recursion is not being used, all "local" variables that have to be + +/* When recursion is not being used, all "local" variables that have to be preserved over calls to RMATCH() are part of a "frame". We set up the top-level frame on the stack here; subsequent instantiations are obtained from the heap whenever RMATCH() does a "recursion". See the macro definitions above. Putting the top-level on the stack rather than malloc-ing them all gives a performance boost in many cases where there is not much "recursion". */ - -#ifdef NO_RECURSE + +#ifdef NO_RECURSE heapframe *frame = (heapframe *)md->match_frames_base; - -/* Copy in the original argument variables */ - -frame->Xeptr = eptr; -frame->Xecode = ecode; -frame->Xmstart = mstart; -frame->Xoffset_top = offset_top; -frame->Xeptrb = eptrb; -frame->Xrdepth = rdepth; - -/* This is where control jumps back to to effect "recursion" */ - -HEAP_RECURSE: - -/* Macros make the argument variables come from the current frame */ - -#define eptr frame->Xeptr -#define ecode frame->Xecode -#define mstart frame->Xmstart -#define offset_top frame->Xoffset_top -#define eptrb frame->Xeptrb -#define rdepth frame->Xrdepth - -/* Ditto for the local variables */ - + +/* Copy in the original argument variables */ + +frame->Xeptr = eptr; +frame->Xecode = ecode; +frame->Xmstart = mstart; +frame->Xoffset_top = offset_top; +frame->Xeptrb = eptrb; +frame->Xrdepth = rdepth; + +/* This is where control jumps back to to effect "recursion" */ + +HEAP_RECURSE: + +/* Macros make the argument variables come from the current frame */ + +#define eptr frame->Xeptr +#define ecode frame->Xecode +#define mstart frame->Xmstart +#define offset_top frame->Xoffset_top +#define eptrb frame->Xeptrb +#define rdepth frame->Xrdepth + +/* Ditto for the local variables */ + #ifdef SUPPORT_UTF -#define charptr frame->Xcharptr -#endif -#define callpat frame->Xcallpat +#define charptr frame->Xcharptr +#endif +#define callpat frame->Xcallpat #define codelink frame->Xcodelink -#define data frame->Xdata -#define next frame->Xnext -#define pp frame->Xpp -#define prev frame->Xprev -#define saved_eptr frame->Xsaved_eptr - -#define new_recursive frame->Xnew_recursive - -#define cur_is_word frame->Xcur_is_word -#define condition frame->Xcondition -#define prev_is_word frame->Xprev_is_word - -#ifdef SUPPORT_UCP -#define prop_type frame->Xprop_type -#define prop_value frame->Xprop_value -#define prop_fail_result frame->Xprop_fail_result -#define oclength frame->Xoclength -#define occhars frame->Xocchars -#endif - -#define ctype frame->Xctype -#define fc frame->Xfc -#define fi frame->Xfi -#define length frame->Xlength -#define max frame->Xmax -#define min frame->Xmin -#define number frame->Xnumber -#define offset frame->Xoffset -#define op frame->Xop -#define save_capture_last frame->Xsave_capture_last -#define save_offset1 frame->Xsave_offset1 -#define save_offset2 frame->Xsave_offset2 -#define save_offset3 frame->Xsave_offset3 -#define stacksave frame->Xstacksave - -#define newptrb frame->Xnewptrb - -/* When recursion is being used, local variables are allocated on the stack and -get preserved during recursion in the normal way. In this environment, fi and -i, and fc and c, can be the same variables. */ - -#else /* NO_RECURSE not defined */ -#define fi i -#define fc c - +#define data frame->Xdata +#define next frame->Xnext +#define pp frame->Xpp +#define prev frame->Xprev +#define saved_eptr frame->Xsaved_eptr + +#define new_recursive frame->Xnew_recursive + +#define cur_is_word frame->Xcur_is_word +#define condition frame->Xcondition +#define prev_is_word frame->Xprev_is_word + +#ifdef SUPPORT_UCP +#define prop_type frame->Xprop_type +#define prop_value frame->Xprop_value +#define prop_fail_result frame->Xprop_fail_result +#define oclength frame->Xoclength +#define occhars frame->Xocchars +#endif + +#define ctype frame->Xctype +#define fc frame->Xfc +#define fi frame->Xfi +#define length frame->Xlength +#define max frame->Xmax +#define min frame->Xmin +#define number frame->Xnumber +#define offset frame->Xoffset +#define op frame->Xop +#define save_capture_last frame->Xsave_capture_last +#define save_offset1 frame->Xsave_offset1 +#define save_offset2 frame->Xsave_offset2 +#define save_offset3 frame->Xsave_offset3 +#define stacksave frame->Xstacksave + +#define newptrb frame->Xnewptrb + +/* When recursion is being used, local variables are allocated on the stack and +get preserved during recursion in the normal way. In this environment, fi and +i, and fc and c, can be the same variables. */ + +#else /* NO_RECURSE not defined */ +#define fi i +#define fc c + /* Many of the following variables are used only in small blocks of the code. My normal style of coding would have declared them within each of those blocks. However, in order to accommodate the version of this code that uses an external @@ -624,7 +624,7 @@ However, in order to accommodate the version of this code that uses an external declarations can be cut out in a block. The only declarations within blocks below are for variables that do not have to be preserved over a recursive call to RMATCH(). */ - + #ifdef SUPPORT_UTF const pcre_uchar *charptr; #endif @@ -638,30 +638,30 @@ PCRE_PUCHAR saved_eptr; recursion_info new_recursive; BOOL cur_is_word; -BOOL condition; -BOOL prev_is_word; - -#ifdef SUPPORT_UCP -int prop_type; +BOOL condition; +BOOL prev_is_word; + +#ifdef SUPPORT_UCP +int prop_type; unsigned int prop_value; -int prop_fail_result; -int oclength; +int prop_fail_result; +int oclength; pcre_uchar occhars[6]; -#endif - +#endif + int codelink; -int ctype; -int length; -int max; -int min; +int ctype; +int length; +int max; +int min; unsigned int number; -int offset; +int offset; unsigned int op; pcre_int32 save_capture_last; -int save_offset1, save_offset2, save_offset3; -int stacksave[REC_STACK_SAVE_MAX]; - -eptrblock newptrb; +int save_offset1, save_offset2, save_offset3; +int stacksave[REC_STACK_SAVE_MAX]; + +eptrblock newptrb; /* There is a special fudge for calling match() in a way that causes it to measure the size of its basic stack frame when the stack is being used for @@ -679,8 +679,8 @@ if (ecode == NULL) return (len > 0)? -len : len; } } -#endif /* NO_RECURSE */ - +#endif /* NO_RECURSE */ + /* To save space on the stack and in the heap frame, I have doubled up on some of the local variables that are used only in localised parts of the code, but still need to be preserved over recursive calls of match(). These macros define @@ -694,47 +694,47 @@ the alternative names that are used. */ #define foc number #define save_mark data -/* These statements are here to stop the compiler complaining about unitialized -variables. */ - -#ifdef SUPPORT_UCP -prop_value = 0; -prop_fail_result = 0; -#endif - - -/* This label is used for tail recursion, which is used in a few cases even -when NO_RECURSE is not defined, in order to reduce the amount of stack that is -used. Thanks to Ian Taylor for noticing this possibility and sending the -original patch. */ - -TAIL_RECURSE: - -/* OK, now we can get on with the real code of the function. Recursive calls -are specified by the macro RMATCH and RRETURN is used to return. When -NO_RECURSE is *not* defined, these just turn into a recursive call to match() +/* These statements are here to stop the compiler complaining about unitialized +variables. */ + +#ifdef SUPPORT_UCP +prop_value = 0; +prop_fail_result = 0; +#endif + + +/* This label is used for tail recursion, which is used in a few cases even +when NO_RECURSE is not defined, in order to reduce the amount of stack that is +used. Thanks to Ian Taylor for noticing this possibility and sending the +original patch. */ + +TAIL_RECURSE: + +/* OK, now we can get on with the real code of the function. Recursive calls +are specified by the macro RMATCH and RRETURN is used to return. When +NO_RECURSE is *not* defined, these just turn into a recursive call to match() and a "return", respectively (possibly with some debugging if PCRE_DEBUG is -defined). However, RMATCH isn't like a function call because it's quite a -complicated macro. It has to be used in one particular way. This shouldn't, -however, impact performance when true recursion is being used. */ - +defined). However, RMATCH isn't like a function call because it's quite a +complicated macro. It has to be used in one particular way. This shouldn't, +however, impact performance when true recursion is being used. */ + #ifdef SUPPORT_UTF utf = md->utf; /* Local copy of the flag */ -#else +#else utf = FALSE; -#endif - -/* First check that we haven't called match() too many times, or that we -haven't exceeded the recursive call limit. */ - -if (md->match_call_count++ >= md->match_limit) RRETURN(PCRE_ERROR_MATCHLIMIT); -if (rdepth >= md->match_limit_recursion) RRETURN(PCRE_ERROR_RECURSIONLIMIT); - -/* At the start of a group with an unlimited repeat that may match an empty +#endif + +/* First check that we haven't called match() too many times, or that we +haven't exceeded the recursive call limit. */ + +if (md->match_call_count++ >= md->match_limit) RRETURN(PCRE_ERROR_MATCHLIMIT); +if (rdepth >= md->match_limit_recursion) RRETURN(PCRE_ERROR_RECURSIONLIMIT); + +/* At the start of a group with an unlimited repeat that may match an empty string, the variable md->match_function_type is set to MATCH_CBEGROUP. It is done this way to save having to use another function argument, which would take up space on the stack. See also MATCH_CONDASSERT below. - + When MATCH_CBEGROUP is set, add the current subject pointer to the chain of such remembered pointers, to be checked when we hit the closing ket, in order to break infinite loops that match no characters. When match() is called in @@ -743,20 +743,20 @@ NOT be used with tail recursion, because the memory block that is used is on the stack, so a new one may be required for each match(). */ if (md->match_function_type == MATCH_CBEGROUP) - { - newptrb.epb_saved_eptr = eptr; - newptrb.epb_prev = eptrb; - eptrb = &newptrb; + { + newptrb.epb_saved_eptr = eptr; + newptrb.epb_prev = eptrb; + eptrb = &newptrb; md->match_function_type = 0; - } - -/* Now start processing the opcodes. */ - -for (;;) - { - minimize = possessive = FALSE; - op = *ecode; - + } + +/* Now start processing the opcodes. */ + +for (;;) + { + minimize = possessive = FALSE; + op = *ecode; + switch(op) { case OP_MARK: @@ -766,14 +766,14 @@ for (;;) eptrb, RM55); if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) && md->mark == NULL) md->mark = ecode + 2; - + /* A return of MATCH_SKIP_ARG means that matching failed at SKIP with an argument, and we must check whether that argument matches this MARK's argument. It is passed back in md->start_match_ptr (an overloading of that variable). If it does match, we reset that variable to the current subject position and return MATCH_SKIP. Otherwise, pass back the return code unaltered. */ - + else if (rrc == MATCH_SKIP_ARG && STRCMP_UC_UC_TEST(ecode + 2, md->start_match_ptr) == 0) { @@ -782,21 +782,21 @@ for (;;) } RRETURN(rrc); - case OP_FAIL: - RRETURN(MATCH_NOMATCH); - + case OP_FAIL: + RRETURN(MATCH_NOMATCH); + case OP_COMMIT: RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb, RM52); if (rrc != MATCH_NOMATCH) RRETURN(rrc); RRETURN(MATCH_COMMIT); - case OP_PRUNE: + case OP_PRUNE: RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb, RM51); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); - RRETURN(MATCH_PRUNE); - + if (rrc != MATCH_NOMATCH) RRETURN(rrc); + RRETURN(MATCH_PRUNE); + case OP_PRUNE_ARG: md->nomatch_mark = ecode + 2; md->mark = NULL; /* In case previously set by assertion */ @@ -804,16 +804,16 @@ for (;;) eptrb, RM56); if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) && md->mark == NULL) md->mark = ecode + 2; - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); RRETURN(MATCH_PRUNE); - - case OP_SKIP: + + case OP_SKIP: RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb, RM53); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); - md->start_match_ptr = eptr; /* Pass back current position */ - RRETURN(MATCH_SKIP); - + if (rrc != MATCH_NOMATCH) RRETURN(rrc); + md->start_match_ptr = eptr; /* Pass back current position */ + RRETURN(MATCH_SKIP); + /* Note that, for Perl compatibility, SKIP with an argument does NOT set nomatch_mark. When a pattern match ends with a SKIP_ARG for which there was not a matching mark, we have to re-run the match, ignoring the SKIP_ARG @@ -845,13 +845,13 @@ for (;;) the branch in which it occurs can be determined. Overload the start of match pointer to do this. */ - case OP_THEN: + case OP_THEN: RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb, RM54); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); md->start_match_ptr = ecode; - RRETURN(MATCH_THEN); - + RRETURN(MATCH_THEN); + case OP_THEN_ARG: md->nomatch_mark = ecode + 2; md->mark = NULL; /* In case previously set by assertion */ @@ -862,7 +862,7 @@ for (;;) if (rrc != MATCH_NOMATCH) RRETURN(rrc); md->start_match_ptr = ecode; RRETURN(MATCH_THEN); - + /* Handle an atomic group that does not contain any capturing parentheses. This can be handled like an assertion. Prior to 8.13, all atomic groups were handled this way. In 8.13, the code was changed as below for ONCE, so @@ -870,7 +870,7 @@ for (;;) However, this uses a lot more stack, so in 8.20, atomic groups that do not contain any captures generate OP_ONCE_NC, which can be handled in the old, less stack intensive way. - + Check the alternative branches in turn - the matching won't pass the KET for this kind of subpattern. If any one branch matches, we carry on as at the end of a normal bracket, leaving the subject pointer, but resetting @@ -955,36 +955,36 @@ for (;;) the working value and also the values of the final offsets, in case they were set by a previous iteration of the same bracket. - If there isn't enough space in the offset vector, treat this as if it were - a non-capturing bracket. Don't worry about setting the flag for the error - case here; that is handled in the code for KET. */ - - case OP_CBRA: - case OP_SCBRA: - number = GET2(ecode, 1+LINK_SIZE); - offset = number << 1; - + If there isn't enough space in the offset vector, treat this as if it were + a non-capturing bracket. Don't worry about setting the flag for the error + case here; that is handled in the code for KET. */ + + case OP_CBRA: + case OP_SCBRA: + number = GET2(ecode, 1+LINK_SIZE); + offset = number << 1; + #ifdef PCRE_DEBUG - printf("start bracket %d\n", number); - printf("subject="); - pchars(eptr, 16, TRUE, md); - printf("\n"); -#endif - - if (offset < md->offset_max) - { - save_offset1 = md->offset_vector[offset]; - save_offset2 = md->offset_vector[offset+1]; - save_offset3 = md->offset_vector[md->offset_end - number]; - save_capture_last = md->capture_last; + printf("start bracket %d\n", number); + printf("subject="); + pchars(eptr, 16, TRUE, md); + printf("\n"); +#endif + + if (offset < md->offset_max) + { + save_offset1 = md->offset_vector[offset]; + save_offset2 = md->offset_vector[offset+1]; + save_offset3 = md->offset_vector[md->offset_end - number]; + save_capture_last = md->capture_last; save_mark = md->mark; - - DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3)); + + DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3)); md->offset_vector[md->offset_end - number] = (int)(eptr - md->start_subject); - + for (;;) - { + { if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP; RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb, RM1); @@ -1012,36 +1012,36 @@ for (;;) /* Anything other than NOMATCH is passed back. */ if (rrc != MATCH_NOMATCH) RRETURN(rrc); - md->capture_last = save_capture_last; - ecode += GET(ecode, 1); + md->capture_last = save_capture_last; + ecode += GET(ecode, 1); md->mark = save_mark; if (*ecode != OP_ALT) break; - } - - DPRINTF(("bracket %d failed\n", number)); - md->offset_vector[offset] = save_offset1; - md->offset_vector[offset+1] = save_offset2; - md->offset_vector[md->offset_end - number] = save_offset3; - + } + + DPRINTF(("bracket %d failed\n", number)); + md->offset_vector[offset] = save_offset1; + md->offset_vector[offset+1] = save_offset2; + md->offset_vector[md->offset_end - number] = save_offset3; + /* At this point, rrc will be one of MATCH_ONCE or MATCH_NOMATCH. */ RRETURN(rrc); - } - - /* FALL THROUGH ... Insufficient room for saving captured contents. Treat - as a non-capturing bracket. */ - - /* VVVVVVVVVVVVVVVVVVVVVVVVV */ - /* VVVVVVVVVVVVVVVVVVVVVVVVV */ - - DPRINTF(("insufficient capture room: treat as non-capturing\n")); - - /* VVVVVVVVVVVVVVVVVVVVVVVVV */ - /* VVVVVVVVVVVVVVVVVVVVVVVVV */ - + } + + /* FALL THROUGH ... Insufficient room for saving captured contents. Treat + as a non-capturing bracket. */ + + /* VVVVVVVVVVVVVVVVVVVVVVVVV */ + /* VVVVVVVVVVVVVVVVVVVVVVVVV */ + + DPRINTF(("insufficient capture room: treat as non-capturing\n")); + + /* VVVVVVVVVVVVVVVVVVVVVVVVV */ + /* VVVVVVVVVVVVVVVVVVVVVVVVV */ + /* Non-capturing or atomic group, except for possessive with unlimited repeat and ONCE group with no captures. Loop for all the alternatives. - + When we get to the final alternative within the brackets, we used to return the result of a recursive call to match() whatever happened so it was possible to reduce stack usage by turning this into a tail recursion, @@ -1060,12 +1060,12 @@ for (;;) previous backup points can be taken. */ case OP_ONCE: - case OP_BRA: - case OP_SBRA: - DPRINTF(("start non-capturing bracket\n")); + case OP_BRA: + case OP_SBRA: + DPRINTF(("start non-capturing bracket\n")); - for (;;) - { + for (;;) + { if (op >= OP_SBRA || op == OP_ONCE) md->match_function_type = MATCH_CBEGROUP; @@ -1074,7 +1074,7 @@ for (;;) above. */ else if (!md->hasthen && ecode[GET(ecode, 1)] != OP_ALT) - { + { ecode += PRIV(OP_lengths)[*ecode]; goto TAIL_RECURSE; } @@ -1100,7 +1100,7 @@ for (;;) if (rrc != MATCH_NOMATCH) { if (rrc == MATCH_ONCE) - { + { const pcre_uchar *scode = ecode; if (*scode != OP_ONCE) /* If not at start, find it */ { @@ -1108,7 +1108,7 @@ for (;;) scode -= GET(scode, 1); } if (md->once_target == scode) rrc = MATCH_NOMATCH; - } + } RRETURN(rrc); } ecode += GET(ecode, 1); @@ -1116,9 +1116,9 @@ for (;;) if (*ecode != OP_ALT) break; md->capture_last = save_capture_last; } - + RRETURN(MATCH_NOMATCH); - + /* Handle possessive capturing brackets with an unlimited repeat. We come here from BRAZERO with allow_zero set TRUE. The offset_vector values are handled similarly to the normal case above. However, the matching is @@ -1184,11 +1184,11 @@ for (;;) } eptr = md->end_match_ptr; continue; - } - + } + /* See comment in the code for capturing groups above about handling THEN. */ - + if (rrc == MATCH_THEN) { next = ecode + GET(ecode,1); @@ -1199,9 +1199,9 @@ for (;;) if (rrc != MATCH_NOMATCH) RRETURN(rrc); md->capture_last = save_capture_last; - ecode += GET(ecode, 1); + ecode += GET(ecode, 1); if (*ecode != OP_ALT) break; - } + } if (!matched_once) { @@ -1276,15 +1276,15 @@ for (;;) } RRETURN(MATCH_NOMATCH); - /* Control never reaches here. */ - + /* Control never reaches here. */ + /* Conditional group: compilation checked that there are no more than two branches. If the condition is false, skipping the first branch takes us past the end of the item if there is only one branch, but that's exactly what we want. */ - - case OP_COND: - case OP_SCOND: + + case OP_COND: + case OP_SCOND: /* The variable codelink will be added to ecode when the condition is false, to get to the second branch. Setting it to the offset to the ALT @@ -1298,7 +1298,7 @@ for (;;) inserted between OP_COND and an assertion condition. */ if (*ecode == OP_CALLOUT) - { + { if (PUBL(callout) != NULL) { PUBL(callout_block) cb; @@ -1332,13 +1332,13 @@ for (;;) ecode += PRIV(OP_lengths)[OP_CALLOUT]; codelink -= PRIV(OP_lengths)[OP_CALLOUT]; - } - + } + /* Test the various possible conditions */ condition = FALSE; switch(condcode = *ecode) - { + { case OP_RREF: /* Numbered group recursion test */ if (md->recursive != NULL) /* Not recursing => FALSE */ { @@ -1364,9 +1364,9 @@ for (;;) case OP_CREF: /* Numbered group used test */ offset = GET2(ecode, 1) << 1; /* Doubled ref number */ - condition = offset < offset_top && md->offset_vector[offset] >= 0; + condition = offset < offset_top && md->offset_vector[offset] >= 0; break; - + case OP_DNCREF: /* Duplicate named group used test */ { int count = GET2(ecode, 1 + IMM2_SIZE); @@ -1380,11 +1380,11 @@ for (;;) } } break; - + case OP_DEF: /* DEFINE - always false */ case OP_FAIL: /* From optimized (?!) condition */ break; - + /* The condition is an assertion. Call match() to evaluate it - setting md->match_function_type to MATCH_CONDASSERT causes it to stop at the end of an assertion. */ @@ -1392,11 +1392,11 @@ for (;;) default: md->match_function_type = MATCH_CONDASSERT; RMATCH(eptr, ecode, offset_top, md, NULL, RM3); - if (rrc == MATCH_MATCH) - { + if (rrc == MATCH_MATCH) + { if (md->end_offset_top > offset_top) offset_top = md->end_offset_top; /* Captures may have happened */ - condition = TRUE; + condition = TRUE; /* Advance ecode past the assertion to the start of the first branch, but adjust it so that the general choosing code below works. If the @@ -1405,23 +1405,23 @@ for (;;) if (*ecode == OP_BRAZERO) ecode++; ecode += GET(ecode, 1); - while (*ecode == OP_ALT) ecode += GET(ecode, 1); + while (*ecode == OP_ALT) ecode += GET(ecode, 1); ecode += 1 + LINK_SIZE - PRIV(OP_lengths)[condcode]; - } + } /* PCRE doesn't allow the effect of (*THEN) to escape beyond an assertion; it is therefore treated as NOMATCH. Any other return is an error. */ - else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) - { - RRETURN(rrc); /* Need braces because of following else */ - } + else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) + { + RRETURN(rrc); /* Need braces because of following else */ + } break; - } - + } + /* Choose branch according to the condition */ - + ecode += condition? PRIV(OP_lengths)[condcode] : codelink; /* We are now at the branch that is to be obeyed. As there is only one, we @@ -1435,28 +1435,28 @@ for (;;) of alternatives, of course). */ if (condition || ecode[-(1+LINK_SIZE)] == OP_ALT) - { + { if (op != OP_SCOND) - { - goto TAIL_RECURSE; - } + { + goto TAIL_RECURSE; + } md->match_function_type = MATCH_CBEGROUP; RMATCH(eptr, ecode, offset_top, md, eptrb, RM49); RRETURN(rrc); - } + } /* Condition false & no alternative; continue after the group. */ else - { - } - break; - - + { + } + break; + + /* Before OP_ACCEPT there may be any number of OP_CLOSE opcodes, to close any currently open capturing brackets. */ - + case OP_CLOSE: number = GET2(ecode, 1); /* Must be less than 65536 */ offset = number << 1; @@ -1468,7 +1468,7 @@ for (;;) md->capture_last = (md->capture_last & OVFLMASK) | number; if (offset >= md->offset_max) md->capture_last |= OVFLBIT; else - { + { md->offset_vector[offset] = md->offset_vector[md->offset_end - number]; md->offset_vector[offset+1] = (int)(eptr - md->start_subject); @@ -1484,11 +1484,11 @@ for (;;) while (iptr < iend) *iptr++ = -1; offset_top = offset + 2; } - } + } ecode += 1 + IMM2_SIZE; break; - - + + /* End of the pattern, either real or forced. */ case OP_END: @@ -1509,29 +1509,29 @@ for (;;) /* Otherwise, we have a match. */ - md->end_match_ptr = eptr; /* Record where we ended */ - md->end_offset_top = offset_top; /* and how many extracts were taken */ - md->start_match_ptr = mstart; /* and the start (\K can modify) */ - + md->end_match_ptr = eptr; /* Record where we ended */ + md->end_offset_top = offset_top; /* and how many extracts were taken */ + md->start_match_ptr = mstart; /* and the start (\K can modify) */ + /* For some reason, the macros don't work properly if an expression is given as the argument to RRETURN when the heap is in use. */ - + rrc = (op == OP_END)? MATCH_MATCH : MATCH_ACCEPT; RRETURN(rrc); - - /* Assertion brackets. Check the alternative branches in turn - the - matching won't pass the KET for an assertion. If any one branch matches, - the assertion is true. Lookbehind assertions have an OP_REVERSE item at the - start of each branch to move the current point backwards, so the code at + + /* Assertion brackets. Check the alternative branches in turn - the + matching won't pass the KET for an assertion. If any one branch matches, + the assertion is true. Lookbehind assertions have an OP_REVERSE item at the + start of each branch to move the current point backwards, so the code at this level is identical to the lookahead case. When the assertion is part of a condition, we want to return immediately afterwards. The caller of this incarnation of the match() function will have set MATCH_CONDASSERT in md->match_function type, and one of these opcodes will be the first opcode that is processed. We use a local variable that is preserved over calls to match() to remember this case. */ - - case OP_ASSERT: - case OP_ASSERTBACK: + + case OP_ASSERT: + case OP_ASSERTBACK: save_mark = md->mark; if (md->match_function_type == MATCH_CONDASSERT) { @@ -1542,8 +1542,8 @@ for (;;) /* Loop for each branch */ - do - { + do + { RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM4); /* A match means that the assertion is true; break out of the loop @@ -1577,32 +1577,32 @@ for (;;) Perl. */ if (rrc != MATCH_NOMATCH) RRETURN(rrc); - ecode += GET(ecode, 1); - } + ecode += GET(ecode, 1); + } while (*ecode == OP_ALT); /* Continue for next alternative */ /* If we have tried all the alternative branches, the assertion has failed. If not, we broke out after a match. */ - if (*ecode == OP_KET) RRETURN(MATCH_NOMATCH); - - /* If checking an assertion for a condition, return MATCH_MATCH. */ - + if (*ecode == OP_KET) RRETURN(MATCH_NOMATCH); + + /* If checking an assertion for a condition, return MATCH_MATCH. */ + if (condassert) RRETURN(MATCH_MATCH); - + /* Continue from after a successful assertion, updating the offsets high water mark, since extracts may have been taken during the assertion. */ - - do ecode += GET(ecode,1); while (*ecode == OP_ALT); - ecode += 1 + LINK_SIZE; - offset_top = md->end_offset_top; - continue; - + + do ecode += GET(ecode,1); while (*ecode == OP_ALT); + ecode += 1 + LINK_SIZE; + offset_top = md->end_offset_top; + continue; + /* Negative assertion: all branches must fail to match for the assertion to succeed. */ - - case OP_ASSERT_NOT: - case OP_ASSERTBACK_NOT: + + case OP_ASSERT_NOT: + case OP_ASSERTBACK_NOT: save_mark = md->mark; if (md->match_function_type == MATCH_CONDASSERT) { @@ -1613,8 +1613,8 @@ for (;;) /* Loop for each alternative branch. */ - do - { + do + { RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5); md->mark = save_mark; /* Always restore the mark setting */ @@ -1660,63 +1660,63 @@ for (;;) /* Continue with next branch */ - ecode += GET(ecode,1); - } - while (*ecode == OP_ALT); - + ecode += GET(ecode,1); + } + while (*ecode == OP_ALT); + /* All branches in the assertion failed to match. */ - + NEG_ASSERT_TRUE: if (condassert) RRETURN(MATCH_MATCH); /* Condition assertion */ ecode += 1 + LINK_SIZE; /* Continue with current branch */ - continue; - - /* Move the subject pointer back. This occurs only at the start of - each branch of a lookbehind assertion. If we are too close to the start to - move back, this match function fails. When working with UTF-8 we move - back a number of characters, not bytes. */ - - case OP_REVERSE: + continue; + + /* Move the subject pointer back. This occurs only at the start of + each branch of a lookbehind assertion. If we are too close to the start to + move back, this match function fails. When working with UTF-8 we move + back a number of characters, not bytes. */ + + case OP_REVERSE: #ifdef SUPPORT_UTF if (utf) - { - i = GET(ecode, 1); - while (i-- > 0) - { - eptr--; - if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH); - BACKCHAR(eptr); - } - } - else -#endif - - /* No UTF-8 support, or not in UTF-8 mode: count is byte count */ - - { - eptr -= GET(ecode, 1); - if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH); - } - + { + i = GET(ecode, 1); + while (i-- > 0) + { + eptr--; + if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH); + BACKCHAR(eptr); + } + } + else +#endif + + /* No UTF-8 support, or not in UTF-8 mode: count is byte count */ + + { + eptr -= GET(ecode, 1); + if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH); + } + /* Save the earliest consulted character, then skip to next op code */ - + if (eptr < md->start_used_ptr) md->start_used_ptr = eptr; - ecode += 1 + LINK_SIZE; - break; - - /* The callout item calls an external function, if one is provided, passing - details of the match so far. This is mainly for debugging, though the - function is able to force a failure. */ - - case OP_CALLOUT: + ecode += 1 + LINK_SIZE; + break; + + /* The callout item calls an external function, if one is provided, passing + details of the match so far. This is mainly for debugging, though the + function is able to force a failure. */ + + case OP_CALLOUT: if (PUBL(callout) != NULL) - { + { PUBL(callout_block) cb; cb.version = 2; /* Version 1 of the callout block */ - cb.callout_number = ecode[1]; - cb.offset_vector = md->offset_vector; + cb.callout_number = ecode[1]; + cb.offset_vector = md->offset_vector; #if defined COMPILE_PCRE8 - cb.subject = (PCRE_SPTR)md->start_subject; + cb.subject = (PCRE_SPTR)md->start_subject; #elif defined COMPILE_PCRE16 cb.subject = (PCRE_SPTR16)md->start_subject; #elif defined COMPILE_PCRE32 @@ -1725,24 +1725,24 @@ for (;;) cb.subject_length = (int)(md->end_subject - md->start_subject); cb.start_match = (int)(mstart - md->start_subject); cb.current_position = (int)(eptr - md->start_subject); - cb.pattern_position = GET(ecode, 2); - cb.next_item_length = GET(ecode, 2 + LINK_SIZE); - cb.capture_top = offset_top/2; + cb.pattern_position = GET(ecode, 2); + cb.next_item_length = GET(ecode, 2 + LINK_SIZE); + cb.capture_top = offset_top/2; cb.capture_last = md->capture_last & CAPLMASK; /* Internal change requires this for API compatibility. */ if (cb.capture_last == 0) cb.capture_last = -1; - cb.callout_data = md->callout_data; + cb.callout_data = md->callout_data; cb.mark = md->nomatch_mark; if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH); - if (rrc < 0) RRETURN(rrc); - } - ecode += 2 + 2*LINK_SIZE; - break; - - /* Recursion either matches the current regex, or some subexpression. The - offset data is the offset to the starting bracket from the start of the - whole pattern. (This is so that it works from duplicated subpatterns.) - + if (rrc < 0) RRETURN(rrc); + } + ecode += 2 + 2*LINK_SIZE; + break; + + /* Recursion either matches the current regex, or some subexpression. The + offset data is the offset to the starting bracket from the start of the + whole pattern. (This is so that it works from duplicated subpatterns.) + The state of the capturing groups is preserved over recursion, and re-instated afterwards. We don't know how many are started and not yet finished (offset_top records the completed total) so we just have to save @@ -1750,21 +1750,21 @@ for (;;) large to put on the stack, but using malloc for small numbers seems expensive. As a compromise, the stack is used when there are no more than REC_STACK_SAVE_MAX values to store; otherwise malloc is used. - - There are also other values that have to be saved. We use a chained - sequence of blocks that actually live on the stack. Thanks to Robin Houston + + There are also other values that have to be saved. We use a chained + sequence of blocks that actually live on the stack. Thanks to Robin Houston for the original version of this logic. It has, however, been hacked around a lot, so he is not to blame for the current way it works. */ - - case OP_RECURSE: - { + + case OP_RECURSE: + { recursion_info *ri; unsigned int recno; - callpat = md->start_code + GET(ecode, 1); + callpat = md->start_code + GET(ecode, 1); recno = (callpat == md->start_code)? 0 : - GET2(callpat, 1 + LINK_SIZE); - + GET2(callpat, 1 + LINK_SIZE); + /* Check for repeating a recursion without advancing the subject pointer. This should catch convoluted mutual recursions. (Some simple cases are caught at compile time.) */ @@ -1773,41 +1773,41 @@ for (;;) if (recno == ri->group_num && eptr == ri->subject_position) RRETURN(PCRE_ERROR_RECURSELOOP); - /* Add to "recursing stack" */ - + /* Add to "recursing stack" */ + new_recursive.group_num = recno; new_recursive.saved_capture_last = md->capture_last; new_recursive.subject_position = eptr; - new_recursive.prevrec = md->recursive; - md->recursive = &new_recursive; - + new_recursive.prevrec = md->recursive; + md->recursive = &new_recursive; + /* Where to continue from afterwards */ - - ecode += 1 + LINK_SIZE; - + + ecode += 1 + LINK_SIZE; + /* Now save the offset data */ - - new_recursive.saved_max = md->offset_end; - if (new_recursive.saved_max <= REC_STACK_SAVE_MAX) - new_recursive.offset_save = stacksave; - else - { - new_recursive.offset_save = + + new_recursive.saved_max = md->offset_end; + if (new_recursive.saved_max <= REC_STACK_SAVE_MAX) + new_recursive.offset_save = stacksave; + else + { + new_recursive.offset_save = (int *)(PUBL(malloc))(new_recursive.saved_max * sizeof(int)); - if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY); - } - memcpy(new_recursive.offset_save, md->offset_vector, - new_recursive.saved_max * sizeof(int)); - + if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY); + } + memcpy(new_recursive.offset_save, md->offset_vector, + new_recursive.saved_max * sizeof(int)); + /* OK, now we can do the recursion. After processing each alternative, restore the offset data and the last captured value. If there were nested recursions, md->recursive might be changed, so reset it before looping. */ - - DPRINTF(("Recursing into group %d\n", new_recursive.group_num)); + + DPRINTF(("Recursing into group %d\n", new_recursive.group_num)); cbegroup = (*callpat >= OP_SBRA); - do - { + do + { if (cbegroup) md->match_function_type = MATCH_CBEGROUP; RMATCH(eptr, callpat + PRIV(OP_lengths)[*callpat], offset_top, md, eptrb, RM6); @@ -1816,9 +1816,9 @@ for (;;) md->capture_last = new_recursive.saved_capture_last; md->recursive = new_recursive.prevrec; if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) - { - DPRINTF(("Recursion matched\n")); - if (new_recursive.offset_save != stacksave) + { + DPRINTF(("Recursion matched\n")); + if (new_recursive.offset_save != stacksave) (PUBL(free))(new_recursive.offset_save); /* Set where we got to in the subject, and reset the start in case @@ -1828,14 +1828,14 @@ for (;;) eptr = md->end_match_ptr; mstart = md->start_match_ptr; goto RECURSION_MATCHED; /* Exit loop; end processing */ - } + } /* PCRE does not allow THEN, SKIP, PRUNE or COMMIT to escape beyond a recursion; they cause a NOMATCH for the entire recursion. These codes are defined in a range that can be tested for. */ if (rrc >= MATCH_BACKTRACK_MIN && rrc <= MATCH_BACKTRACK_MAX) - { + { if (new_recursive.offset_save != stacksave) (PUBL(free))(new_recursive.offset_save); RRETURN(MATCH_NOMATCH); @@ -1845,56 +1845,56 @@ for (;;) if (rrc != MATCH_NOMATCH) { - DPRINTF(("Recursion gave error %d\n", rrc)); + DPRINTF(("Recursion gave error %d\n", rrc)); if (new_recursive.offset_save != stacksave) (PUBL(free))(new_recursive.offset_save); - RRETURN(rrc); - } - - md->recursive = &new_recursive; - callpat += GET(callpat, 1); - } - while (*callpat == OP_ALT); - - DPRINTF(("Recursion didn't match\n")); - md->recursive = new_recursive.prevrec; - if (new_recursive.offset_save != stacksave) + RRETURN(rrc); + } + + md->recursive = &new_recursive; + callpat += GET(callpat, 1); + } + while (*callpat == OP_ALT); + + DPRINTF(("Recursion didn't match\n")); + md->recursive = new_recursive.prevrec; + if (new_recursive.offset_save != stacksave) (PUBL(free))(new_recursive.offset_save); - RRETURN(MATCH_NOMATCH); - } - + RRETURN(MATCH_NOMATCH); + } + RECURSION_MATCHED: break; - - /* An alternation is the end of a branch; scan along to find the end of the - bracketed group and go to there. */ - - case OP_ALT: - do ecode += GET(ecode,1); while (*ecode == OP_ALT); - break; - + + /* An alternation is the end of a branch; scan along to find the end of the + bracketed group and go to there. */ + + case OP_ALT: + do ecode += GET(ecode,1); while (*ecode == OP_ALT); + break; + /* BRAZERO, BRAMINZERO and SKIPZERO occur just before a bracket group, indicating that it may occur zero times. It may repeat infinitely, or not at all - i.e. it could be ()* or ()? or even (){0} in the pattern. Brackets with fixed upper repeat limits are compiled as a number of copies, with the optional ones preceded by BRAZERO or BRAMINZERO. */ - - case OP_BRAZERO: + + case OP_BRAZERO: next = ecode + 1; RMATCH(eptr, next, offset_top, md, eptrb, RM10); if (rrc != MATCH_NOMATCH) RRETURN(rrc); do next += GET(next, 1); while (*next == OP_ALT); ecode = next + 1 + LINK_SIZE; - break; - - case OP_BRAMINZERO: + break; + + case OP_BRAMINZERO: next = ecode + 1; do next += GET(next, 1); while (*next == OP_ALT); RMATCH(eptr, next + 1+LINK_SIZE, offset_top, md, eptrb, RM11); if (rrc != MATCH_NOMATCH) RRETURN(rrc); ecode++; - break; - + break; + case OP_SKIPZERO: next = ecode+1; do next += GET(next,1); while (*next == OP_ALT); @@ -1910,72 +1910,72 @@ for (;;) if (op == OP_CBRAPOS || op == OP_SCBRAPOS) goto POSSESSIVE_CAPTURE; goto POSSESSIVE_NON_CAPTURE; - /* End of a group, repeated or non-repeating. */ - - case OP_KET: - case OP_KETRMIN: - case OP_KETRMAX: + /* End of a group, repeated or non-repeating. */ + + case OP_KET: + case OP_KETRMIN: + case OP_KETRMAX: case OP_KETRPOS: - prev = ecode - GET(ecode, 1); - - /* If this was a group that remembered the subject start, in order to break - infinite repeats of empty string matches, retrieve the subject start from - the chain. Otherwise, set it NULL. */ - + prev = ecode - GET(ecode, 1); + + /* If this was a group that remembered the subject start, in order to break + infinite repeats of empty string matches, retrieve the subject start from + the chain. Otherwise, set it NULL. */ + if (*prev >= OP_SBRA || *prev == OP_ONCE) - { - saved_eptr = eptrb->epb_saved_eptr; /* Value at start of group */ - eptrb = eptrb->epb_prev; /* Backup to previous group */ - } - else saved_eptr = NULL; - + { + saved_eptr = eptrb->epb_saved_eptr; /* Value at start of group */ + eptrb = eptrb->epb_prev; /* Backup to previous group */ + } + else saved_eptr = NULL; + /* If we are at the end of an assertion group or a non-capturing atomic group, stop matching and return MATCH_MATCH, but record the current high water mark for use by positive assertions. We also need to record the match start in case it was changed by \K. */ - + if ((*prev >= OP_ASSERT && *prev <= OP_ASSERTBACK_NOT) || *prev == OP_ONCE_NC) - { + { md->end_match_ptr = eptr; /* For ONCE_NC */ - md->end_offset_top = offset_top; + md->end_offset_top = offset_top; md->start_match_ptr = mstart; RRETURN(MATCH_MATCH); /* Sets md->mark */ - } - - /* For capturing groups we have to check the group number back at the start - and if necessary complete handling an extraction by setting the offsets and + } + + /* For capturing groups we have to check the group number back at the start + and if necessary complete handling an extraction by setting the offsets and bumping the high water mark. Whole-pattern recursion is coded as a recurse into group 0, so it won't be picked up here. Instead, we catch it when the OP_END is reached. Other recursion is handled here. We just have to record the current subject position and start match pointer and give a MATCH return. */ - + if (*prev == OP_CBRA || *prev == OP_SCBRA || *prev == OP_CBRAPOS || *prev == OP_SCBRAPOS) - { - number = GET2(prev, 1+LINK_SIZE); - offset = number << 1; - + { + number = GET2(prev, 1+LINK_SIZE); + offset = number << 1; + #ifdef PCRE_DEBUG - printf("end bracket %d", number); - printf("\n"); -#endif - + printf("end bracket %d", number); + printf("\n"); +#endif + /* Handle a recursively called group. */ if (md->recursive != NULL && md->recursive->group_num == number) - { + { md->end_match_ptr = eptr; md->start_match_ptr = mstart; RRETURN(MATCH_MATCH); - } - + } + /* Deal with capturing */ - + md->capture_last = (md->capture_last & OVFLMASK) | number; if (offset >= md->offset_max) md->capture_last |= OVFLBIT; else - { + { /* If offset is greater than offset_top, it means that we are "skipping" a capturing group, and that group's offsets must be marked unset. In earlier versions of PCRE, all the offsets were unset at the @@ -1999,14 +1999,14 @@ for (;;) md->offset_vector[md->offset_end - number]; md->offset_vector[offset+1] = (int)(eptr - md->start_subject); if (offset_top <= offset) offset_top = offset + 2; - } - } - + } + } + /* OP_KETRPOS is a possessive repeating ket. Remember the current position, and return the MATCH_KETRPOS. This makes it possible to do the repeats one at a time from the outer level, thus saving stack. This must precede the empty string test - in this case that test is done at the outer level. */ - + if (*ecode == OP_KETRPOS) { md->start_match_ptr = mstart; /* In case \K reset it */ @@ -2014,7 +2014,7 @@ for (;;) md->end_offset_top = offset_top; RRETURN(MATCH_KETRPOS); } - + /* For an ordinary non-repeating ket, just continue at this level. This also happens for a repeating ket if no characters were matched in the group. This is the forcible breaking of infinite loops as implemented in @@ -2023,9 +2023,9 @@ for (;;) level. If this results in a NOMATCH return, pass MATCH_ONCE back to the original OP_ONCE level, thereby bypassing intermediate backup points, but resetting any captures that happened along the way. */ - - if (*ecode == OP_KET || eptr == saved_eptr) - { + + if (*ecode == OP_KET || eptr == saved_eptr) + { if (*prev == OP_ONCE) { RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM12); @@ -2034,21 +2034,21 @@ for (;;) RRETURN(MATCH_ONCE); } ecode += 1 + LINK_SIZE; /* Carry on at this level */ - break; - } - + break; + } + /* The normal repeating kets try the rest of the pattern or restart from the preceding bracket, in the appropriate order. In the second case, we can use tail recursion to avoid using another stack frame, unless we have an an atomic group or an unlimited repeat of a group that can match an empty string. */ - - if (*ecode == OP_KETRMIN) - { + + if (*ecode == OP_KETRMIN) + { RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM7); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (*prev == OP_ONCE) - { + { RMATCH(eptr, prev, offset_top, md, eptrb, RM8); if (rrc != MATCH_NOMATCH) RRETURN(rrc); md->once_target = prev; /* Level at which to change to MATCH_NOMATCH */ @@ -2057,16 +2057,16 @@ for (;;) if (*prev >= OP_SBRA) /* Could match an empty string */ { RMATCH(eptr, prev, offset_top, md, eptrb, RM50); - RRETURN(rrc); - } - ecode = prev; - goto TAIL_RECURSE; - } - else /* OP_KETRMAX */ - { + RRETURN(rrc); + } + ecode = prev; + goto TAIL_RECURSE; + } + else /* OP_KETRMAX */ + { RMATCH(eptr, prev, offset_top, md, eptrb, RM13); if (rrc == MATCH_ONCE && md->once_target == prev) rrc = MATCH_NOMATCH; - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (*prev == OP_ONCE) { RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM9); @@ -2074,23 +2074,23 @@ for (;;) md->once_target = prev; RRETURN(MATCH_ONCE); } - ecode += 1 + LINK_SIZE; - goto TAIL_RECURSE; - } - /* Control never gets here */ - + ecode += 1 + LINK_SIZE; + goto TAIL_RECURSE; + } + /* Control never gets here */ + /* Not multiline mode: start of subject assertion, unless notbol. */ - - case OP_CIRC: - if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH); - - /* Start of subject assertion */ - - case OP_SOD: - if (eptr != md->start_subject) RRETURN(MATCH_NOMATCH); - ecode++; - break; - + + case OP_CIRC: + if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH); + + /* Start of subject assertion */ + + case OP_SOD: + if (eptr != md->start_subject) RRETURN(MATCH_NOMATCH); + ecode++; + break; + /* Multiline mode: start of subject unless notbol, or after any newline. */ case OP_CIRCM: @@ -2101,26 +2101,26 @@ for (;;) ecode++; break; - /* Start of match assertion */ - - case OP_SOM: - if (eptr != md->start_subject + md->start_offset) RRETURN(MATCH_NOMATCH); - ecode++; - break; - - /* Reset the start of match point */ - - case OP_SET_SOM: - mstart = eptr; - ecode++; - break; - + /* Start of match assertion */ + + case OP_SOM: + if (eptr != md->start_subject + md->start_offset) RRETURN(MATCH_NOMATCH); + ecode++; + break; + + /* Reset the start of match point */ + + case OP_SET_SOM: + mstart = eptr; + ecode++; + break; + /* Multiline mode: assert before any newline, or before end of subject unless noteol is set. */ - + case OP_DOLLM: if (eptr < md->end_subject) - { + { if (!IS_NEWLINE(eptr)) { if (md->partial != 0 && @@ -2134,12 +2134,12 @@ for (;;) } RRETURN(MATCH_NOMATCH); } - } - else - { - if (md->noteol) RRETURN(MATCH_NOMATCH); + } + else + { + if (md->noteol) RRETURN(MATCH_NOMATCH); SCHECK_PARTIAL(); - } + } ecode++; break; @@ -2150,22 +2150,22 @@ for (;;) if (md->noteol) RRETURN(MATCH_NOMATCH); if (!md->endonly) goto ASSERT_NL_OR_EOS; - /* ... else fall through for endonly */ - - /* End of subject assertion (\z) */ - - case OP_EOD: - if (eptr < md->end_subject) RRETURN(MATCH_NOMATCH); + /* ... else fall through for endonly */ + + /* End of subject assertion (\z) */ + + case OP_EOD: + if (eptr < md->end_subject) RRETURN(MATCH_NOMATCH); SCHECK_PARTIAL(); - ecode++; - break; - - /* End of subject or ending \n assertion (\Z) */ - - case OP_EODN: + ecode++; + break; + + /* End of subject or ending \n assertion (\Z) */ + + case OP_EODN: ASSERT_NL_OR_EOS: if (eptr < md->end_subject && - (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen)) + (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen)) { if (md->partial != 0 && eptr + 1 >= md->end_subject && @@ -2176,37 +2176,37 @@ for (;;) md->hitend = TRUE; if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); } - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } /* Either at end of string or \n before end. */ SCHECK_PARTIAL(); - ecode++; - break; - - /* Word boundary assertions */ - - case OP_NOT_WORD_BOUNDARY: - case OP_WORD_BOUNDARY: - { - - /* Find out if the previous and current characters are "word" characters. - It takes a bit more work in UTF-8 mode. Characters > 255 are assumed to + ecode++; + break; + + /* Word boundary assertions */ + + case OP_NOT_WORD_BOUNDARY: + case OP_WORD_BOUNDARY: + { + + /* Find out if the previous and current characters are "word" characters. + It takes a bit more work in UTF-8 mode. Characters > 255 are assumed to be "non-word" characters. Remember the earliest consulted character for partial matching. */ - + #ifdef SUPPORT_UTF if (utf) - { + { /* Get status of previous character */ - if (eptr == md->start_subject) prev_is_word = FALSE; else - { + if (eptr == md->start_subject) prev_is_word = FALSE; else + { PCRE_PUCHAR lastptr = eptr - 1; BACKCHAR(lastptr); if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr; - GETCHAR(c, lastptr); + GETCHAR(c, lastptr); #ifdef SUPPORT_UCP if (md->use_ucp) { @@ -2218,19 +2218,19 @@ for (;;) } else #endif - prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0; - } + prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0; + } /* Get status of next character */ if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); cur_is_word = FALSE; } else { - GETCHAR(c, eptr); + GETCHAR(c, eptr); #ifdef SUPPORT_UCP if (md->use_ucp) { @@ -2242,16 +2242,16 @@ for (;;) } else #endif - cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0; - } - } - else -#endif - + cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0; + } + } + else +#endif + /* Not in UTF-8 mode, but we may still have PCRE_UCP set, and for consistency with the behaviour of \w we do use it in this case. */ - - { + + { /* Get status of previous character */ if (eptr == md->start_subject) prev_is_word = FALSE; else @@ -2295,30 +2295,30 @@ for (;;) #endif cur_is_word = MAX_255(*eptr) && ((md->ctypes[*eptr] & ctype_word) != 0); - } - - /* Now see if the situation is what we want */ - - if ((*ecode++ == OP_WORD_BOUNDARY)? - cur_is_word == prev_is_word : cur_is_word != prev_is_word) - RRETURN(MATCH_NOMATCH); - } - break; - + } + + /* Now see if the situation is what we want */ + + if ((*ecode++ == OP_WORD_BOUNDARY)? + cur_is_word == prev_is_word : cur_is_word != prev_is_word) + RRETURN(MATCH_NOMATCH); + } + break; + /* Match any single character type except newline; have to take care with CRLF newlines and partial matching. */ - - case OP_ANY: + + case OP_ANY: if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); if (md->partial != 0 && eptr == md->end_subject - 1 && NLBLOCK->nltype == NLTYPE_FIXED && NLBLOCK->nllen == 2 && UCHAR21TEST(eptr) == NLBLOCK->nl[0]) - { + { md->hitend = TRUE; if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); - } + } /* Fall through */ @@ -2334,134 +2334,134 @@ for (;;) #ifdef SUPPORT_UTF if (utf) ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++); #endif - ecode++; - break; - - /* Match a single byte, even in UTF-8 mode. This opcode really does match - any byte, even newline, independent of the setting of PCRE_DOTALL. */ - - case OP_ANYBYTE: + ecode++; + break; + + /* Match a single byte, even in UTF-8 mode. This opcode really does match + any byte, even newline, independent of the setting of PCRE_DOTALL. */ + + case OP_ANYBYTE: if (eptr >= md->end_subject) /* DO NOT merge the eptr++ here; it must */ { /* not be updated before SCHECK_PARTIAL. */ SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } eptr++; - ecode++; - break; - - case OP_NOT_DIGIT: + ecode++; + break; + + case OP_NOT_DIGIT: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - if ( + GETCHARINCTEST(c, eptr); + if ( #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8) - c < 256 && -#endif - (md->ctypes[c] & ctype_digit) != 0 - ) - RRETURN(MATCH_NOMATCH); - ecode++; - break; - - case OP_DIGIT: + c < 256 && +#endif + (md->ctypes[c] & ctype_digit) != 0 + ) + RRETURN(MATCH_NOMATCH); + ecode++; + break; + + case OP_DIGIT: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - if ( + GETCHARINCTEST(c, eptr); + if ( #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8) c > 255 || -#endif - (md->ctypes[c] & ctype_digit) == 0 - ) - RRETURN(MATCH_NOMATCH); - ecode++; - break; - - case OP_NOT_WHITESPACE: +#endif + (md->ctypes[c] & ctype_digit) == 0 + ) + RRETURN(MATCH_NOMATCH); + ecode++; + break; + + case OP_NOT_WHITESPACE: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - if ( + GETCHARINCTEST(c, eptr); + if ( #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8) - c < 256 && -#endif - (md->ctypes[c] & ctype_space) != 0 - ) - RRETURN(MATCH_NOMATCH); - ecode++; - break; - - case OP_WHITESPACE: + c < 256 && +#endif + (md->ctypes[c] & ctype_space) != 0 + ) + RRETURN(MATCH_NOMATCH); + ecode++; + break; + + case OP_WHITESPACE: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - if ( + GETCHARINCTEST(c, eptr); + if ( #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8) c > 255 || -#endif - (md->ctypes[c] & ctype_space) == 0 - ) - RRETURN(MATCH_NOMATCH); - ecode++; - break; - - case OP_NOT_WORDCHAR: +#endif + (md->ctypes[c] & ctype_space) == 0 + ) + RRETURN(MATCH_NOMATCH); + ecode++; + break; + + case OP_NOT_WORDCHAR: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - if ( + GETCHARINCTEST(c, eptr); + if ( #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8) - c < 256 && -#endif - (md->ctypes[c] & ctype_word) != 0 - ) - RRETURN(MATCH_NOMATCH); - ecode++; - break; - - case OP_WORDCHAR: + c < 256 && +#endif + (md->ctypes[c] & ctype_word) != 0 + ) + RRETURN(MATCH_NOMATCH); + ecode++; + break; + + case OP_WORDCHAR: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - if ( + GETCHARINCTEST(c, eptr); + if ( #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8) c > 255 || -#endif - (md->ctypes[c] & ctype_word) == 0 - ) - RRETURN(MATCH_NOMATCH); - ecode++; - break; - - case OP_ANYNL: +#endif + (md->ctypes[c] & ctype_word) == 0 + ) + RRETURN(MATCH_NOMATCH); + ecode++; + break; + + case OP_ANYNL: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - switch(c) - { - default: RRETURN(MATCH_NOMATCH); + GETCHARINCTEST(c, eptr); + switch(c) + { + default: RRETURN(MATCH_NOMATCH); case CHAR_CR: if (eptr >= md->end_subject) @@ -2469,128 +2469,128 @@ for (;;) SCHECK_PARTIAL(); } else if (UCHAR21TEST(eptr) == CHAR_LF) eptr++; - break; - + break; + case CHAR_LF: - break; - + break; + case CHAR_VT: case CHAR_FF: case CHAR_NEL: #ifndef EBCDIC - case 0x2028: - case 0x2029: + case 0x2028: + case 0x2029: #endif /* Not EBCDIC */ - if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH); - break; - } - ecode++; - break; - - case OP_NOT_HSPACE: + if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH); + break; + } + ecode++; + break; + + case OP_NOT_HSPACE: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - switch(c) - { + GETCHARINCTEST(c, eptr); + switch(c) + { HSPACE_CASES: RRETURN(MATCH_NOMATCH); /* Byte and multibyte cases */ - default: break; - } - ecode++; - break; - - case OP_HSPACE: + default: break; + } + ecode++; + break; + + case OP_HSPACE: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - switch(c) - { + GETCHARINCTEST(c, eptr); + switch(c) + { HSPACE_CASES: break; /* Byte and multibyte cases */ - default: RRETURN(MATCH_NOMATCH); - } - ecode++; - break; - - case OP_NOT_VSPACE: + default: RRETURN(MATCH_NOMATCH); + } + ecode++; + break; + + case OP_NOT_VSPACE: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - switch(c) - { + GETCHARINCTEST(c, eptr); + switch(c) + { VSPACE_CASES: RRETURN(MATCH_NOMATCH); - default: break; - } - ecode++; - break; - - case OP_VSPACE: + default: break; + } + ecode++; + break; + + case OP_VSPACE: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - switch(c) - { + GETCHARINCTEST(c, eptr); + switch(c) + { VSPACE_CASES: break; - default: RRETURN(MATCH_NOMATCH); - } - ecode++; - break; - -#ifdef SUPPORT_UCP - /* Check the next character by Unicode property. We will get here only - if the support is in the binary; otherwise a compile-time error occurs. */ - - case OP_PROP: - case OP_NOTPROP: + default: RRETURN(MATCH_NOMATCH); + } + ecode++; + break; + +#ifdef SUPPORT_UCP + /* Check the next character by Unicode property. We will get here only + if the support is in the binary; otherwise a compile-time error occurs. */ + + case OP_PROP: + case OP_NOTPROP: if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - { + GETCHARINCTEST(c, eptr); + { const pcre_uint32 *cp; const ucd_record *prop = GET_UCD(c); - - switch(ecode[1]) - { - case PT_ANY: - if (op == OP_NOTPROP) RRETURN(MATCH_NOMATCH); - break; - - case PT_LAMP: + + switch(ecode[1]) + { + case PT_ANY: + if (op == OP_NOTPROP) RRETURN(MATCH_NOMATCH); + break; + + case PT_LAMP: if ((prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt) == (op == OP_NOTPROP)) - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); break; - - case PT_GC: + + case PT_GC: if ((ecode[2] != PRIV(ucp_gentype)[prop->chartype]) == (op == OP_PROP)) - RRETURN(MATCH_NOMATCH); - break; - - case PT_PC: + RRETURN(MATCH_NOMATCH); + break; + + case PT_PC: if ((ecode[2] != prop->chartype) == (op == OP_PROP)) - RRETURN(MATCH_NOMATCH); - break; - - case PT_SC: + RRETURN(MATCH_NOMATCH); + break; + + case PT_SC: if ((ecode[2] != prop->script) == (op == OP_PROP)) - RRETURN(MATCH_NOMATCH); - break; - + RRETURN(MATCH_NOMATCH); + break; + /* These are specials */ case PT_ALNUM: @@ -2646,20 +2646,20 @@ for (;;) /* This should never occur */ - default: - RRETURN(PCRE_ERROR_INTERNAL); - } - - ecode += 3; - } - break; - - /* Match an extended Unicode sequence. We will get here only if the support - is in the binary; otherwise a compile-time error occurs. */ - - case OP_EXTUNI: + default: + RRETURN(PCRE_ERROR_INTERNAL); + } + + ecode += 3; + } + break; + + /* Match an extended Unicode sequence. We will get here only if the support + is in the binary; otherwise a compile-time error occurs. */ + + case OP_EXTUNI: if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } @@ -2668,30 +2668,30 @@ for (;;) int lgb, rgb; GETCHARINCTEST(c, eptr); lgb = UCD_GRAPHBREAK(c); - while (eptr < md->end_subject) - { - int len = 1; + while (eptr < md->end_subject) + { + int len = 1; if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); } rgb = UCD_GRAPHBREAK(c); if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; lgb = rgb; - eptr += len; - } - } + eptr += len; + } + } CHECK_PARTIAL(); - ecode++; - break; + ecode++; + break; #endif /* SUPPORT_UCP */ - - - /* Match a back reference, possibly repeatedly. Look past the end of the - item to see if there is repeat information following. The code is similar - to that for character classes, but repeated for efficiency. Then obey - similar code to character type repeats - written out again for speed. - However, if the referenced string is the empty string, always treat - it as matched, any number of times (otherwise there could be infinite + + + /* Match a back reference, possibly repeatedly. Look past the end of the + item to see if there is repeat information following. The code is similar + to that for character classes, but repeated for efficiency. Then obey + similar code to character type repeats - written out again for speed. + However, if the referenced string is the empty string, always treat + it as matched, any number of times (otherwise there could be infinite loops). If the reference is unset, there are two possibilities: - + (a) In the default, Perl-compatible state, set the length negative; this ensures that every attempt at a match fails. We can't just fail here, because of the possibility of quantifiers with zero minima. @@ -2710,19 +2710,19 @@ for (;;) case OP_DNREF: case OP_DNREFI: caseless = op == OP_DNREFI; - { + { int count = GET2(ecode, 1+IMM2_SIZE); pcre_uchar *slot = md->name_table + GET2(ecode, 1) * md->name_entry_size; ecode += 1 + 2*IMM2_SIZE; - + /* Setting the default length first and initializing 'offset' avoids compiler warnings in the REF_REPEAT code. */ - + length = (md->jscript_compat)? 0 : -1; offset = 0; - + while (count-- > 0) - { + { offset = GET2(slot, 0) << 1; if (offset < offset_top && md->offset_vector[offset] >= 0) { @@ -2733,7 +2733,7 @@ for (;;) } } goto REF_REPEAT; - + case OP_REF: case OP_REFI: caseless = op == OP_REFI; @@ -2743,7 +2743,7 @@ for (;;) length = (md->jscript_compat)? 0 : -1; else length = md->offset_vector[offset+1] - md->offset_vector[offset]; - + /* Set up for repetition, or handle the non-repeated case */ REF_REPEAT: @@ -2777,72 +2777,72 @@ for (;;) if (length == -2) eptr = md->end_subject; /* Partial match */ CHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); - } + } eptr += length; continue; /* With the main loop */ } - + /* Handle repeated back references. If the length of the reference is zero, just continue with the main loop. If the length is negative, it means the reference is unset in non-Java-compatible mode. If the minimum is zero, we can continue at the same level without recursion. For any other minimum, carrying on will result in NOMATCH. */ - + if (length == 0) continue; if (length < 0 && min == 0) continue; - + /* First, ensure the minimum number of matches are present. We get back the length of the reference string explicitly rather than passing the address of eptr, so that eptr can be a register variable. */ - + for (i = 1; i <= min; i++) { int slength; if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0) - { + { if (slength == -2) eptr = md->end_subject; /* Partial match */ CHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); - } + } eptr += slength; } - + /* If min = max, continue at the same level without recursion. They are not both allowed to be zero. */ - + if (min == max) continue; - + /* If minimizing, keep trying and advancing the pointer */ - + if (minimize) { for (fi = min;; fi++) - { + { int slength; RMATCH(eptr, ecode, offset_top, md, eptrb, RM14); if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0) - { + { if (slength == -2) eptr = md->end_subject; /* Partial match */ CHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); - } + } eptr += slength; - } + } /* Control never gets here */ } - + /* If maximizing, find the longest string and work backwards */ - + else { pp = eptr; for (i = min; i < max; i++) - { + { int slength; if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0) - { + { /* Can't use CHECK_PARTIAL because we don't want to update eptr in the soft partial matching case. */ @@ -2853,9 +2853,9 @@ for (;;) if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); } break; - } + } eptr += slength; - } + } while (eptr >= pp) { @@ -2864,97 +2864,97 @@ for (;;) eptr -= length; } RRETURN(MATCH_NOMATCH); - } - /* Control never gets here */ - - /* Match a bit-mapped character class, possibly repeatedly. This op code is - used when all the characters in the class have values in the range 0-255, - and either the matching is caseful, or the characters are in the range - 0-127 when UTF-8 processing is enabled. The only difference between - OP_CLASS and OP_NCLASS occurs when a data character outside the range is - encountered. - - First, look past the end of the item to see if there is repeat information - following. Then obey similar code to character type repeats - written out - again for speed. */ - - case OP_NCLASS: - case OP_CLASS: - { + } + /* Control never gets here */ + + /* Match a bit-mapped character class, possibly repeatedly. This op code is + used when all the characters in the class have values in the range 0-255, + and either the matching is caseful, or the characters are in the range + 0-127 when UTF-8 processing is enabled. The only difference between + OP_CLASS and OP_NCLASS occurs when a data character outside the range is + encountered. + + First, look past the end of the item to see if there is repeat information + following. Then obey similar code to character type repeats - written out + again for speed. */ + + case OP_NCLASS: + case OP_CLASS: + { /* The data variable is saved across frames, so the byte map needs to be stored there. */ #define BYTE_MAP ((pcre_uint8 *)data) - data = ecode + 1; /* Save for matching */ + data = ecode + 1; /* Save for matching */ ecode += 1 + (32 / sizeof(pcre_uchar)); /* Advance past the item */ - - switch (*ecode) - { - case OP_CRSTAR: - case OP_CRMINSTAR: - case OP_CRPLUS: - case OP_CRMINPLUS: - case OP_CRQUERY: - case OP_CRMINQUERY: + + switch (*ecode) + { + case OP_CRSTAR: + case OP_CRMINSTAR: + case OP_CRPLUS: + case OP_CRMINPLUS: + case OP_CRQUERY: + case OP_CRMINQUERY: case OP_CRPOSSTAR: case OP_CRPOSPLUS: case OP_CRPOSQUERY: - c = *ecode++ - OP_CRSTAR; + c = *ecode++ - OP_CRSTAR; if (c < OP_CRPOSSTAR - OP_CRSTAR) minimize = (c & 1) != 0; else possessive = TRUE; - min = rep_min[c]; /* Pick up values from tables; */ - max = rep_max[c]; /* zero for max => infinity */ - if (max == 0) max = INT_MAX; - break; - - case OP_CRRANGE: - case OP_CRMINRANGE: + min = rep_min[c]; /* Pick up values from tables; */ + max = rep_max[c]; /* zero for max => infinity */ + if (max == 0) max = INT_MAX; + break; + + case OP_CRRANGE: + case OP_CRMINRANGE: case OP_CRPOSRANGE: - minimize = (*ecode == OP_CRMINRANGE); + minimize = (*ecode == OP_CRMINRANGE); possessive = (*ecode == OP_CRPOSRANGE); - min = GET2(ecode, 1); + min = GET2(ecode, 1); max = GET2(ecode, 1 + IMM2_SIZE); - if (max == 0) max = INT_MAX; + if (max == 0) max = INT_MAX; ecode += 1 + 2 * IMM2_SIZE; - break; - - default: /* No repeat follows */ - min = max = 1; - break; - } - - /* First, ensure the minimum number of matches are present. */ - + break; + + default: /* No repeat follows */ + min = max = 1; + break; + } + + /* First, ensure the minimum number of matches are present. */ + #ifdef SUPPORT_UTF if (utf) - { - for (i = 1; i <= min; i++) - { + { + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(c, eptr); - if (c > 255) - { - if (op == OP_CLASS) RRETURN(MATCH_NOMATCH); - } - else + GETCHARINC(c, eptr); + if (c > 255) + { + if (op == OP_CLASS) RRETURN(MATCH_NOMATCH); + } + else if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH); - } - } - else -#endif + } + } + else +#endif /* Not UTF mode */ - { - for (i = 1; i <= min; i++) - { + { + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - c = *eptr++; + c = *eptr++; #ifndef COMPILE_PCRE8 if (c > 255) { @@ -2963,56 +2963,56 @@ for (;;) else #endif if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH); - } - } - - /* If max == min we can continue with the main loop without the - need to recurse. */ - - if (min == max) continue; - - /* If minimizing, keep testing the rest of the expression and advancing - the pointer while it matches the class. */ - - if (minimize) - { + } + } + + /* If max == min we can continue with the main loop without the + need to recurse. */ + + if (min == max) continue; + + /* If minimizing, keep testing the rest of the expression and advancing + the pointer while it matches the class. */ + + if (minimize) + { #ifdef SUPPORT_UTF if (utf) - { - for (fi = min;; fi++) - { + { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM16); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(c, eptr); - if (c > 255) - { - if (op == OP_CLASS) RRETURN(MATCH_NOMATCH); - } - else + GETCHARINC(c, eptr); + if (c > 255) + { + if (op == OP_CLASS) RRETURN(MATCH_NOMATCH); + } + else if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH); - } - } - else -#endif + } + } + else +#endif /* Not UTF mode */ - { - for (fi = min;; fi++) - { + { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM17); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - c = *eptr++; + c = *eptr++; #ifndef COMPILE_PCRE8 if (c > 255) { @@ -3021,60 +3021,60 @@ for (;;) else #endif if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH); - } - } - /* Control never gets here */ - } - - /* If maximizing, find the longest possible run, then work backwards. */ - - else - { - pp = eptr; - + } + } + /* Control never gets here */ + } + + /* If maximizing, find the longest possible run, then work backwards. */ + + else + { + pp = eptr; + #ifdef SUPPORT_UTF if (utf) - { - for (i = min; i < max; i++) - { - int len = 1; + { + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(c, eptr, len); - if (c > 255) - { - if (op == OP_CLASS) break; - } - else + GETCHARLEN(c, eptr, len); + if (c > 255) + { + if (op == OP_CLASS) break; + } + else if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break; - eptr += len; - } + eptr += len; + } if (possessive) continue; /* No backtracking */ - for (;;) - { + for (;;) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM18); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (eptr-- <= pp) break; /* Stop if tried at original pos */ - BACKCHAR(eptr); - } - } - else -#endif + BACKCHAR(eptr); + } + } + else +#endif /* Not UTF mode */ - { - for (i = min; i < max; i++) - { + { + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - c = *eptr; + c = *eptr; #ifndef COMPILE_PCRE8 if (c > 255) { @@ -3083,76 +3083,76 @@ for (;;) else #endif if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break; - eptr++; - } + eptr++; + } if (possessive) continue; /* No backtracking */ - while (eptr >= pp) - { + while (eptr >= pp) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM19); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); - eptr--; - } - } - - RRETURN(MATCH_NOMATCH); - } + if (rrc != MATCH_NOMATCH) RRETURN(rrc); + eptr--; + } + } + + RRETURN(MATCH_NOMATCH); + } #undef BYTE_MAP - } - /* Control never gets here */ - - + } + /* Control never gets here */ + + /* Match an extended character class. In the 8-bit library, this opcode is encountered only when UTF-8 mode mode is supported. In the 16-bit and 32-bit libraries, codepoints greater than 255 may be encountered even when UTF is not supported. */ - + #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 - case OP_XCLASS: - { - data = ecode + 1 + LINK_SIZE; /* Save for matching */ - ecode += GET(ecode, 1); /* Advance past the item */ - - switch (*ecode) - { - case OP_CRSTAR: - case OP_CRMINSTAR: - case OP_CRPLUS: - case OP_CRMINPLUS: - case OP_CRQUERY: - case OP_CRMINQUERY: + case OP_XCLASS: + { + data = ecode + 1 + LINK_SIZE; /* Save for matching */ + ecode += GET(ecode, 1); /* Advance past the item */ + + switch (*ecode) + { + case OP_CRSTAR: + case OP_CRMINSTAR: + case OP_CRPLUS: + case OP_CRMINPLUS: + case OP_CRQUERY: + case OP_CRMINQUERY: case OP_CRPOSSTAR: case OP_CRPOSPLUS: case OP_CRPOSQUERY: - c = *ecode++ - OP_CRSTAR; + c = *ecode++ - OP_CRSTAR; if (c < OP_CRPOSSTAR - OP_CRSTAR) minimize = (c & 1) != 0; else possessive = TRUE; - min = rep_min[c]; /* Pick up values from tables; */ - max = rep_max[c]; /* zero for max => infinity */ - if (max == 0) max = INT_MAX; - break; - - case OP_CRRANGE: - case OP_CRMINRANGE: + min = rep_min[c]; /* Pick up values from tables; */ + max = rep_max[c]; /* zero for max => infinity */ + if (max == 0) max = INT_MAX; + break; + + case OP_CRRANGE: + case OP_CRMINRANGE: case OP_CRPOSRANGE: - minimize = (*ecode == OP_CRMINRANGE); + minimize = (*ecode == OP_CRMINRANGE); possessive = (*ecode == OP_CRPOSRANGE); - min = GET2(ecode, 1); + min = GET2(ecode, 1); max = GET2(ecode, 1 + IMM2_SIZE); - if (max == 0) max = INT_MAX; + if (max == 0) max = INT_MAX; ecode += 1 + 2 * IMM2_SIZE; - break; - - default: /* No repeat follows */ - min = max = 1; - break; - } - - /* First, ensure the minimum number of matches are present. */ - - for (i = 1; i <= min; i++) - { + break; + + default: /* No repeat follows */ + min = max = 1; + break; + } + + /* First, ensure the minimum number of matches are present. */ + + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); @@ -3160,22 +3160,22 @@ for (;;) } GETCHARINCTEST(c, eptr); if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH); - } - - /* If max == min we can continue with the main loop without the - need to recurse. */ - - if (min == max) continue; - - /* If minimizing, keep testing the rest of the expression and advancing - the pointer while it matches the class. */ - - if (minimize) - { - for (fi = min;; fi++) - { + } + + /* If max == min we can continue with the main loop without the + need to recurse. */ + + if (min == max) continue; + + /* If minimizing, keep testing the rest of the expression and advancing + the pointer while it matches the class. */ + + if (minimize) + { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM20); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { @@ -3184,18 +3184,18 @@ for (;;) } GETCHARINCTEST(c, eptr); if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH); - } - /* Control never gets here */ - } - - /* If maximizing, find the longest possible run, then work backwards. */ - - else - { - pp = eptr; - for (i = min; i < max; i++) - { - int len = 1; + } + /* Control never gets here */ + } + + /* If maximizing, find the longest possible run, then work backwards. */ + + else + { + pp = eptr; + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); @@ -3207,63 +3207,63 @@ for (;;) c = *eptr; #endif if (!PRIV(xclass)(c, data, utf)) break; - eptr += len; - } + eptr += len; + } if (possessive) continue; /* No backtracking */ - for(;;) - { + for(;;) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM21); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (eptr-- <= pp) break; /* Stop if tried at original pos */ #ifdef SUPPORT_UTF if (utf) BACKCHAR(eptr); #endif - } - RRETURN(MATCH_NOMATCH); - } - - /* Control never gets here */ - } -#endif /* End of XCLASS */ - - /* Match a single character, casefully */ - - case OP_CHAR: + } + RRETURN(MATCH_NOMATCH); + } + + /* Control never gets here */ + } +#endif /* End of XCLASS */ + + /* Match a single character, casefully */ + + case OP_CHAR: #ifdef SUPPORT_UTF if (utf) - { - length = 1; - ecode++; - GETCHARLEN(fc, ecode, length); + { + length = 1; + ecode++; + GETCHARLEN(fc, ecode, length); if (length > md->end_subject - eptr) { CHECK_PARTIAL(); /* Not SCHECK_PARTIAL() */ RRETURN(MATCH_NOMATCH); } while (length-- > 0) if (*ecode++ != UCHAR21INC(eptr)) RRETURN(MATCH_NOMATCH); - } - else -#endif + } + else +#endif /* Not UTF mode */ - { + { if (md->end_subject - eptr < 1) { SCHECK_PARTIAL(); /* This one can use SCHECK_PARTIAL() */ RRETURN(MATCH_NOMATCH); } - if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH); - ecode += 2; - } - break; - + if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH); + ecode += 2; + } + break; + /* Match a single character, caselessly. If we are at the end of the subject, give up immediately. */ - + case OP_CHARI: if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } @@ -3271,128 +3271,128 @@ for (;;) #ifdef SUPPORT_UTF if (utf) { - length = 1; - ecode++; - GETCHARLEN(fc, ecode, length); - - /* If the pattern character's value is < 128, we have only one byte, and + length = 1; + ecode++; + GETCHARLEN(fc, ecode, length); + + /* If the pattern character's value is < 128, we have only one byte, and we know that its other case must also be one byte long, so we can use the fast lookup table. We know that there is at least one byte left in the subject. */ - - if (fc < 128) - { + + if (fc < 128) + { pcre_uint32 cc = UCHAR21(eptr); if (md->lcc[fc] != TABLE_GET(cc, md->lcc, cc)) RRETURN(MATCH_NOMATCH); ecode++; eptr++; - } - + } + /* Otherwise we must pick up the subject character. Note that we cannot use the value of "length" to check for sufficient bytes left, because the other case of the character may have more or fewer bytes. */ - - else - { + + else + { pcre_uint32 dc; - GETCHARINC(dc, eptr); - ecode += length; - - /* If we have Unicode property support, we can use it to test the other - case of the character, if there is one. */ - - if (fc != dc) - { -#ifdef SUPPORT_UCP + GETCHARINC(dc, eptr); + ecode += length; + + /* If we have Unicode property support, we can use it to test the other + case of the character, if there is one. */ + + if (fc != dc) + { +#ifdef SUPPORT_UCP if (dc != UCD_OTHERCASE(fc)) -#endif - RRETURN(MATCH_NOMATCH); - } - } - } - else +#endif + RRETURN(MATCH_NOMATCH); + } + } + } + else #endif /* SUPPORT_UTF */ - + /* Not UTF mode */ - { + { if (TABLE_GET(ecode[1], md->lcc, ecode[1]) != TABLE_GET(*eptr, md->lcc, *eptr)) RRETURN(MATCH_NOMATCH); eptr++; - ecode += 2; - } - break; - - /* Match a single character repeatedly. */ - - case OP_EXACT: + ecode += 2; + } + break; + + /* Match a single character repeatedly. */ + + case OP_EXACT: case OP_EXACTI: - min = max = GET2(ecode, 1); + min = max = GET2(ecode, 1); ecode += 1 + IMM2_SIZE; - goto REPEATCHAR; - - case OP_POSUPTO: + goto REPEATCHAR; + + case OP_POSUPTO: case OP_POSUPTOI: - possessive = TRUE; - /* Fall through */ - - case OP_UPTO: + possessive = TRUE; + /* Fall through */ + + case OP_UPTO: case OP_UPTOI: - case OP_MINUPTO: + case OP_MINUPTO: case OP_MINUPTOI: - min = 0; - max = GET2(ecode, 1); + min = 0; + max = GET2(ecode, 1); minimize = *ecode == OP_MINUPTO || *ecode == OP_MINUPTOI; ecode += 1 + IMM2_SIZE; - goto REPEATCHAR; - - case OP_POSSTAR: + goto REPEATCHAR; + + case OP_POSSTAR: case OP_POSSTARI: - possessive = TRUE; - min = 0; - max = INT_MAX; - ecode++; - goto REPEATCHAR; - - case OP_POSPLUS: + possessive = TRUE; + min = 0; + max = INT_MAX; + ecode++; + goto REPEATCHAR; + + case OP_POSPLUS: case OP_POSPLUSI: - possessive = TRUE; - min = 1; - max = INT_MAX; - ecode++; - goto REPEATCHAR; - - case OP_POSQUERY: + possessive = TRUE; + min = 1; + max = INT_MAX; + ecode++; + goto REPEATCHAR; + + case OP_POSQUERY: case OP_POSQUERYI: - possessive = TRUE; - min = 0; - max = 1; - ecode++; - goto REPEATCHAR; - - case OP_STAR: + possessive = TRUE; + min = 0; + max = 1; + ecode++; + goto REPEATCHAR; + + case OP_STAR: case OP_STARI: - case OP_MINSTAR: + case OP_MINSTAR: case OP_MINSTARI: - case OP_PLUS: + case OP_PLUS: case OP_PLUSI: - case OP_MINPLUS: + case OP_MINPLUS: case OP_MINPLUSI: - case OP_QUERY: + case OP_QUERY: case OP_QUERYI: - case OP_MINQUERY: + case OP_MINQUERY: case OP_MINQUERYI: c = *ecode++ - ((op < OP_STARI)? OP_STAR : OP_STARI); - minimize = (c & 1) != 0; - min = rep_min[c]; /* Pick up values from tables; */ - max = rep_max[c]; /* zero for max => infinity */ - if (max == 0) max = INT_MAX; - + minimize = (c & 1) != 0; + min = rep_min[c]; /* Pick up values from tables; */ + max = rep_max[c]; /* zero for max => infinity */ + if (max == 0) max = INT_MAX; + /* Common code for all repeated single-character matches. We first check for the minimum number of characters. If the minimum equals the maximum, we are done. Otherwise, if minimizing, check the rest of the pattern for a match; if there isn't one, advance up to the maximum, one character at a time. - + If maximizing, advance up to the maximum number of matching characters, until eptr is past the end of the maximum run. If possessive, we are then done (no backing up). Otherwise, match at this position; anything @@ -3404,128 +3404,128 @@ for (;;) The various UTF/non-UTF and caseful/caseless cases are handled separately, for speed. */ - REPEATCHAR: + REPEATCHAR: #ifdef SUPPORT_UTF if (utf) - { - length = 1; - charptr = ecode; - GETCHARLEN(fc, ecode, length); - ecode += length; - - /* Handle multibyte character matching specially here. There is - support for caseless matching if UCP support is present. */ - - if (length > 1) - { -#ifdef SUPPORT_UCP + { + length = 1; + charptr = ecode; + GETCHARLEN(fc, ecode, length); + ecode += length; + + /* Handle multibyte character matching specially here. There is + support for caseless matching if UCP support is present. */ + + if (length > 1) + { +#ifdef SUPPORT_UCP pcre_uint32 othercase; if (op >= OP_STARI && /* Caseless */ (othercase = UCD_OTHERCASE(fc)) != fc) oclength = PRIV(ord2utf)(othercase, occhars); - else oclength = 0; -#endif /* SUPPORT_UCP */ - - for (i = 1; i <= min; i++) - { + else oclength = 0; +#endif /* SUPPORT_UCP */ + + for (i = 1; i <= min; i++) + { if (eptr <= md->end_subject - length && memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length; -#ifdef SUPPORT_UCP +#ifdef SUPPORT_UCP else if (oclength > 0 && eptr <= md->end_subject - oclength && memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength; #endif /* SUPPORT_UCP */ - else - { + else + { CHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); - } - } - - if (min == max) continue; - - if (minimize) - { - for (fi = min;; fi++) - { + } + } + + if (min == max) continue; + + if (minimize) + { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM22); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr <= md->end_subject - length && memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length; -#ifdef SUPPORT_UCP +#ifdef SUPPORT_UCP else if (oclength > 0 && eptr <= md->end_subject - oclength && memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength; #endif /* SUPPORT_UCP */ - else - { + else + { CHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); - } - } - /* Control never gets here */ - } - - else /* Maximize */ - { - pp = eptr; - for (i = min; i < max; i++) - { + } + } + /* Control never gets here */ + } + + else /* Maximize */ + { + pp = eptr; + for (i = min; i < max; i++) + { if (eptr <= md->end_subject - length && memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length; -#ifdef SUPPORT_UCP +#ifdef SUPPORT_UCP else if (oclength > 0 && eptr <= md->end_subject - oclength && memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength; #endif /* SUPPORT_UCP */ - else - { + else + { CHECK_PARTIAL(); break; - } - } - + } + } + if (possessive) continue; /* No backtracking */ - for(;;) + for(;;) { if (eptr <= pp) goto TAIL_RECURSE; RMATCH(eptr, ecode, offset_top, md, eptrb, RM23); if (rrc != MATCH_NOMATCH) RRETURN(rrc); -#ifdef SUPPORT_UCP +#ifdef SUPPORT_UCP eptr--; BACKCHAR(eptr); -#else /* without SUPPORT_UCP */ +#else /* without SUPPORT_UCP */ eptr -= length; -#endif /* SUPPORT_UCP */ - } - } - /* Control never gets here */ - } - - /* If the length of a UTF-8 character is 1, we fall through here, and - obey the code as for non-UTF-8 characters below, though in this case the - value of fc will always be < 128. */ - } - else +#endif /* SUPPORT_UCP */ + } + } + /* Control never gets here */ + } + + /* If the length of a UTF-8 character is 1, we fall through here, and + obey the code as for non-UTF-8 characters below, though in this case the + value of fc will always be < 128. */ + } + else #endif /* SUPPORT_UTF */ /* When not in UTF-8 mode, load a single-byte character. */ - fc = *ecode++; - + fc = *ecode++; + /* The value of fc at this point is always one character, though we may or may not be in UTF mode. The code is duplicated for the caseless and - caseful cases, for speed, since matching characters is likely to be quite - common. First, ensure the minimum number of matches are present. If min = - max, continue at the same level without recursing. Otherwise, if - minimizing, keep trying the rest of the expression and advancing one - matching character if failing, up to the maximum. Alternatively, if - maximizing, find the maximum number of characters and work backwards. */ - - DPRINTF(("matching %c{%d,%d} against subject %.*s\n", fc, min, max, + caseful cases, for speed, since matching characters is likely to be quite + common. First, ensure the minimum number of matches are present. If min = + max, continue at the same level without recursing. Otherwise, if + minimizing, keep trying the rest of the expression and advancing one + matching character if failing, up to the maximum. Alternatively, if + maximizing, find the maximum number of characters and work backwards. */ + + DPRINTF(("matching %c{%d,%d} against subject %.*s\n", fc, min, max, max, (char *)eptr)); - + if (op >= OP_STARI) /* Caseless */ - { + { #ifdef COMPILE_PCRE8 /* fc must be < 128 if UTF is enabled. */ foc = md->fcc[fc]; @@ -3543,7 +3543,7 @@ for (;;) foc = TABLE_GET(fc, md->fcc, fc); #endif /* COMPILE_PCRE8 */ - for (i = 1; i <= min; i++) + for (i = 1; i <= min; i++) { pcre_uint32 cc; /* Faster than pcre_uchar */ if (eptr >= md->end_subject) @@ -3555,31 +3555,31 @@ for (;;) if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH); eptr++; } - if (min == max) continue; - if (minimize) - { - for (fi = min;; fi++) - { + if (min == max) continue; + if (minimize) + { + for (fi = min;; fi++) + { pcre_uint32 cc; /* Faster than pcre_uchar */ RMATCH(eptr, ecode, offset_top, md, eptrb, RM24); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } cc = UCHAR21TEST(eptr); if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH); eptr++; - } - /* Control never gets here */ - } - else /* Maximize */ - { - pp = eptr; - for (i = min; i < max; i++) - { + } + /* Control never gets here */ + } + else /* Maximize */ + { + pp = eptr; + for (i = min; i < max; i++) + { pcre_uint32 cc; /* Faster than pcre_uchar */ if (eptr >= md->end_subject) { @@ -3588,24 +3588,24 @@ for (;;) } cc = UCHAR21TEST(eptr); if (fc != cc && foc != cc) break; - eptr++; - } + eptr++; + } if (possessive) continue; /* No backtracking */ for (;;) - { + { if (eptr == pp) goto TAIL_RECURSE; RMATCH(eptr, ecode, offset_top, md, eptrb, RM25); - eptr--; - if (rrc != MATCH_NOMATCH) RRETURN(rrc); - } + eptr--; + if (rrc != MATCH_NOMATCH) RRETURN(rrc); + } /* Control never gets here */ - } - } - - /* Caseful comparisons (includes all multi-byte characters) */ - - else - { + } + } + + /* Caseful comparisons (includes all multi-byte characters) */ + + else + { for (i = 1; i <= min; i++) { if (eptr >= md->end_subject) @@ -3616,60 +3616,60 @@ for (;;) if (fc != UCHAR21INCTEST(eptr)) RRETURN(MATCH_NOMATCH); } - if (min == max) continue; + if (min == max) continue; - if (minimize) - { - for (fi = min;; fi++) - { + if (minimize) + { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM26); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } if (fc != UCHAR21INCTEST(eptr)) RRETURN(MATCH_NOMATCH); - } - /* Control never gets here */ - } - else /* Maximize */ - { - pp = eptr; - for (i = min; i < max; i++) - { + } + /* Control never gets here */ + } + else /* Maximize */ + { + pp = eptr; + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } if (fc != UCHAR21TEST(eptr)) break; - eptr++; - } + eptr++; + } if (possessive) continue; /* No backtracking */ for (;;) - { + { if (eptr == pp) goto TAIL_RECURSE; RMATCH(eptr, ecode, offset_top, md, eptrb, RM27); - eptr--; - if (rrc != MATCH_NOMATCH) RRETURN(rrc); - } + eptr--; + if (rrc != MATCH_NOMATCH) RRETURN(rrc); + } /* Control never gets here */ - } - } - /* Control never gets here */ - - /* Match a negated single one-byte character. The character we are - checking can be multibyte. */ - - case OP_NOT: + } + } + /* Control never gets here */ + + /* Match a negated single one-byte character. The character we are + checking can be multibyte. */ + + case OP_NOT: case OP_NOTI: if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); - } + } #ifdef SUPPORT_UTF if (utf) { @@ -3697,108 +3697,108 @@ for (;;) if (ch == c || och == c) RRETURN(MATCH_NOMATCH); } } - else + else #endif - { + { register pcre_uint32 ch = ecode[1]; c = *eptr++; if (ch == c || (op == OP_NOTI && TABLE_GET(ch, md->fcc, ch) == c)) RRETURN(MATCH_NOMATCH); ecode += 2; - } - break; - - /* Match a negated single one-byte character repeatedly. This is almost a - repeat of the code for a repeated single character, but I haven't found a - nice way of commoning these up that doesn't require a test of the - positive/negative option for each character match. Maybe that wouldn't add - very much to the time taken, but character matching *is* what this is all - about... */ - - case OP_NOTEXACT: + } + break; + + /* Match a negated single one-byte character repeatedly. This is almost a + repeat of the code for a repeated single character, but I haven't found a + nice way of commoning these up that doesn't require a test of the + positive/negative option for each character match. Maybe that wouldn't add + very much to the time taken, but character matching *is* what this is all + about... */ + + case OP_NOTEXACT: case OP_NOTEXACTI: - min = max = GET2(ecode, 1); + min = max = GET2(ecode, 1); ecode += 1 + IMM2_SIZE; - goto REPEATNOTCHAR; - - case OP_NOTUPTO: + goto REPEATNOTCHAR; + + case OP_NOTUPTO: case OP_NOTUPTOI: - case OP_NOTMINUPTO: + case OP_NOTMINUPTO: case OP_NOTMINUPTOI: - min = 0; - max = GET2(ecode, 1); + min = 0; + max = GET2(ecode, 1); minimize = *ecode == OP_NOTMINUPTO || *ecode == OP_NOTMINUPTOI; ecode += 1 + IMM2_SIZE; - goto REPEATNOTCHAR; - - case OP_NOTPOSSTAR: + goto REPEATNOTCHAR; + + case OP_NOTPOSSTAR: case OP_NOTPOSSTARI: - possessive = TRUE; - min = 0; - max = INT_MAX; - ecode++; - goto REPEATNOTCHAR; - - case OP_NOTPOSPLUS: + possessive = TRUE; + min = 0; + max = INT_MAX; + ecode++; + goto REPEATNOTCHAR; + + case OP_NOTPOSPLUS: case OP_NOTPOSPLUSI: - possessive = TRUE; - min = 1; - max = INT_MAX; - ecode++; - goto REPEATNOTCHAR; - - case OP_NOTPOSQUERY: + possessive = TRUE; + min = 1; + max = INT_MAX; + ecode++; + goto REPEATNOTCHAR; + + case OP_NOTPOSQUERY: case OP_NOTPOSQUERYI: - possessive = TRUE; - min = 0; - max = 1; - ecode++; - goto REPEATNOTCHAR; - - case OP_NOTPOSUPTO: + possessive = TRUE; + min = 0; + max = 1; + ecode++; + goto REPEATNOTCHAR; + + case OP_NOTPOSUPTO: case OP_NOTPOSUPTOI: - possessive = TRUE; - min = 0; - max = GET2(ecode, 1); + possessive = TRUE; + min = 0; + max = GET2(ecode, 1); ecode += 1 + IMM2_SIZE; - goto REPEATNOTCHAR; - - case OP_NOTSTAR: + goto REPEATNOTCHAR; + + case OP_NOTSTAR: case OP_NOTSTARI: - case OP_NOTMINSTAR: + case OP_NOTMINSTAR: case OP_NOTMINSTARI: - case OP_NOTPLUS: + case OP_NOTPLUS: case OP_NOTPLUSI: - case OP_NOTMINPLUS: + case OP_NOTMINPLUS: case OP_NOTMINPLUSI: - case OP_NOTQUERY: + case OP_NOTQUERY: case OP_NOTQUERYI: - case OP_NOTMINQUERY: + case OP_NOTMINQUERY: case OP_NOTMINQUERYI: c = *ecode++ - ((op >= OP_NOTSTARI)? OP_NOTSTARI: OP_NOTSTAR); - minimize = (c & 1) != 0; - min = rep_min[c]; /* Pick up values from tables; */ - max = rep_max[c]; /* zero for max => infinity */ - if (max == 0) max = INT_MAX; - + minimize = (c & 1) != 0; + min = rep_min[c]; /* Pick up values from tables; */ + max = rep_max[c]; /* zero for max => infinity */ + if (max == 0) max = INT_MAX; + /* Common code for all repeated single-byte matches. */ - - REPEATNOTCHAR: + + REPEATNOTCHAR: GETCHARINCTEST(fc, ecode); - - /* The code is duplicated for the caseless and caseful cases, for speed, - since matching characters is likely to be quite common. First, ensure the - minimum number of matches are present. If min = max, continue at the same - level without recursing. Otherwise, if minimizing, keep trying the rest of - the expression and advancing one matching character if failing, up to the - maximum. Alternatively, if maximizing, find the maximum number of - characters and work backwards. */ - - DPRINTF(("negative matching %c{%d,%d} against subject %.*s\n", fc, min, max, + + /* The code is duplicated for the caseless and caseful cases, for speed, + since matching characters is likely to be quite common. First, ensure the + minimum number of matches are present. If min = max, continue at the same + level without recursing. Otherwise, if minimizing, keep trying the rest of + the expression and advancing one matching character if failing, up to the + maximum. Alternatively, if maximizing, find the maximum number of + characters and work backwards. */ + + DPRINTF(("negative matching %c{%d,%d} against subject %.*s\n", fc, min, max, max, (char *)eptr)); - + if (op >= OP_NOTSTARI) /* Caseless */ - { + { #ifdef SUPPORT_UTF #ifdef SUPPORT_UCP if (utf && fc > 127) @@ -3810,27 +3810,27 @@ for (;;) else #endif /* SUPPORT_UTF */ foc = TABLE_GET(fc, md->fcc, fc); - + #ifdef SUPPORT_UTF if (utf) - { + { register pcre_uint32 d; - for (i = 1; i <= min; i++) - { + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(d, eptr); + GETCHARINC(d, eptr); if (fc == d || (unsigned int)foc == d) RRETURN(MATCH_NOMATCH); - } - } - else + } + } + else #endif /* SUPPORT_UTF */ /* Not UTF mode */ - { - for (i = 1; i <= min; i++) + { + for (i = 1; i <= min; i++) { if (eptr >= md->end_subject) { @@ -3840,407 +3840,407 @@ for (;;) if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH); eptr++; } - } - - if (min == max) continue; - - if (minimize) - { + } + + if (min == max) continue; + + if (minimize) + { #ifdef SUPPORT_UTF if (utf) - { + { register pcre_uint32 d; - for (fi = min;; fi++) - { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM28); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(d, eptr); + GETCHARINC(d, eptr); if (fc == d || (unsigned int)foc == d) RRETURN(MATCH_NOMATCH); - } - } - else + } + } + else #endif /*SUPPORT_UTF */ /* Not UTF mode */ - { - for (fi = min;; fi++) - { + { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM29); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH); eptr++; - } - } - /* Control never gets here */ - } - - /* Maximize case */ - - else - { - pp = eptr; - + } + } + /* Control never gets here */ + } + + /* Maximize case */ + + else + { + pp = eptr; + #ifdef SUPPORT_UTF if (utf) - { + { register pcre_uint32 d; - for (i = min; i < max; i++) - { - int len = 1; + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(d, eptr, len); + GETCHARLEN(d, eptr, len); if (fc == d || (unsigned int)foc == d) break; - eptr += len; - } + eptr += len; + } if (possessive) continue; /* No backtracking */ for(;;) - { + { if (eptr <= pp) goto TAIL_RECURSE; RMATCH(eptr, ecode, offset_top, md, eptrb, RM30); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); eptr--; - BACKCHAR(eptr); - } - } - else + BACKCHAR(eptr); + } + } + else #endif /* SUPPORT_UTF */ /* Not UTF mode */ - { - for (i = min; i < max; i++) - { + { + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } if (fc == *eptr || foc == *eptr) break; - eptr++; - } + eptr++; + } if (possessive) continue; /* No backtracking */ for (;;) - { + { if (eptr == pp) goto TAIL_RECURSE; RMATCH(eptr, ecode, offset_top, md, eptrb, RM31); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); - eptr--; - } - } + if (rrc != MATCH_NOMATCH) RRETURN(rrc); + eptr--; + } + } /* Control never gets here */ - } - } - - /* Caseful comparisons */ - - else - { + } + } + + /* Caseful comparisons */ + + else + { #ifdef SUPPORT_UTF if (utf) - { + { register pcre_uint32 d; - for (i = 1; i <= min; i++) - { + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(d, eptr); - if (fc == d) RRETURN(MATCH_NOMATCH); - } - } - else -#endif + GETCHARINC(d, eptr); + if (fc == d) RRETURN(MATCH_NOMATCH); + } + } + else +#endif /* Not UTF mode */ - { - for (i = 1; i <= min; i++) + { + for (i = 1; i <= min; i++) { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - if (fc == *eptr++) RRETURN(MATCH_NOMATCH); + if (fc == *eptr++) RRETURN(MATCH_NOMATCH); } - } - - if (min == max) continue; - - if (minimize) - { + } + + if (min == max) continue; + + if (minimize) + { #ifdef SUPPORT_UTF if (utf) - { + { register pcre_uint32 d; - for (fi = min;; fi++) - { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM32); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(d, eptr); + GETCHARINC(d, eptr); if (fc == d) RRETURN(MATCH_NOMATCH); - } - } - else -#endif + } + } + else +#endif /* Not UTF mode */ - { - for (fi = min;; fi++) - { + { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM33); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } if (fc == *eptr++) RRETURN(MATCH_NOMATCH); - } - } - /* Control never gets here */ - } - - /* Maximize case */ - - else - { - pp = eptr; - + } + } + /* Control never gets here */ + } + + /* Maximize case */ + + else + { + pp = eptr; + #ifdef SUPPORT_UTF if (utf) - { + { register pcre_uint32 d; - for (i = min; i < max; i++) - { - int len = 1; + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(d, eptr, len); - if (fc == d) break; - eptr += len; - } + GETCHARLEN(d, eptr, len); + if (fc == d) break; + eptr += len; + } if (possessive) continue; /* No backtracking */ - for(;;) - { + for(;;) + { if (eptr <= pp) goto TAIL_RECURSE; RMATCH(eptr, ecode, offset_top, md, eptrb, RM34); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); eptr--; - BACKCHAR(eptr); - } - } - else -#endif + BACKCHAR(eptr); + } + } + else +#endif /* Not UTF mode */ - { - for (i = min; i < max; i++) - { + { + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } if (fc == *eptr) break; - eptr++; - } + eptr++; + } if (possessive) continue; /* No backtracking */ for (;;) - { + { if (eptr == pp) goto TAIL_RECURSE; RMATCH(eptr, ecode, offset_top, md, eptrb, RM35); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); - eptr--; - } - } + if (rrc != MATCH_NOMATCH) RRETURN(rrc); + eptr--; + } + } /* Control never gets here */ - } - } - /* Control never gets here */ - - /* Match a single character type repeatedly; several different opcodes - share code. This is very similar to the code for single characters, but we - repeat it in the interests of efficiency. */ - - case OP_TYPEEXACT: - min = max = GET2(ecode, 1); - minimize = TRUE; + } + } + /* Control never gets here */ + + /* Match a single character type repeatedly; several different opcodes + share code. This is very similar to the code for single characters, but we + repeat it in the interests of efficiency. */ + + case OP_TYPEEXACT: + min = max = GET2(ecode, 1); + minimize = TRUE; ecode += 1 + IMM2_SIZE; - goto REPEATTYPE; - - case OP_TYPEUPTO: - case OP_TYPEMINUPTO: - min = 0; - max = GET2(ecode, 1); - minimize = *ecode == OP_TYPEMINUPTO; + goto REPEATTYPE; + + case OP_TYPEUPTO: + case OP_TYPEMINUPTO: + min = 0; + max = GET2(ecode, 1); + minimize = *ecode == OP_TYPEMINUPTO; ecode += 1 + IMM2_SIZE; - goto REPEATTYPE; - - case OP_TYPEPOSSTAR: - possessive = TRUE; - min = 0; - max = INT_MAX; - ecode++; - goto REPEATTYPE; - - case OP_TYPEPOSPLUS: - possessive = TRUE; - min = 1; - max = INT_MAX; - ecode++; - goto REPEATTYPE; - - case OP_TYPEPOSQUERY: - possessive = TRUE; - min = 0; - max = 1; - ecode++; - goto REPEATTYPE; - - case OP_TYPEPOSUPTO: - possessive = TRUE; - min = 0; - max = GET2(ecode, 1); + goto REPEATTYPE; + + case OP_TYPEPOSSTAR: + possessive = TRUE; + min = 0; + max = INT_MAX; + ecode++; + goto REPEATTYPE; + + case OP_TYPEPOSPLUS: + possessive = TRUE; + min = 1; + max = INT_MAX; + ecode++; + goto REPEATTYPE; + + case OP_TYPEPOSQUERY: + possessive = TRUE; + min = 0; + max = 1; + ecode++; + goto REPEATTYPE; + + case OP_TYPEPOSUPTO: + possessive = TRUE; + min = 0; + max = GET2(ecode, 1); ecode += 1 + IMM2_SIZE; - goto REPEATTYPE; - - case OP_TYPESTAR: - case OP_TYPEMINSTAR: - case OP_TYPEPLUS: - case OP_TYPEMINPLUS: - case OP_TYPEQUERY: - case OP_TYPEMINQUERY: - c = *ecode++ - OP_TYPESTAR; - minimize = (c & 1) != 0; - min = rep_min[c]; /* Pick up values from tables; */ - max = rep_max[c]; /* zero for max => infinity */ - if (max == 0) max = INT_MAX; - - /* Common code for all repeated single character type matches. Note that - in UTF-8 mode, '.' matches a character of any length, but for the other - character types, the valid characters are all one-byte long. */ - - REPEATTYPE: - ctype = *ecode++; /* Code for the character type */ - -#ifdef SUPPORT_UCP - if (ctype == OP_PROP || ctype == OP_NOTPROP) - { - prop_fail_result = ctype == OP_NOTPROP; - prop_type = *ecode++; - prop_value = *ecode++; - } - else prop_type = -1; -#endif - - /* First, ensure the minimum number of matches are present. Use inline - code for maximizing the speed, and do the type test once at the start + goto REPEATTYPE; + + case OP_TYPESTAR: + case OP_TYPEMINSTAR: + case OP_TYPEPLUS: + case OP_TYPEMINPLUS: + case OP_TYPEQUERY: + case OP_TYPEMINQUERY: + c = *ecode++ - OP_TYPESTAR; + minimize = (c & 1) != 0; + min = rep_min[c]; /* Pick up values from tables; */ + max = rep_max[c]; /* zero for max => infinity */ + if (max == 0) max = INT_MAX; + + /* Common code for all repeated single character type matches. Note that + in UTF-8 mode, '.' matches a character of any length, but for the other + character types, the valid characters are all one-byte long. */ + + REPEATTYPE: + ctype = *ecode++; /* Code for the character type */ + +#ifdef SUPPORT_UCP + if (ctype == OP_PROP || ctype == OP_NOTPROP) + { + prop_fail_result = ctype == OP_NOTPROP; + prop_type = *ecode++; + prop_value = *ecode++; + } + else prop_type = -1; +#endif + + /* First, ensure the minimum number of matches are present. Use inline + code for maximizing the speed, and do the type test once at the start (i.e. keep it out of the loop). Separate the UTF-8 code completely as that - is tidier. Also separate the UCP code, which can be the same for both UTF-8 - and single-bytes. */ - - if (min > 0) - { -#ifdef SUPPORT_UCP - if (prop_type >= 0) - { - switch(prop_type) - { - case PT_ANY: - if (prop_fail_result) RRETURN(MATCH_NOMATCH); - for (i = 1; i <= min; i++) - { + is tidier. Also separate the UCP code, which can be the same for both UTF-8 + and single-bytes. */ + + if (min > 0) + { +#ifdef SUPPORT_UCP + if (prop_type >= 0) + { + switch(prop_type) + { + case PT_ANY: + if (prop_fail_result) RRETURN(MATCH_NOMATCH); + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); - } - break; - - case PT_LAMP: - for (i = 1; i <= min; i++) - { + GETCHARINCTEST(c, eptr); + } + break; + + case PT_LAMP: + for (i = 1; i <= min; i++) + { int chartype; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); + GETCHARINCTEST(c, eptr); chartype = UCD_CHARTYPE(c); if ((chartype == ucp_Lu || chartype == ucp_Ll || chartype == ucp_Lt) == prop_fail_result) - RRETURN(MATCH_NOMATCH); - } - break; - - case PT_GC: - for (i = 1; i <= min; i++) - { + RRETURN(MATCH_NOMATCH); + } + break; + + case PT_GC: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); + GETCHARINCTEST(c, eptr); if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result) - RRETURN(MATCH_NOMATCH); - } - break; - - case PT_PC: - for (i = 1; i <= min; i++) - { + RRETURN(MATCH_NOMATCH); + } + break; + + case PT_PC: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); + GETCHARINCTEST(c, eptr); if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result) - RRETURN(MATCH_NOMATCH); - } - break; - - case PT_SC: - for (i = 1; i <= min; i++) - { + RRETURN(MATCH_NOMATCH); + } + break; + + case PT_SC: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINCTEST(c, eptr); + GETCHARINCTEST(c, eptr); if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result) - RRETURN(MATCH_NOMATCH); - } - break; - + RRETURN(MATCH_NOMATCH); + } + break; + case PT_ALNUM: for (i = 1; i <= min; i++) { @@ -4342,20 +4342,20 @@ for (;;) /* This should not occur */ - default: - RRETURN(PCRE_ERROR_INTERNAL); - } - } - - /* Match extended Unicode sequences. We will get here only if the - support is in the binary; otherwise a compile-time error occurs. */ - - else if (ctype == OP_EXTUNI) - { - for (i = 1; i <= min; i++) - { + default: + RRETURN(PCRE_ERROR_INTERNAL); + } + } + + /* Match extended Unicode sequences. We will get here only if the + support is in the binary; otherwise a compile-time error occurs. */ + + else if (ctype == OP_EXTUNI) + { + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } @@ -4365,34 +4365,34 @@ for (;;) GETCHARINCTEST(c, eptr); lgb = UCD_GRAPHBREAK(c); while (eptr < md->end_subject) - { + { int len = 1; if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); } rgb = UCD_GRAPHBREAK(c); if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; lgb = rgb; eptr += len; - } - } + } + } CHECK_PARTIAL(); - } - } - - else -#endif /* SUPPORT_UCP */ - -/* Handle all other cases when the coding is UTF-8 */ - + } + } + + else +#endif /* SUPPORT_UCP */ + +/* Handle all other cases when the coding is UTF-8 */ + #ifdef SUPPORT_UTF if (utf) switch(ctype) - { - case OP_ANY: - for (i = 1; i <= min; i++) - { + { + case OP_ANY: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); if (md->partial != 0 && @@ -4404,11 +4404,11 @@ for (;;) md->hitend = TRUE; if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); } - eptr++; + eptr++; ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++); - } - break; - + } + break; + case OP_ALLANY: for (i = 1; i <= min; i++) { @@ -4422,231 +4422,231 @@ for (;;) } break; - case OP_ANYBYTE: + case OP_ANYBYTE: if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH); - eptr += min; - break; - - case OP_ANYNL: - for (i = 1; i <= min; i++) - { + eptr += min; + break; + + case OP_ANYNL: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(c, eptr); - switch(c) - { - default: RRETURN(MATCH_NOMATCH); + GETCHARINC(c, eptr); + switch(c) + { + default: RRETURN(MATCH_NOMATCH); case CHAR_CR: if (eptr < md->end_subject && UCHAR21(eptr) == CHAR_LF) eptr++; - break; - + break; + case CHAR_LF: - break; - + break; + case CHAR_VT: case CHAR_FF: case CHAR_NEL: #ifndef EBCDIC - case 0x2028: - case 0x2029: + case 0x2028: + case 0x2029: #endif /* Not EBCDIC */ - if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH); - break; - } - } - break; - - case OP_NOT_HSPACE: - for (i = 1; i <= min; i++) - { + if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH); + break; + } + } + break; + + case OP_NOT_HSPACE: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(c, eptr); - switch(c) - { + GETCHARINC(c, eptr); + switch(c) + { HSPACE_CASES: RRETURN(MATCH_NOMATCH); /* Byte and multibyte cases */ - default: break; - } - } - break; - - case OP_HSPACE: - for (i = 1; i <= min; i++) - { + default: break; + } + } + break; + + case OP_HSPACE: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(c, eptr); - switch(c) - { + GETCHARINC(c, eptr); + switch(c) + { HSPACE_CASES: break; /* Byte and multibyte cases */ - default: RRETURN(MATCH_NOMATCH); - } - } - break; - - case OP_NOT_VSPACE: - for (i = 1; i <= min; i++) - { + default: RRETURN(MATCH_NOMATCH); + } + } + break; + + case OP_NOT_VSPACE: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(c, eptr); - switch(c) - { + GETCHARINC(c, eptr); + switch(c) + { VSPACE_CASES: RRETURN(MATCH_NOMATCH); - default: break; - } - } - break; - - case OP_VSPACE: - for (i = 1; i <= min; i++) - { + default: break; + } + } + break; + + case OP_VSPACE: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(c, eptr); - switch(c) - { + GETCHARINC(c, eptr); + switch(c) + { VSPACE_CASES: break; - default: RRETURN(MATCH_NOMATCH); - } - } - break; - - case OP_NOT_DIGIT: - for (i = 1; i <= min; i++) - { + default: RRETURN(MATCH_NOMATCH); + } + } + break; + + case OP_NOT_DIGIT: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - GETCHARINC(c, eptr); - if (c < 128 && (md->ctypes[c] & ctype_digit) != 0) - RRETURN(MATCH_NOMATCH); - } - break; - - case OP_DIGIT: - for (i = 1; i <= min; i++) - { + GETCHARINC(c, eptr); + if (c < 128 && (md->ctypes[c] & ctype_digit) != 0) + RRETURN(MATCH_NOMATCH); + } + break; + + case OP_DIGIT: + for (i = 1; i <= min; i++) + { pcre_uint32 cc; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } cc = UCHAR21(eptr); if (cc >= 128 || (md->ctypes[cc] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH); eptr++; - /* No need to skip more bytes - we know it's a 1-byte character */ - } - break; - - case OP_NOT_WHITESPACE: - for (i = 1; i <= min; i++) - { + /* No need to skip more bytes - we know it's a 1-byte character */ + } + break; + + case OP_NOT_WHITESPACE: + for (i = 1; i <= min; i++) + { pcre_uint32 cc; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } cc = UCHAR21(eptr); if (cc < 128 && (md->ctypes[cc] & ctype_space) != 0) RRETURN(MATCH_NOMATCH); eptr++; ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++); - } - break; - - case OP_WHITESPACE: - for (i = 1; i <= min; i++) - { + } + break; + + case OP_WHITESPACE: + for (i = 1; i <= min; i++) + { pcre_uint32 cc; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } cc = UCHAR21(eptr); if (cc >= 128 || (md->ctypes[cc] & ctype_space) == 0) RRETURN(MATCH_NOMATCH); eptr++; - /* No need to skip more bytes - we know it's a 1-byte character */ - } - break; - - case OP_NOT_WORDCHAR: - for (i = 1; i <= min; i++) - { + /* No need to skip more bytes - we know it's a 1-byte character */ + } + break; + + case OP_NOT_WORDCHAR: + for (i = 1; i <= min; i++) + { pcre_uint32 cc; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } cc = UCHAR21(eptr); if (cc < 128 && (md->ctypes[cc] & ctype_word) != 0) RRETURN(MATCH_NOMATCH); eptr++; ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++); - } - break; - - case OP_WORDCHAR: - for (i = 1; i <= min; i++) - { + } + break; + + case OP_WORDCHAR: + for (i = 1; i <= min; i++) + { pcre_uint32 cc; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } cc = UCHAR21(eptr); if (cc >= 128 || (md->ctypes[cc] & ctype_word) == 0) RRETURN(MATCH_NOMATCH); eptr++; - /* No need to skip more bytes - we know it's a 1-byte character */ - } - break; - - default: - RRETURN(PCRE_ERROR_INTERNAL); - } /* End switch(ctype) */ - - else + /* No need to skip more bytes - we know it's a 1-byte character */ + } + break; + + default: + RRETURN(PCRE_ERROR_INTERNAL); + } /* End switch(ctype) */ + + else #endif /* SUPPORT_UTF */ - - /* Code for the non-UTF-8 case for minimum matching of operators other + + /* Code for the non-UTF-8 case for minimum matching of operators other than OP_PROP and OP_NOTPROP. */ - - switch(ctype) - { - case OP_ANY: + + switch(ctype) + { + case OP_ANY: for (i = 1; i <= min; i++) - { + { if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); - } + } if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); if (md->partial != 0 && eptr + 1 >= md->end_subject && @@ -4658,9 +4658,9 @@ for (;;) if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); } eptr++; - } - break; - + } + break; + case OP_ALLANY: if (eptr > md->end_subject - min) { @@ -4670,34 +4670,34 @@ for (;;) eptr += min; break; - case OP_ANYBYTE: + case OP_ANYBYTE: if (eptr > md->end_subject - min) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - eptr += min; - break; - - case OP_ANYNL: - for (i = 1; i <= min; i++) - { + eptr += min; + break; + + case OP_ANYNL: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - switch(*eptr++) - { - default: RRETURN(MATCH_NOMATCH); + switch(*eptr++) + { + default: RRETURN(MATCH_NOMATCH); case CHAR_CR: if (eptr < md->end_subject && *eptr == CHAR_LF) eptr++; - break; + break; case CHAR_LF: - break; - + break; + case CHAR_VT: case CHAR_FF: case CHAR_NEL: @@ -4705,94 +4705,94 @@ for (;;) case 0x2028: case 0x2029: #endif - if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH); - break; - } - } - break; - - case OP_NOT_HSPACE: - for (i = 1; i <= min; i++) - { + if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH); + break; + } + } + break; + + case OP_NOT_HSPACE: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - switch(*eptr++) - { - default: break; + switch(*eptr++) + { + default: break; HSPACE_BYTE_CASES: #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 HSPACE_MULTIBYTE_CASES: #endif - RRETURN(MATCH_NOMATCH); - } - } - break; - - case OP_HSPACE: - for (i = 1; i <= min; i++) - { + RRETURN(MATCH_NOMATCH); + } + } + break; + + case OP_HSPACE: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - switch(*eptr++) - { - default: RRETURN(MATCH_NOMATCH); + switch(*eptr++) + { + default: RRETURN(MATCH_NOMATCH); HSPACE_BYTE_CASES: #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 HSPACE_MULTIBYTE_CASES: #endif - break; - } - } - break; - - case OP_NOT_VSPACE: - for (i = 1; i <= min; i++) - { + break; + } + } + break; + + case OP_NOT_VSPACE: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - switch(*eptr++) - { + switch(*eptr++) + { VSPACE_BYTE_CASES: #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 VSPACE_MULTIBYTE_CASES: #endif RRETURN(MATCH_NOMATCH); - default: break; - } - } - break; - - case OP_VSPACE: - for (i = 1; i <= min; i++) - { + default: break; + } + } + break; + + case OP_VSPACE: + for (i = 1; i <= min; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } - switch(*eptr++) - { - default: RRETURN(MATCH_NOMATCH); + switch(*eptr++) + { + default: RRETURN(MATCH_NOMATCH); VSPACE_BYTE_CASES: #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 VSPACE_MULTIBYTE_CASES: #endif - break; - } - } - break; - - case OP_NOT_DIGIT: - for (i = 1; i <= min; i++) + break; + } + } + break; + + case OP_NOT_DIGIT: + for (i = 1; i <= min; i++) { if (eptr >= md->end_subject) { @@ -4803,10 +4803,10 @@ for (;;) RRETURN(MATCH_NOMATCH); eptr++; } - break; - - case OP_DIGIT: - for (i = 1; i <= min; i++) + break; + + case OP_DIGIT: + for (i = 1; i <= min; i++) { if (eptr >= md->end_subject) { @@ -4817,10 +4817,10 @@ for (;;) RRETURN(MATCH_NOMATCH); eptr++; } - break; - - case OP_NOT_WHITESPACE: - for (i = 1; i <= min; i++) + break; + + case OP_NOT_WHITESPACE: + for (i = 1; i <= min; i++) { if (eptr >= md->end_subject) { @@ -4831,10 +4831,10 @@ for (;;) RRETURN(MATCH_NOMATCH); eptr++; } - break; - - case OP_WHITESPACE: - for (i = 1; i <= min; i++) + break; + + case OP_WHITESPACE: + for (i = 1; i <= min; i++) { if (eptr >= md->end_subject) { @@ -4845,61 +4845,61 @@ for (;;) RRETURN(MATCH_NOMATCH); eptr++; } - break; - - case OP_NOT_WORDCHAR: - for (i = 1; i <= min; i++) + break; + + case OP_NOT_WORDCHAR: + for (i = 1; i <= min; i++) { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_word) != 0) RRETURN(MATCH_NOMATCH); eptr++; } - break; - - case OP_WORDCHAR: - for (i = 1; i <= min; i++) + break; + + case OP_WORDCHAR: + for (i = 1; i <= min; i++) { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_word) == 0) RRETURN(MATCH_NOMATCH); eptr++; } - break; - - default: - RRETURN(PCRE_ERROR_INTERNAL); - } - } - - /* If min = max, continue at the same level without recursing */ - - if (min == max) continue; - - /* If minimizing, we have to test the rest of the pattern before each - subsequent match. Again, separate the UTF-8 case for speed, and also - separate the UCP cases. */ - - if (minimize) - { -#ifdef SUPPORT_UCP - if (prop_type >= 0) - { - switch(prop_type) - { - case PT_ANY: - for (fi = min;; fi++) - { + break; + + default: + RRETURN(PCRE_ERROR_INTERNAL); + } + } + + /* If min = max, continue at the same level without recursing */ + + if (min == max) continue; + + /* If minimizing, we have to test the rest of the pattern before each + subsequent match. Again, separate the UTF-8 case for speed, and also + separate the UCP cases. */ + + if (minimize) + { +#ifdef SUPPORT_UCP + if (prop_type >= 0) + { + switch(prop_type) + { + case PT_ANY: + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM36); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { @@ -4907,21 +4907,21 @@ for (;;) RRETURN(MATCH_NOMATCH); } GETCHARINCTEST(c, eptr); - if (prop_fail_result) RRETURN(MATCH_NOMATCH); - } - /* Control never gets here */ - - case PT_LAMP: - for (fi = min;; fi++) - { + if (prop_fail_result) RRETURN(MATCH_NOMATCH); + } + /* Control never gets here */ + + case PT_LAMP: + for (fi = min;; fi++) + { int chartype; RMATCH(eptr, ecode, offset_top, md, eptrb, RM37); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } GETCHARINCTEST(c, eptr); chartype = UCD_CHARTYPE(c); @@ -4929,60 +4929,60 @@ for (;;) chartype == ucp_Ll || chartype == ucp_Lt) == prop_fail_result) RRETURN(MATCH_NOMATCH); - } - /* Control never gets here */ - - case PT_GC: - for (fi = min;; fi++) - { + } + /* Control never gets here */ + + case PT_GC: + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM38); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } GETCHARINCTEST(c, eptr); if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result) RRETURN(MATCH_NOMATCH); - } - /* Control never gets here */ - - case PT_PC: - for (fi = min;; fi++) - { + } + /* Control never gets here */ + + case PT_PC: + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM39); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } GETCHARINCTEST(c, eptr); if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result) RRETURN(MATCH_NOMATCH); - } - /* Control never gets here */ - - case PT_SC: - for (fi = min;; fi++) - { + } + /* Control never gets here */ + + case PT_SC: + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM40); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } GETCHARINCTEST(c, eptr); if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result) RRETURN(MATCH_NOMATCH); - } - /* Control never gets here */ - + } + /* Control never gets here */ + case PT_ALNUM: for (fi = min;; fi++) { @@ -5100,23 +5100,23 @@ for (;;) /* Control never gets here */ /* This should never occur */ - default: - RRETURN(PCRE_ERROR_INTERNAL); - } - } - - /* Match extended Unicode sequences. We will get here only if the - support is in the binary; otherwise a compile-time error occurs. */ - - else if (ctype == OP_EXTUNI) - { - for (fi = min;; fi++) - { + default: + RRETURN(PCRE_ERROR_INTERNAL); + } + } + + /* Match extended Unicode sequences. We will get here only if the + support is in the binary; otherwise a compile-time error occurs. */ + + else if (ctype == OP_EXTUNI) + { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM41); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); RRETURN(MATCH_NOMATCH); } @@ -5126,39 +5126,39 @@ for (;;) GETCHARINCTEST(c, eptr); lgb = UCD_GRAPHBREAK(c); while (eptr < md->end_subject) - { + { int len = 1; if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); } rgb = UCD_GRAPHBREAK(c); if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; lgb = rgb; eptr += len; - } - } + } + } CHECK_PARTIAL(); - } - } - else -#endif /* SUPPORT_UCP */ - + } + } + else +#endif /* SUPPORT_UCP */ + #ifdef SUPPORT_UTF if (utf) - { - for (fi = min;; fi++) - { + { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM42); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } if (ctype == OP_ANY && IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); - GETCHARINC(c, eptr); - switch(ctype) - { + GETCHARINC(c, eptr); + switch(ctype) + { case OP_ANY: /* This is the non-NL case */ if (md->partial != 0 && /* Take care with CRLF partial */ eptr >= md->end_subject && @@ -5169,121 +5169,121 @@ for (;;) md->hitend = TRUE; if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); } - break; - + break; + case OP_ALLANY: - case OP_ANYBYTE: - break; - - case OP_ANYNL: - switch(c) - { - default: RRETURN(MATCH_NOMATCH); + case OP_ANYBYTE: + break; + + case OP_ANYNL: + switch(c) + { + default: RRETURN(MATCH_NOMATCH); case CHAR_CR: if (eptr < md->end_subject && UCHAR21(eptr) == CHAR_LF) eptr++; - break; + break; case CHAR_LF: - break; - + break; + case CHAR_VT: case CHAR_FF: case CHAR_NEL: #ifndef EBCDIC - case 0x2028: - case 0x2029: + case 0x2028: + case 0x2029: #endif /* Not EBCDIC */ - if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH); - break; - } - break; - - case OP_NOT_HSPACE: - switch(c) - { + if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH); + break; + } + break; + + case OP_NOT_HSPACE: + switch(c) + { HSPACE_CASES: RRETURN(MATCH_NOMATCH); - default: break; - } - break; - - case OP_HSPACE: - switch(c) - { + default: break; + } + break; + + case OP_HSPACE: + switch(c) + { HSPACE_CASES: break; - default: RRETURN(MATCH_NOMATCH); - } - break; - - case OP_NOT_VSPACE: - switch(c) - { + default: RRETURN(MATCH_NOMATCH); + } + break; + + case OP_NOT_VSPACE: + switch(c) + { VSPACE_CASES: RRETURN(MATCH_NOMATCH); - default: break; - } - break; - - case OP_VSPACE: - switch(c) - { + default: break; + } + break; + + case OP_VSPACE: + switch(c) + { VSPACE_CASES: break; - default: RRETURN(MATCH_NOMATCH); - } - break; - - case OP_NOT_DIGIT: - if (c < 256 && (md->ctypes[c] & ctype_digit) != 0) - RRETURN(MATCH_NOMATCH); - break; - - case OP_DIGIT: - if (c >= 256 || (md->ctypes[c] & ctype_digit) == 0) - RRETURN(MATCH_NOMATCH); - break; - - case OP_NOT_WHITESPACE: - if (c < 256 && (md->ctypes[c] & ctype_space) != 0) - RRETURN(MATCH_NOMATCH); - break; - - case OP_WHITESPACE: + default: RRETURN(MATCH_NOMATCH); + } + break; + + case OP_NOT_DIGIT: + if (c < 256 && (md->ctypes[c] & ctype_digit) != 0) + RRETURN(MATCH_NOMATCH); + break; + + case OP_DIGIT: + if (c >= 256 || (md->ctypes[c] & ctype_digit) == 0) + RRETURN(MATCH_NOMATCH); + break; + + case OP_NOT_WHITESPACE: + if (c < 256 && (md->ctypes[c] & ctype_space) != 0) + RRETURN(MATCH_NOMATCH); + break; + + case OP_WHITESPACE: if (c >= 256 || (md->ctypes[c] & ctype_space) == 0) - RRETURN(MATCH_NOMATCH); - break; - - case OP_NOT_WORDCHAR: - if (c < 256 && (md->ctypes[c] & ctype_word) != 0) - RRETURN(MATCH_NOMATCH); - break; - - case OP_WORDCHAR: - if (c >= 256 || (md->ctypes[c] & ctype_word) == 0) - RRETURN(MATCH_NOMATCH); - break; - - default: - RRETURN(PCRE_ERROR_INTERNAL); - } - } - } - else -#endif + RRETURN(MATCH_NOMATCH); + break; + + case OP_NOT_WORDCHAR: + if (c < 256 && (md->ctypes[c] & ctype_word) != 0) + RRETURN(MATCH_NOMATCH); + break; + + case OP_WORDCHAR: + if (c >= 256 || (md->ctypes[c] & ctype_word) == 0) + RRETURN(MATCH_NOMATCH); + break; + + default: + RRETURN(PCRE_ERROR_INTERNAL); + } + } + } + else +#endif /* Not UTF mode */ - { - for (fi = min;; fi++) - { + { + for (fi = min;; fi++) + { RMATCH(eptr, ecode, offset_top, md, eptrb, RM43); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); if (fi >= max) RRETURN(MATCH_NOMATCH); if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); } if (ctype == OP_ANY && IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); - c = *eptr++; - switch(ctype) - { + c = *eptr++; + switch(ctype) + { case OP_ANY: /* This is the non-NL case */ if (md->partial != 0 && /* Take care with CRLF partial */ eptr >= md->end_subject && @@ -5294,23 +5294,23 @@ for (;;) md->hitend = TRUE; if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); } - break; - + break; + case OP_ALLANY: - case OP_ANYBYTE: - break; - - case OP_ANYNL: - switch(c) - { - default: RRETURN(MATCH_NOMATCH); + case OP_ANYBYTE: + break; + + case OP_ANYNL: + switch(c) + { + default: RRETURN(MATCH_NOMATCH); case CHAR_CR: if (eptr < md->end_subject && *eptr == CHAR_LF) eptr++; - break; - + break; + case CHAR_LF: - break; - + break; + case CHAR_VT: case CHAR_FF: case CHAR_NEL: @@ -5318,128 +5318,128 @@ for (;;) case 0x2028: case 0x2029: #endif - if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH); - break; - } - break; - - case OP_NOT_HSPACE: - switch(c) - { - default: break; + if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH); + break; + } + break; + + case OP_NOT_HSPACE: + switch(c) + { + default: break; HSPACE_BYTE_CASES: #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 HSPACE_MULTIBYTE_CASES: #endif - RRETURN(MATCH_NOMATCH); - } - break; - - case OP_HSPACE: - switch(c) - { - default: RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); + } + break; + + case OP_HSPACE: + switch(c) + { + default: RRETURN(MATCH_NOMATCH); HSPACE_BYTE_CASES: #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 HSPACE_MULTIBYTE_CASES: #endif - break; - } - break; - - case OP_NOT_VSPACE: - switch(c) - { - default: break; + break; + } + break; + + case OP_NOT_VSPACE: + switch(c) + { + default: break; VSPACE_BYTE_CASES: #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 VSPACE_MULTIBYTE_CASES: #endif - RRETURN(MATCH_NOMATCH); - } - break; - - case OP_VSPACE: - switch(c) - { - default: RRETURN(MATCH_NOMATCH); + RRETURN(MATCH_NOMATCH); + } + break; + + case OP_VSPACE: + switch(c) + { + default: RRETURN(MATCH_NOMATCH); VSPACE_BYTE_CASES: #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 VSPACE_MULTIBYTE_CASES: #endif - break; - } - break; - - case OP_NOT_DIGIT: + break; + } + break; + + case OP_NOT_DIGIT: if (MAX_255(c) && (md->ctypes[c] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH); - break; - - case OP_DIGIT: + break; + + case OP_DIGIT: if (!MAX_255(c) || (md->ctypes[c] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH); - break; - - case OP_NOT_WHITESPACE: + break; + + case OP_NOT_WHITESPACE: if (MAX_255(c) && (md->ctypes[c] & ctype_space) != 0) RRETURN(MATCH_NOMATCH); - break; - - case OP_WHITESPACE: + break; + + case OP_WHITESPACE: if (!MAX_255(c) || (md->ctypes[c] & ctype_space) == 0) RRETURN(MATCH_NOMATCH); - break; - - case OP_NOT_WORDCHAR: + break; + + case OP_NOT_WORDCHAR: if (MAX_255(c) && (md->ctypes[c] & ctype_word) != 0) RRETURN(MATCH_NOMATCH); - break; - - case OP_WORDCHAR: + break; + + case OP_WORDCHAR: if (!MAX_255(c) || (md->ctypes[c] & ctype_word) == 0) RRETURN(MATCH_NOMATCH); - break; - - default: - RRETURN(PCRE_ERROR_INTERNAL); - } - } - } - /* Control never gets here */ - } - - /* If maximizing, it is worth using inline code for speed, doing the type - test once at the start (i.e. keep it out of the loop). Again, keep the - UTF-8 and UCP stuff separate. */ - - else - { - pp = eptr; /* Remember where we started */ - -#ifdef SUPPORT_UCP - if (prop_type >= 0) - { - switch(prop_type) - { - case PT_ANY: - for (i = min; i < max; i++) - { - int len = 1; + break; + + default: + RRETURN(PCRE_ERROR_INTERNAL); + } + } + } + /* Control never gets here */ + } + + /* If maximizing, it is worth using inline code for speed, doing the type + test once at the start (i.e. keep it out of the loop). Again, keep the + UTF-8 and UCP stuff separate. */ + + else + { + pp = eptr; /* Remember where we started */ + +#ifdef SUPPORT_UCP + if (prop_type >= 0) + { + switch(prop_type) + { + case PT_ANY: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } GETCHARLENTEST(c, eptr, len); - if (prop_fail_result) break; - eptr+= len; - } - break; - - case PT_LAMP: - for (i = min; i < max; i++) - { + if (prop_fail_result) break; + eptr+= len; + } + break; + + case PT_LAMP: + for (i = min; i < max; i++) + { int chartype; - int len = 1; + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } GETCHARLENTEST(c, eptr, len); chartype = UCD_CHARTYPE(c); @@ -5447,54 +5447,54 @@ for (;;) chartype == ucp_Ll || chartype == ucp_Lt) == prop_fail_result) break; - eptr+= len; - } - break; - - case PT_GC: - for (i = min; i < max; i++) - { - int len = 1; + eptr+= len; + } + break; + + case PT_GC: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } GETCHARLENTEST(c, eptr, len); if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result) break; - eptr+= len; - } - break; - - case PT_PC: - for (i = min; i < max; i++) - { - int len = 1; + eptr+= len; + } + break; + + case PT_PC: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } GETCHARLENTEST(c, eptr, len); if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result) break; - eptr+= len; - } - break; - - case PT_SC: - for (i = min; i < max; i++) - { - int len = 1; + eptr+= len; + } + break; + + case PT_SC: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } GETCHARLENTEST(c, eptr, len); if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result) break; - eptr+= len; - } - break; + eptr+= len; + } + break; case PT_ALNUM: for (i = min; i < max; i++) @@ -5609,30 +5609,30 @@ for (;;) default: RRETURN(PCRE_ERROR_INTERNAL); - } - - /* eptr is now past the end of the maximum run */ - + } + + /* eptr is now past the end of the maximum run */ + if (possessive) continue; /* No backtracking */ - for(;;) - { + for(;;) + { if (eptr <= pp) goto TAIL_RECURSE; RMATCH(eptr, ecode, offset_top, md, eptrb, RM44); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); eptr--; if (utf) BACKCHAR(eptr); - } - } - + } + } + /* Match extended Unicode grapheme clusters. We will get here only if the - support is in the binary; otherwise a compile-time error occurs. */ - - else if (ctype == OP_EXTUNI) - { - for (i = min; i < max; i++) - { + support is in the binary; otherwise a compile-time error occurs. */ + + else if (ctype == OP_EXTUNI) + { + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); break; } @@ -5642,20 +5642,20 @@ for (;;) GETCHARINCTEST(c, eptr); lgb = UCD_GRAPHBREAK(c); while (eptr < md->end_subject) - { + { int len = 1; if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); } rgb = UCD_GRAPHBREAK(c); if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; lgb = rgb; eptr += len; - } - } + } + } CHECK_PARTIAL(); - } - - /* eptr is now past the end of the maximum run */ - + } + + /* eptr is now past the end of the maximum run */ + if (possessive) continue; /* No backtracking */ /* We use <= pp rather than == pp to detect the start of the run while @@ -5663,14 +5663,14 @@ for (;;) move back past pp. This is just palliative; the use of \C in UTF mode is fraught with danger. */ - for(;;) - { + for(;;) + { int lgb, rgb; PCRE_PUCHAR fptr; if (eptr <= pp) goto TAIL_RECURSE; /* At start of char run */ RMATCH(eptr, ecode, offset_top, md, eptrb, RM45); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); /* Backtracking over an extended grapheme cluster involves inspecting the previous two characters (if present) to see if a break is @@ -5678,7 +5678,7 @@ for (;;) eptr--; if (!utf) c = *eptr; else - { + { BACKCHAR(eptr); GETCHAR(c, eptr); } @@ -5689,276 +5689,276 @@ for (;;) if (eptr <= pp) goto TAIL_RECURSE; /* At start of char run */ fptr = eptr - 1; if (!utf) c = *fptr; else - { + { BACKCHAR(fptr); GETCHAR(c, fptr); - } + } lgb = UCD_GRAPHBREAK(c); if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; eptr = fptr; rgb = lgb; - } - } - } - - else -#endif /* SUPPORT_UCP */ - + } + } + } + + else +#endif /* SUPPORT_UCP */ + #ifdef SUPPORT_UTF if (utf) - { - switch(ctype) - { - case OP_ANY: + { + switch(ctype) + { + case OP_ANY: for (i = min; i < max; i++) - { + { if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); break; - } + } if (IS_NEWLINE(eptr)) break; if (md->partial != 0 && /* Take care with CRLF partial */ eptr + 1 >= md->end_subject && NLBLOCK->nltype == NLTYPE_FIXED && NLBLOCK->nllen == 2 && UCHAR21(eptr) == NLBLOCK->nl[0]) - { + { md->hitend = TRUE; if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); - } + } eptr++; ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++); - } + } break; - + case OP_ALLANY: if (max < INT_MAX) - { + { for (i = min; i < max; i++) - { + { if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); break; - } + } eptr++; ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++); - } - } + } + } else { eptr = md->end_subject; /* Unlimited UTF-8 repeat */ SCHECK_PARTIAL(); } - break; - - /* The byte case is the same as non-UTF8 */ - - case OP_ANYBYTE: - c = max - min; - if (c > (unsigned int)(md->end_subject - eptr)) + break; + + /* The byte case is the same as non-UTF8 */ + + case OP_ANYBYTE: + c = max - min; + if (c > (unsigned int)(md->end_subject - eptr)) { eptr = md->end_subject; SCHECK_PARTIAL(); } else eptr += c; - break; - - case OP_ANYNL: - for (i = min; i < max; i++) - { - int len = 1; + break; + + case OP_ANYNL: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(c, eptr, len); + GETCHARLEN(c, eptr, len); if (c == CHAR_CR) - { - if (++eptr >= md->end_subject) break; + { + if (++eptr >= md->end_subject) break; if (UCHAR21(eptr) == CHAR_LF) eptr++; - } - else - { + } + else + { if (c != CHAR_LF && - (md->bsr_anycrlf || + (md->bsr_anycrlf || (c != CHAR_VT && c != CHAR_FF && c != CHAR_NEL #ifndef EBCDIC && c != 0x2028 && c != 0x2029 #endif /* Not EBCDIC */ ))) - break; - eptr += len; - } - } - break; - - case OP_NOT_HSPACE: - case OP_HSPACE: - for (i = min; i < max; i++) - { - BOOL gotspace; - int len = 1; + break; + eptr += len; + } + } + break; + + case OP_NOT_HSPACE: + case OP_HSPACE: + for (i = min; i < max; i++) + { + BOOL gotspace; + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(c, eptr, len); - switch(c) - { + GETCHARLEN(c, eptr, len); + switch(c) + { HSPACE_CASES: gotspace = TRUE; break; - default: gotspace = FALSE; break; - } - if (gotspace == (ctype == OP_NOT_HSPACE)) break; - eptr += len; - } - break; - - case OP_NOT_VSPACE: - case OP_VSPACE: - for (i = min; i < max; i++) - { - BOOL gotspace; - int len = 1; + default: gotspace = FALSE; break; + } + if (gotspace == (ctype == OP_NOT_HSPACE)) break; + eptr += len; + } + break; + + case OP_NOT_VSPACE: + case OP_VSPACE: + for (i = min; i < max; i++) + { + BOOL gotspace; + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(c, eptr, len); - switch(c) - { + GETCHARLEN(c, eptr, len); + switch(c) + { VSPACE_CASES: gotspace = TRUE; break; - default: gotspace = FALSE; break; - } - if (gotspace == (ctype == OP_NOT_VSPACE)) break; - eptr += len; - } - break; - - case OP_NOT_DIGIT: - for (i = min; i < max; i++) - { - int len = 1; + default: gotspace = FALSE; break; + } + if (gotspace == (ctype == OP_NOT_VSPACE)) break; + eptr += len; + } + break; + + case OP_NOT_DIGIT: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(c, eptr, len); - if (c < 256 && (md->ctypes[c] & ctype_digit) != 0) break; - eptr+= len; - } - break; - - case OP_DIGIT: - for (i = min; i < max; i++) - { - int len = 1; + GETCHARLEN(c, eptr, len); + if (c < 256 && (md->ctypes[c] & ctype_digit) != 0) break; + eptr+= len; + } + break; + + case OP_DIGIT: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(c, eptr, len); - if (c >= 256 ||(md->ctypes[c] & ctype_digit) == 0) break; - eptr+= len; - } - break; - - case OP_NOT_WHITESPACE: - for (i = min; i < max; i++) - { - int len = 1; + GETCHARLEN(c, eptr, len); + if (c >= 256 ||(md->ctypes[c] & ctype_digit) == 0) break; + eptr+= len; + } + break; + + case OP_NOT_WHITESPACE: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(c, eptr, len); - if (c < 256 && (md->ctypes[c] & ctype_space) != 0) break; - eptr+= len; - } - break; - - case OP_WHITESPACE: - for (i = min; i < max; i++) - { - int len = 1; + GETCHARLEN(c, eptr, len); + if (c < 256 && (md->ctypes[c] & ctype_space) != 0) break; + eptr+= len; + } + break; + + case OP_WHITESPACE: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(c, eptr, len); - if (c >= 256 ||(md->ctypes[c] & ctype_space) == 0) break; - eptr+= len; - } - break; - - case OP_NOT_WORDCHAR: - for (i = min; i < max; i++) - { - int len = 1; + GETCHARLEN(c, eptr, len); + if (c >= 256 ||(md->ctypes[c] & ctype_space) == 0) break; + eptr+= len; + } + break; + + case OP_NOT_WORDCHAR: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(c, eptr, len); - if (c < 256 && (md->ctypes[c] & ctype_word) != 0) break; - eptr+= len; - } - break; - - case OP_WORDCHAR: - for (i = min; i < max; i++) - { - int len = 1; + GETCHARLEN(c, eptr, len); + if (c < 256 && (md->ctypes[c] & ctype_word) != 0) break; + eptr+= len; + } + break; + + case OP_WORDCHAR: + for (i = min; i < max; i++) + { + int len = 1; if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - GETCHARLEN(c, eptr, len); - if (c >= 256 || (md->ctypes[c] & ctype_word) == 0) break; - eptr+= len; - } - break; - - default: - RRETURN(PCRE_ERROR_INTERNAL); - } - + GETCHARLEN(c, eptr, len); + if (c >= 256 || (md->ctypes[c] & ctype_word) == 0) break; + eptr+= len; + } + break; + + default: + RRETURN(PCRE_ERROR_INTERNAL); + } + if (possessive) continue; /* No backtracking */ - for(;;) - { + for(;;) + { if (eptr <= pp) goto TAIL_RECURSE; RMATCH(eptr, ecode, offset_top, md, eptrb, RM46); - if (rrc != MATCH_NOMATCH) RRETURN(rrc); + if (rrc != MATCH_NOMATCH) RRETURN(rrc); eptr--; - BACKCHAR(eptr); + BACKCHAR(eptr); if (ctype == OP_ANYNL && eptr > pp && UCHAR21(eptr) == CHAR_NL && UCHAR21(eptr - 1) == CHAR_CR) eptr--; - } - } - else + } + } + else #endif /* SUPPORT_UTF */ /* Not UTF mode */ - { - switch(ctype) - { - case OP_ANY: + { + switch(ctype) + { + case OP_ANY: for (i = min; i < max; i++) - { + { if (eptr >= md->end_subject) - { + { SCHECK_PARTIAL(); break; - } + } if (IS_NEWLINE(eptr)) break; if (md->partial != 0 && /* Take care with CRLF partial */ eptr + 1 >= md->end_subject && @@ -5970,50 +5970,50 @@ for (;;) if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); } eptr++; - } + } break; - + case OP_ALLANY: - case OP_ANYBYTE: - c = max - min; - if (c > (unsigned int)(md->end_subject - eptr)) + case OP_ANYBYTE: + c = max - min; + if (c > (unsigned int)(md->end_subject - eptr)) { eptr = md->end_subject; SCHECK_PARTIAL(); } else eptr += c; - break; - - case OP_ANYNL: - for (i = min; i < max; i++) - { + break; + + case OP_ANYNL: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); break; } - c = *eptr; + c = *eptr; if (c == CHAR_CR) - { - if (++eptr >= md->end_subject) break; + { + if (++eptr >= md->end_subject) break; if (*eptr == CHAR_LF) eptr++; - } - else - { + } + else + { if (c != CHAR_LF && (md->bsr_anycrlf || (c != CHAR_VT && c != CHAR_FF && c != CHAR_NEL #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 && c != 0x2028 && c != 0x2029 #endif ))) break; - eptr++; - } - } - break; - - case OP_NOT_HSPACE: - for (i = min; i < max; i++) - { + eptr++; + } + } + break; + + case OP_NOT_HSPACE: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); @@ -6028,13 +6028,13 @@ for (;;) #endif goto ENDLOOP00; } - } + } ENDLOOP00: - break; - - case OP_HSPACE: - for (i = min; i < max; i++) - { + break; + + case OP_HSPACE: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); @@ -6049,17 +6049,17 @@ for (;;) #endif eptr++; break; } - } + } ENDLOOP01: - break; - - case OP_NOT_VSPACE: - for (i = min; i < max; i++) - { + break; + + case OP_NOT_VSPACE: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } switch(*eptr) { @@ -6070,17 +6070,17 @@ for (;;) #endif goto ENDLOOP02; } - } + } ENDLOOP02: - break; - - case OP_VSPACE: - for (i = min; i < max; i++) - { + break; + + case OP_VSPACE: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } switch(*eptr) { @@ -6091,136 +6091,136 @@ for (;;) #endif eptr++; break; } - } + } ENDLOOP03: - break; - - case OP_NOT_DIGIT: - for (i = min; i < max; i++) - { + break; + + case OP_NOT_DIGIT: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_digit) != 0) break; - eptr++; - } - break; - - case OP_DIGIT: - for (i = min; i < max; i++) - { + eptr++; + } + break; + + case OP_DIGIT: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_digit) == 0) break; - eptr++; - } - break; - - case OP_NOT_WHITESPACE: - for (i = min; i < max; i++) - { + eptr++; + } + break; + + case OP_NOT_WHITESPACE: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_space) != 0) break; - eptr++; - } - break; - - case OP_WHITESPACE: - for (i = min; i < max; i++) - { + eptr++; + } + break; + + case OP_WHITESPACE: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_space) == 0) break; - eptr++; - } - break; - - case OP_NOT_WORDCHAR: - for (i = min; i < max; i++) - { + eptr++; + } + break; + + case OP_NOT_WORDCHAR: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_word) != 0) break; - eptr++; - } - break; - - case OP_WORDCHAR: - for (i = min; i < max; i++) - { + eptr++; + } + break; + + case OP_WORDCHAR: + for (i = min; i < max; i++) + { if (eptr >= md->end_subject) { SCHECK_PARTIAL(); - break; + break; } if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_word) == 0) break; - eptr++; - } - break; - - default: - RRETURN(PCRE_ERROR_INTERNAL); - } - + eptr++; + } + break; + + default: + RRETURN(PCRE_ERROR_INTERNAL); + } + if (possessive) continue; /* No backtracking */ for (;;) - { + { if (eptr == pp) goto TAIL_RECURSE; RMATCH(eptr, ecode, offset_top, md, eptrb, RM47); if (rrc != MATCH_NOMATCH) RRETURN(rrc); - eptr--; + eptr--; if (ctype == OP_ANYNL && eptr > pp && *eptr == CHAR_LF && eptr[-1] == CHAR_CR) eptr--; - } - } - + } + } + /* Control never gets here */ - } - - /* There's been some horrible disaster. Arrival here can only mean there is - something seriously wrong in the code above or the OP_xxx definitions. */ - - default: - DPRINTF(("Unknown opcode %d\n", *ecode)); - RRETURN(PCRE_ERROR_UNKNOWN_OPCODE); - } - - /* Do not stick any code in here without much thought; it is assumed - that "continue" in the code above comes out to here to repeat the main - loop. */ - - } /* End of main loop */ -/* Control never reaches here */ - - -/* When compiling to use the heap rather than the stack for recursive calls to -match(), the RRETURN() macro jumps here. The number that is saved in -frame->Xwhere indicates which label we actually want to return to. */ - -#ifdef NO_RECURSE -#define LBL(val) case val: goto L_RM##val; -HEAP_RETURN: -switch (frame->Xwhere) - { - LBL( 1) LBL( 2) LBL( 3) LBL( 4) LBL( 5) LBL( 6) LBL( 7) LBL( 8) - LBL( 9) LBL(10) LBL(11) LBL(12) LBL(13) LBL(14) LBL(15) LBL(17) - LBL(19) LBL(24) LBL(25) LBL(26) LBL(27) LBL(29) LBL(31) LBL(33) - LBL(35) LBL(43) LBL(47) LBL(48) LBL(49) LBL(50) LBL(51) LBL(52) + } + + /* There's been some horrible disaster. Arrival here can only mean there is + something seriously wrong in the code above or the OP_xxx definitions. */ + + default: + DPRINTF(("Unknown opcode %d\n", *ecode)); + RRETURN(PCRE_ERROR_UNKNOWN_OPCODE); + } + + /* Do not stick any code in here without much thought; it is assumed + that "continue" in the code above comes out to here to repeat the main + loop. */ + + } /* End of main loop */ +/* Control never reaches here */ + + +/* When compiling to use the heap rather than the stack for recursive calls to +match(), the RRETURN() macro jumps here. The number that is saved in +frame->Xwhere indicates which label we actually want to return to. */ + +#ifdef NO_RECURSE +#define LBL(val) case val: goto L_RM##val; +HEAP_RETURN: +switch (frame->Xwhere) + { + LBL( 1) LBL( 2) LBL( 3) LBL( 4) LBL( 5) LBL( 6) LBL( 7) LBL( 8) + LBL( 9) LBL(10) LBL(11) LBL(12) LBL(13) LBL(14) LBL(15) LBL(17) + LBL(19) LBL(24) LBL(25) LBL(26) LBL(27) LBL(29) LBL(31) LBL(33) + LBL(35) LBL(43) LBL(47) LBL(48) LBL(49) LBL(50) LBL(51) LBL(52) LBL(53) LBL(54) LBL(55) LBL(56) LBL(57) LBL(58) LBL(63) LBL(64) LBL(65) LBL(66) #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 @@ -6229,80 +6229,80 @@ switch (frame->Xwhere) #ifdef SUPPORT_UTF LBL(16) LBL(18) LBL(22) LBL(23) LBL(28) LBL(30) - LBL(32) LBL(34) LBL(42) LBL(46) -#ifdef SUPPORT_UCP - LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45) + LBL(32) LBL(34) LBL(42) LBL(46) +#ifdef SUPPORT_UCP + LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45) LBL(59) LBL(60) LBL(61) LBL(62) LBL(67) -#endif /* SUPPORT_UCP */ +#endif /* SUPPORT_UCP */ #endif /* SUPPORT_UTF */ - default: - DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere)); - return PCRE_ERROR_INTERNAL; - } -#undef LBL -#endif /* NO_RECURSE */ -} - - -/*************************************************************************** -**************************************************************************** - RECURSION IN THE match() FUNCTION - -Undefine all the macros that were defined above to handle this. */ - -#ifdef NO_RECURSE -#undef eptr -#undef ecode -#undef mstart -#undef offset_top -#undef eptrb -#undef flags - -#undef callpat -#undef charptr -#undef data -#undef next -#undef pp -#undef prev -#undef saved_eptr - -#undef new_recursive - -#undef cur_is_word -#undef condition -#undef prev_is_word - -#undef ctype -#undef length -#undef max -#undef min -#undef number -#undef offset -#undef op -#undef save_capture_last -#undef save_offset1 -#undef save_offset2 -#undef save_offset3 -#undef stacksave - -#undef newptrb - -#endif - -/* These two are defined as macros in both cases */ - -#undef fc -#undef fi - -/*************************************************************************** -***************************************************************************/ - - + default: + DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere)); + return PCRE_ERROR_INTERNAL; + } +#undef LBL +#endif /* NO_RECURSE */ +} + + +/*************************************************************************** +**************************************************************************** + RECURSION IN THE match() FUNCTION + +Undefine all the macros that were defined above to handle this. */ + +#ifdef NO_RECURSE +#undef eptr +#undef ecode +#undef mstart +#undef offset_top +#undef eptrb +#undef flags + +#undef callpat +#undef charptr +#undef data +#undef next +#undef pp +#undef prev +#undef saved_eptr + +#undef new_recursive + +#undef cur_is_word +#undef condition +#undef prev_is_word + +#undef ctype +#undef length +#undef max +#undef min +#undef number +#undef offset +#undef op +#undef save_capture_last +#undef save_offset1 +#undef save_offset2 +#undef save_offset3 +#undef stacksave + +#undef newptrb + +#endif + +/* These two are defined as macros in both cases */ + +#undef fc +#undef fi + +/*************************************************************************** +***************************************************************************/ + + #ifdef NO_RECURSE /************************************************* * Release allocated heap frames * *************************************************/ - + /* This function releases all the allocated frames. The base frame is on the machine stack, and so must not be freed. @@ -6324,35 +6324,35 @@ while (nextframe != NULL) #endif -/************************************************* -* Execute a Regular Expression * -*************************************************/ - -/* This function applies a compiled re to a subject string and picks out -portions of the string if it matches. Two elements in the vector are set for -each substring: the offsets to the start and end of the substring. - -Arguments: - argument_re points to the compiled expression - extra_data points to extra data or is NULL - subject points to the subject string - length length of subject string (may contain binary zeros) - start_offset where to start in the subject string - options option bits - offsets points to a vector of ints to be filled in with offsets - offsetcount the number of elements in the vector - -Returns: > 0 => success; value is the number of elements filled in - = 0 => success, but offsets is not big enough - -1 => failed to match - < -1 => some kind of unexpected problem -*/ - +/************************************************* +* Execute a Regular Expression * +*************************************************/ + +/* This function applies a compiled re to a subject string and picks out +portions of the string if it matches. Two elements in the vector are set for +each substring: the offsets to the start and end of the substring. + +Arguments: + argument_re points to the compiled expression + extra_data points to extra data or is NULL + subject points to the subject string + length length of subject string (may contain binary zeros) + start_offset where to start in the subject string + options option bits + offsets points to a vector of ints to be filled in with offsets + offsetcount the number of elements in the vector + +Returns: > 0 => success; value is the number of elements filled in + = 0 => success, but offsets is not big enough + -1 => failed to match + < -1 => some kind of unexpected problem +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION -pcre_exec(const pcre *argument_re, const pcre_extra *extra_data, - PCRE_SPTR subject, int length, int start_offset, int options, int *offsets, - int offsetcount) +pcre_exec(const pcre *argument_re, const pcre_extra *extra_data, + PCRE_SPTR subject, int length, int start_offset, int options, int *offsets, + int offsetcount) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre16_exec(const pcre16 *argument_re, const pcre16_extra *extra_data, @@ -6364,13 +6364,13 @@ pcre32_exec(const pcre32 *argument_re, const pcre32_extra *extra_data, PCRE_SPTR32 subject, int length, int start_offset, int options, int *offsets, int offsetcount) #endif -{ +{ int rc, ocount, arg_offset_max; -int newline; -BOOL using_temporary_offsets = FALSE; -BOOL anchored; -BOOL startline; -BOOL firstline; +int newline; +BOOL using_temporary_offsets = FALSE; +BOOL anchored; +BOOL startline; +BOOL firstline; BOOL utf; BOOL has_first_char = FALSE; BOOL has_req_char = FALSE; @@ -6378,8 +6378,8 @@ pcre_uchar first_char = 0; pcre_uchar first_char2 = 0; pcre_uchar req_char = 0; pcre_uchar req_char2 = 0; -match_data match_block; -match_data *md = &match_block; +match_data match_block; +match_data *md = &match_block; const pcre_uint8 *tables; const pcre_uint8 *start_bits = NULL; PCRE_PUCHAR start_match = (PCRE_PUCHAR)subject + start_offset; @@ -6387,17 +6387,17 @@ PCRE_PUCHAR end_subject; PCRE_PUCHAR start_partial = NULL; PCRE_PUCHAR match_partial = NULL; PCRE_PUCHAR req_char_ptr = start_match - 1; - -const pcre_study_data *study; + +const pcre_study_data *study; const REAL_PCRE *re = (const REAL_PCRE *)argument_re; - + #ifdef NO_RECURSE heapframe frame_zero; frame_zero.Xprevframe = NULL; /* Marks the top level */ frame_zero.Xnextframe = NULL; /* None are allocated yet */ md->match_frames_base = &frame_zero; #endif - + /* Check for the special magic call that measures the size of the stack used per recursive call of match(). Without the funny casting for sizeof, a Windows compiler gave this error: "unary minus operator applied to unsigned type, @@ -6411,15 +6411,15 @@ if (re == NULL && extra_data == NULL && subject == NULL && length == -999 && return match(NULL, NULL, NULL, 0, NULL, NULL, 0); #endif -/* Plausibility checks */ - -if ((options & ~PUBLIC_EXEC_OPTIONS) != 0) return PCRE_ERROR_BADOPTION; +/* Plausibility checks */ + +if ((options & ~PUBLIC_EXEC_OPTIONS) != 0) return PCRE_ERROR_BADOPTION; if (re == NULL || subject == NULL || (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL; -if (offsetcount < 0) return PCRE_ERROR_BADCOUNT; +if (offsetcount < 0) return PCRE_ERROR_BADCOUNT; if (length < 0) return PCRE_ERROR_BADLENGTH; if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET; - + /* Check that the first field in the block is the magic number. If it is not, return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which @@ -6503,34 +6503,34 @@ md->name_table = (pcre_uchar *)re + re->name_table_offset; md->name_count = re->name_count; md->name_entry_size = re->name_entry_size; -/* Fish out the optional data from the extra_data structure, first setting -the default values. */ - -study = NULL; -md->match_limit = MATCH_LIMIT; -md->match_limit_recursion = MATCH_LIMIT_RECURSION; -md->callout_data = NULL; - -/* The table pointer is always in native byte order. */ - +/* Fish out the optional data from the extra_data structure, first setting +the default values. */ + +study = NULL; +md->match_limit = MATCH_LIMIT; +md->match_limit_recursion = MATCH_LIMIT_RECURSION; +md->callout_data = NULL; + +/* The table pointer is always in native byte order. */ + tables = re->tables; - + /* The two limit values override the defaults, whatever their value. */ -if (extra_data != NULL) - { +if (extra_data != NULL) + { unsigned long int flags = extra_data->flags; - if ((flags & PCRE_EXTRA_STUDY_DATA) != 0) - study = (const pcre_study_data *)extra_data->study_data; - if ((flags & PCRE_EXTRA_MATCH_LIMIT) != 0) - md->match_limit = extra_data->match_limit; - if ((flags & PCRE_EXTRA_MATCH_LIMIT_RECURSION) != 0) - md->match_limit_recursion = extra_data->match_limit_recursion; - if ((flags & PCRE_EXTRA_CALLOUT_DATA) != 0) - md->callout_data = extra_data->callout_data; - if ((flags & PCRE_EXTRA_TABLES) != 0) tables = extra_data->tables; - } - + if ((flags & PCRE_EXTRA_STUDY_DATA) != 0) + study = (const pcre_study_data *)extra_data->study_data; + if ((flags & PCRE_EXTRA_MATCH_LIMIT) != 0) + md->match_limit = extra_data->match_limit; + if ((flags & PCRE_EXTRA_MATCH_LIMIT_RECURSION) != 0) + md->match_limit_recursion = extra_data->match_limit_recursion; + if ((flags & PCRE_EXTRA_CALLOUT_DATA) != 0) + md->callout_data = extra_data->callout_data; + if ((flags & PCRE_EXTRA_TABLES) != 0) tables = extra_data->tables; + } + /* Limits in the regex override only if they are smaller. */ if ((re->flags & PCRE_MLSET) != 0 && re->limit_match < md->match_limit) @@ -6540,171 +6540,171 @@ if ((re->flags & PCRE_RLSET) != 0 && re->limit_recursion < md->match_limit_recursion) md->match_limit_recursion = re->limit_recursion; -/* If the exec call supplied NULL for tables, use the inbuilt ones. This -is a feature that makes it possible to save compiled regex and re-use them -in other programs later. */ - +/* If the exec call supplied NULL for tables, use the inbuilt ones. This +is a feature that makes it possible to save compiled regex and re-use them +in other programs later. */ + if (tables == NULL) tables = PRIV(default_tables); - -/* Set up other data */ - -anchored = ((re->options | options) & PCRE_ANCHORED) != 0; -startline = (re->flags & PCRE_STARTLINE) != 0; -firstline = (re->options & PCRE_FIRSTLINE) != 0; - -/* The code starts after the real_pcre block and the capture name table. */ - + +/* Set up other data */ + +anchored = ((re->options | options) & PCRE_ANCHORED) != 0; +startline = (re->flags & PCRE_STARTLINE) != 0; +firstline = (re->options & PCRE_FIRSTLINE) != 0; + +/* The code starts after the real_pcre block and the capture name table. */ + md->start_code = (const pcre_uchar *)re + re->name_table_offset + - re->name_count * re->name_entry_size; - + re->name_count * re->name_entry_size; + md->start_subject = (PCRE_PUCHAR)subject; -md->start_offset = start_offset; -md->end_subject = md->start_subject + length; -end_subject = md->end_subject; - -md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0; +md->start_offset = start_offset; +md->end_subject = md->start_subject + length; +end_subject = md->end_subject; + +md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0; md->use_ucp = (re->options & PCRE_UCP) != 0; md->jscript_compat = (re->options & PCRE_JAVASCRIPT_COMPAT) != 0; md->ignore_skip_arg = 0; - + /* Some options are unpacked into BOOL variables in the hope that testing them will be faster than individual option bits. */ -md->notbol = (options & PCRE_NOTBOL) != 0; -md->noteol = (options & PCRE_NOTEOL) != 0; -md->notempty = (options & PCRE_NOTEMPTY) != 0; +md->notbol = (options & PCRE_NOTBOL) != 0; +md->noteol = (options & PCRE_NOTEOL) != 0; +md->notempty = (options & PCRE_NOTEMPTY) != 0; md->notempty_atstart = (options & PCRE_NOTEMPTY_ATSTART) != 0; -md->hitend = FALSE; +md->hitend = FALSE; md->mark = md->nomatch_mark = NULL; /* In case never set */ - -md->recursive = NULL; /* No recursion at top level */ + +md->recursive = NULL; /* No recursion at top level */ md->hasthen = (re->flags & PCRE_HASTHEN) != 0; - -md->lcc = tables + lcc_offset; + +md->lcc = tables + lcc_offset; md->fcc = tables + fcc_offset; -md->ctypes = tables + ctypes_offset; - -/* Handle different \R options. */ - -switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) - { - case 0: - if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0) - md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0; - else -#ifdef BSR_ANYCRLF - md->bsr_anycrlf = TRUE; -#else - md->bsr_anycrlf = FALSE; -#endif - break; - - case PCRE_BSR_ANYCRLF: - md->bsr_anycrlf = TRUE; - break; - - case PCRE_BSR_UNICODE: - md->bsr_anycrlf = FALSE; - break; - - default: return PCRE_ERROR_BADNEWLINE; - } - -/* Handle different types of newline. The three bits give eight cases. If -nothing is set at run time, whatever was used at compile time applies. */ - -switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : - (pcre_uint32)options) & PCRE_NEWLINE_BITS) - { - case 0: newline = NEWLINE; break; /* Compile-time default */ +md->ctypes = tables + ctypes_offset; + +/* Handle different \R options. */ + +switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) + { + case 0: + if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0) + md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0; + else +#ifdef BSR_ANYCRLF + md->bsr_anycrlf = TRUE; +#else + md->bsr_anycrlf = FALSE; +#endif + break; + + case PCRE_BSR_ANYCRLF: + md->bsr_anycrlf = TRUE; + break; + + case PCRE_BSR_UNICODE: + md->bsr_anycrlf = FALSE; + break; + + default: return PCRE_ERROR_BADNEWLINE; + } + +/* Handle different types of newline. The three bits give eight cases. If +nothing is set at run time, whatever was used at compile time applies. */ + +switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : + (pcre_uint32)options) & PCRE_NEWLINE_BITS) + { + case 0: newline = NEWLINE; break; /* Compile-time default */ case PCRE_NEWLINE_CR: newline = CHAR_CR; break; case PCRE_NEWLINE_LF: newline = CHAR_NL; break; - case PCRE_NEWLINE_CR+ + case PCRE_NEWLINE_CR+ PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break; - case PCRE_NEWLINE_ANY: newline = -1; break; - case PCRE_NEWLINE_ANYCRLF: newline = -2; break; - default: return PCRE_ERROR_BADNEWLINE; - } - -if (newline == -2) - { - md->nltype = NLTYPE_ANYCRLF; - } -else if (newline < 0) - { - md->nltype = NLTYPE_ANY; - } -else - { - md->nltype = NLTYPE_FIXED; - if (newline > 255) - { - md->nllen = 2; - md->nl[0] = (newline >> 8) & 255; - md->nl[1] = newline & 255; - } - else - { - md->nllen = 1; - md->nl[0] = newline; - } - } - + case PCRE_NEWLINE_ANY: newline = -1; break; + case PCRE_NEWLINE_ANYCRLF: newline = -2; break; + default: return PCRE_ERROR_BADNEWLINE; + } + +if (newline == -2) + { + md->nltype = NLTYPE_ANYCRLF; + } +else if (newline < 0) + { + md->nltype = NLTYPE_ANY; + } +else + { + md->nltype = NLTYPE_FIXED; + if (newline > 255) + { + md->nllen = 2; + md->nl[0] = (newline >> 8) & 255; + md->nl[1] = newline & 255; + } + else + { + md->nllen = 1; + md->nl[0] = newline; + } + } + /* Partial matching was originally supported only for a restricted set of regexes; from release 8.00 there are no restrictions, but the bits are still defined (though never set). So there's no harm in leaving this code. */ - -if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0) - return PCRE_ERROR_BADPARTIAL; - -/* If the expression has got more back references than the offsets supplied can -hold, we get a temporary chunk of working store to use during the matching. -Otherwise, we can use the vector supplied, rounding down its size to a multiple -of 3. */ - -ocount = offsetcount - (offsetcount % 3); + +if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0) + return PCRE_ERROR_BADPARTIAL; + +/* If the expression has got more back references than the offsets supplied can +hold, we get a temporary chunk of working store to use during the matching. +Otherwise, we can use the vector supplied, rounding down its size to a multiple +of 3. */ + +ocount = offsetcount - (offsetcount % 3); arg_offset_max = (2*ocount)/3; - -if (re->top_backref > 0 && re->top_backref >= ocount/3) - { - ocount = re->top_backref * 3 + 3; + +if (re->top_backref > 0 && re->top_backref >= ocount/3) + { + ocount = re->top_backref * 3 + 3; md->offset_vector = (int *)(PUBL(malloc))(ocount * sizeof(int)); - if (md->offset_vector == NULL) return PCRE_ERROR_NOMEMORY; - using_temporary_offsets = TRUE; - DPRINTF(("Got memory to hold back references\n")); - } -else md->offset_vector = offsets; -md->offset_end = ocount; -md->offset_max = (2*ocount)/3; + if (md->offset_vector == NULL) return PCRE_ERROR_NOMEMORY; + using_temporary_offsets = TRUE; + DPRINTF(("Got memory to hold back references\n")); + } +else md->offset_vector = offsets; +md->offset_end = ocount; +md->offset_max = (2*ocount)/3; md->capture_last = 0; - -/* Reset the working variable associated with each extraction. These should -never be used unless previously set, but they get saved and restored, and so we + +/* Reset the working variable associated with each extraction. These should +never be used unless previously set, but they get saved and restored, and so we initialize them to avoid reading uninitialized locations. Also, unset the offsets for the matched string. This is really just for tidiness with callouts, in case they inspect these fields. */ - -if (md->offset_vector != NULL) - { + +if (md->offset_vector != NULL) + { register int *iptr = md->offset_vector + ocount; register int *iend = iptr - re->top_bracket; if (iend < md->offset_vector + 2) iend = md->offset_vector + 2; - while (--iptr >= iend) *iptr = -1; + while (--iptr >= iend) *iptr = -1; if (offsetcount > 0) md->offset_vector[0] = -1; if (offsetcount > 1) md->offset_vector[1] = -1; - } - + } + /* Set up the first character to match, if available. The first_char value is -never set for an anchored regular expression, but the anchoring may be forced -at run time, so we have to test for anchoring. The first char may be unset for -an unanchored pattern, of course. If there's no first char and the pattern was -studied, there may be a bitmap of possible first characters. */ - -if (!anchored) - { - if ((re->flags & PCRE_FIRSTSET) != 0) - { +never set for an anchored regular expression, but the anchoring may be forced +at run time, so we have to test for anchoring. The first char may be unset for +an unanchored pattern, of course. If there's no first char and the pattern was +studied, there may be a bitmap of possible first characters. */ + +if (!anchored) + { + if ((re->flags & PCRE_FIRSTSET) != 0) + { has_first_char = TRUE; first_char = first_char2 = (pcre_uchar)(re->first_char); if ((re->flags & PCRE_FCH_CASELESS) != 0) @@ -6715,18 +6715,18 @@ if (!anchored) first_char2 = UCD_OTHERCASE(first_char); #endif } - } - else - if (!startline && study != NULL && + } + else + if (!startline && study != NULL && (study->flags & PCRE_STUDY_MAPPED) != 0) - start_bits = study->start_bits; - } - -/* For anchored or unanchored matches, there may be a "last known required -character" set. */ - -if ((re->flags & PCRE_REQCHSET) != 0) - { + start_bits = study->start_bits; + } + +/* For anchored or unanchored matches, there may be a "last known required +character" set. */ + +if ((re->flags & PCRE_REQCHSET) != 0) + { has_req_char = TRUE; req_char = req_char2 = (pcre_uchar)(re->req_char); if ((re->flags & PCRE_RCH_CASELESS) != 0) @@ -6737,27 +6737,27 @@ if ((re->flags & PCRE_REQCHSET) != 0) req_char2 = UCD_OTHERCASE(req_char); #endif } - } - - -/* ==========================================================================*/ - -/* Loop for handling unanchored repeated matching attempts; for anchored regexs -the loop runs just once. */ - -for(;;) - { + } + + +/* ==========================================================================*/ + +/* Loop for handling unanchored repeated matching attempts; for anchored regexs +the loop runs just once. */ + +for(;;) + { PCRE_PUCHAR save_end_subject = end_subject; PCRE_PUCHAR new_start_match; - + /* If firstline is TRUE, the start of the match is constrained to the first line of a multiline string. That is, the match must be before or at the first newline. Implement this by temporarily adjusting end_subject so that we stop scanning at a newline. If the match fails at the newline, later code breaks this loop. */ - - if (firstline) - { + + if (firstline) + { PCRE_PUCHAR t = start_match; #ifdef SUPPORT_UTF if (utf) @@ -6770,24 +6770,24 @@ for(;;) } else #endif - while (t < md->end_subject && !IS_NEWLINE(t)) t++; - end_subject = t; - } - + while (t < md->end_subject && !IS_NEWLINE(t)) t++; + end_subject = t; + } + /* There are some optimizations that avoid running the match if a known starting point is not found, or if a known later character is not present. However, there is an option that disables these, for testing and for ensuring that all callouts do actually occur. The option can be set in the regex by (*NO_START_OPT) or passed in match-time options. */ - + if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0) - { + { /* Advance to a unique first char if there is one. */ - + if (has_first_char) { pcre_uchar smc; - + if (first_char != first_char2) while (start_match < end_subject && (smc = UCHAR21TEST(start_match)) != first_char && smc != first_char2) @@ -6800,7 +6800,7 @@ for(;;) /* Or to just after a linebreak for a multiline match */ else if (startline) - { + { if (start_match > md->start_subject + start_offset) { #ifdef SUPPORT_UTF @@ -6817,7 +6817,7 @@ for(;;) #endif while (start_match < end_subject && !WAS_NEWLINE(start_match)) start_match++; - + /* If we have just passed a CR and the newline option is ANY or ANYCRLF, and we are now at a LF, advance the match position by one more character. */ @@ -6828,12 +6828,12 @@ for(;;) UCHAR21TEST(start_match) == CHAR_NL) start_match++; } - } - + } + /* Or to a non-unique first byte after study */ - + else if (start_bits != NULL) - { + { while (start_match < end_subject) { register pcre_uint32 c = UCHAR21TEST(start_match); @@ -6843,30 +6843,30 @@ for(;;) if ((start_bits[c/8] & (1 << (c&7))) != 0) break; start_match++; } - } + } } /* Starting optimizations */ - - /* Restore fudged end_subject */ - - end_subject = save_end_subject; - + + /* Restore fudged end_subject */ + + end_subject = save_end_subject; + /* The following two optimizations are disabled for partial matching or if disabling is explicitly requested. */ - + if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0 && !md->partial) { /* If the pattern was studied, a minimum subject length may be set. This is a lower bound; no actual string of that length may actually match the pattern. Although the value is, strictly, in characters, we treat it as bytes to avoid spending too much time in this optimization. */ - + if (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0 && (pcre_uint32)(end_subject - start_match) < study->minlength) { rc = MATCH_NOMATCH; break; } - + /* If req_char is set, we know that that character must appear in the subject for the match to succeed. If the first character is set, req_char must be later in the subject; otherwise the test starts at the match point. @@ -6874,56 +6874,56 @@ for(;;) nested unlimited repeats that aren't going to match. Writing separate code for cased/caseless versions makes it go faster, as does using an autoincrement and backing off on a match. - + HOWEVER: when the subject string is very, very long, searching to its end can take a long time, and give bad performance on quite ordinary patterns. This showed up when somebody was matching something like /^\d+C/ on a 32-megabyte string... so we don't do this when the string is sufficiently long. */ - + if (has_req_char && end_subject - start_match < REQ_BYTE_MAX) { register PCRE_PUCHAR p = start_match + (has_first_char? 1:0); - + /* We don't need to repeat the search if we haven't yet reached the place we found it at last time. */ if (p > req_char_ptr) - { + { if (req_char != req_char2) - { + { while (p < end_subject) { register pcre_uint32 pp = UCHAR21INCTEST(p); if (pp == req_char || pp == req_char2) { p--; break; } } - } + } else - { + { while (p < end_subject) { if (UCHAR21INCTEST(p) == req_char) { p--; break; } } - } - + } + /* If we can't find the required character, break the matching loop, forcing a match failure. */ - + if (p >= end_subject) { rc = MATCH_NOMATCH; break; } - + /* If we have found the required character, save the point where we found it, so that we don't search again next time round the loop if the start hasn't passed this character yet. */ - + req_char_ptr = p; } - } - } - + } + } + #ifdef PCRE_DEBUG /* Sigh. Some compilers never learn. */ printf(">>>> Match against: "); pchars(start_match, end_subject - start_match, TRUE, md); @@ -6933,9 +6933,9 @@ for(;;) /* OK, we can now run the match. If "hitend" is set afterwards, remember the first starting point for which a partial match was found. */ - md->start_match_ptr = start_match; + md->start_match_ptr = start_match; md->start_used_ptr = start_match; - md->match_call_count = 0; + md->match_call_count = 0; md->match_function_type = 0; md->end_offset_top = 0; md->skip_arg_count = 0; @@ -6945,9 +6945,9 @@ for(;;) start_partial = md->start_used_ptr; match_partial = start_match; } - - switch(rc) - { + + switch(rc) + { /* If MATCH_SKIP_ARG reaches this level it means that a MARK that matched the SKIP's arg was not found. In this circumstance, Perl ignores the SKIP entirely. The only way we can do that is to re-do the match at the same @@ -6971,117 +6971,117 @@ for(;;) } /* Fall through */ - /* NOMATCH and PRUNE advance by one character. THEN at this level acts + /* NOMATCH and PRUNE advance by one character. THEN at this level acts exactly like PRUNE. Unset ignore SKIP-with-argument. */ - - case MATCH_NOMATCH: - case MATCH_PRUNE: - case MATCH_THEN: + + case MATCH_NOMATCH: + case MATCH_PRUNE: + case MATCH_THEN: md->ignore_skip_arg = 0; - new_start_match = start_match + 1; + new_start_match = start_match + 1; #ifdef SUPPORT_UTF if (utf) ACROSSCHAR(new_start_match < end_subject, *new_start_match, new_start_match++); -#endif - break; - - /* COMMIT disables the bumpalong, but otherwise behaves as NOMATCH. */ - - case MATCH_COMMIT: - rc = MATCH_NOMATCH; - goto ENDLOOP; - +#endif + break; + + /* COMMIT disables the bumpalong, but otherwise behaves as NOMATCH. */ + + case MATCH_COMMIT: + rc = MATCH_NOMATCH; + goto ENDLOOP; + /* Any other return is either a match, or some kind of error. */ - - default: - goto ENDLOOP; - } - - /* Control reaches here for the various types of "no match at this point" - result. Reset the code to MATCH_NOMATCH for subsequent checking. */ - - rc = MATCH_NOMATCH; - - /* If PCRE_FIRSTLINE is set, the match must happen before or at the first - newline in the subject (though it may continue over the newline). Therefore, - if we have just failed to match, starting at a newline, do not continue. */ - - if (firstline && IS_NEWLINE(start_match)) break; - - /* Advance to new matching position */ - - start_match = new_start_match; - - /* Break the loop if the pattern is anchored or if we have passed the end of - the subject. */ - - if (anchored || start_match > end_subject) break; - - /* If we have just passed a CR and we are now at a LF, and the pattern does - not contain any explicit matches for \r or \n, and the newline option is CRLF + + default: + goto ENDLOOP; + } + + /* Control reaches here for the various types of "no match at this point" + result. Reset the code to MATCH_NOMATCH for subsequent checking. */ + + rc = MATCH_NOMATCH; + + /* If PCRE_FIRSTLINE is set, the match must happen before or at the first + newline in the subject (though it may continue over the newline). Therefore, + if we have just failed to match, starting at a newline, do not continue. */ + + if (firstline && IS_NEWLINE(start_match)) break; + + /* Advance to new matching position */ + + start_match = new_start_match; + + /* Break the loop if the pattern is anchored or if we have passed the end of + the subject. */ + + if (anchored || start_match > end_subject) break; + + /* If we have just passed a CR and we are now at a LF, and the pattern does + not contain any explicit matches for \r or \n, and the newline option is CRLF or ANY or ANYCRLF, advance the match position by one more character. In normal matching start_match will aways be greater than the first position at this stage, but a failed *SKIP can cause a return at the same point, which is why the first test exists. */ - + if (start_match > (PCRE_PUCHAR)subject + start_offset && start_match[-1] == CHAR_CR && - start_match < end_subject && + start_match < end_subject && *start_match == CHAR_NL && - (re->flags & PCRE_HASCRORLF) == 0 && - (md->nltype == NLTYPE_ANY || - md->nltype == NLTYPE_ANYCRLF || - md->nllen == 2)) - start_match++; - + (re->flags & PCRE_HASCRORLF) == 0 && + (md->nltype == NLTYPE_ANY || + md->nltype == NLTYPE_ANYCRLF || + md->nllen == 2)) + start_match++; + md->mark = NULL; /* Reset for start of next match attempt */ } /* End of for(;;) "bumpalong" loop */ - -/* ==========================================================================*/ - -/* We reach here when rc is not MATCH_NOMATCH, or if one of the stopping -conditions is true: - -(1) The pattern is anchored or the match was failed by (*COMMIT); - -(2) We are past the end of the subject; - -(3) PCRE_FIRSTLINE is set and we have failed to match at a newline, because - this option requests that a match occur at or before the first newline in - the subject. - -When we have a match and the offset vector is big enough to deal with any -backreferences, captured substring offsets will already be set up. In the case -where we had to get some local store to hold offsets for backreference -processing, copy those that we can. In this case there need not be overflow if -certain parts of the pattern were not used, even though there are more -capturing parentheses than vector slots. */ - -ENDLOOP: - + +/* ==========================================================================*/ + +/* We reach here when rc is not MATCH_NOMATCH, or if one of the stopping +conditions is true: + +(1) The pattern is anchored or the match was failed by (*COMMIT); + +(2) We are past the end of the subject; + +(3) PCRE_FIRSTLINE is set and we have failed to match at a newline, because + this option requests that a match occur at or before the first newline in + the subject. + +When we have a match and the offset vector is big enough to deal with any +backreferences, captured substring offsets will already be set up. In the case +where we had to get some local store to hold offsets for backreference +processing, copy those that we can. In this case there need not be overflow if +certain parts of the pattern were not used, even though there are more +capturing parentheses than vector slots. */ + +ENDLOOP: + if (rc == MATCH_MATCH || rc == MATCH_ACCEPT) - { - if (using_temporary_offsets) - { + { + if (using_temporary_offsets) + { if (arg_offset_max >= 4) - { - memcpy(offsets + 2, md->offset_vector + 2, + { + memcpy(offsets + 2, md->offset_vector + 2, (arg_offset_max - 2) * sizeof(int)); - DPRINTF(("Copied offsets from temporary memory\n")); - } + DPRINTF(("Copied offsets from temporary memory\n")); + } if (md->end_offset_top > arg_offset_max) md->capture_last |= OVFLBIT; - DPRINTF(("Freeing temporary memory\n")); + DPRINTF(("Freeing temporary memory\n")); (PUBL(free))(md->offset_vector); - } - + } + /* Set the return code to the number of captured strings, or 0 if there were - too many to fit into the vector. */ - + too many to fit into the vector. */ + rc = ((md->capture_last & OVFLBIT) != 0 && md->end_offset_top >= arg_offset_max)? 0 : md->end_offset_top/2; - + /* If there is space in the offset vector, set any unused pairs at the end of the pattern to -1 for backwards compatibility. It is documented that this happens. In earlier versions, the whole set of potential capturing offsets @@ -7101,52 +7101,52 @@ if (rc == MATCH_MATCH || rc == MATCH_ACCEPT) while (iptr < iend) *iptr++ = -1; } - /* If there is space, set up the whole thing as substring 0. The value of - md->start_match_ptr might be modified if \K was encountered on the success - matching path. */ - - if (offsetcount < 2) rc = 0; else - { + /* If there is space, set up the whole thing as substring 0. The value of + md->start_match_ptr might be modified if \K was encountered on the success + matching path. */ + + if (offsetcount < 2) rc = 0; else + { offsets[0] = (int)(md->start_match_ptr - md->start_subject); offsets[1] = (int)(md->end_match_ptr - md->start_subject); - } - + } + /* Return MARK data if requested */ if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0) *(extra_data->mark) = (pcre_uchar *)md->mark; - DPRINTF((">>>> returning %d\n", rc)); + DPRINTF((">>>> returning %d\n", rc)); #ifdef NO_RECURSE release_match_heapframes(&frame_zero); #endif - return rc; - } - -/* Control gets here if there has been an error, or if the overall match -attempt has failed at all permitted starting positions. */ - -if (using_temporary_offsets) - { - DPRINTF(("Freeing temporary memory\n")); + return rc; + } + +/* Control gets here if there has been an error, or if the overall match +attempt has failed at all permitted starting positions. */ + +if (using_temporary_offsets) + { + DPRINTF(("Freeing temporary memory\n")); (PUBL(free))(md->offset_vector); - } - + } + /* For anything other than nomatch or partial match, just return the code. */ if (rc != MATCH_NOMATCH && rc != PCRE_ERROR_PARTIAL) - { - DPRINTF((">>>> error: returning %d\n", rc)); + { + DPRINTF((">>>> error: returning %d\n", rc)); #ifdef NO_RECURSE release_match_heapframes(&frame_zero); #endif - return rc; - } + return rc; + } /* Handle partial matches - disable any mark data */ if (match_partial != NULL) - { - DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n")); + { + DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n")); md->mark = NULL; if (offsetcount > 1) { @@ -7156,15 +7156,15 @@ if (match_partial != NULL) offsets[2] = (int)(match_partial - (PCRE_PUCHAR)subject); } rc = PCRE_ERROR_PARTIAL; - } + } /* This is the classic nomatch case */ -else - { - DPRINTF((">>>> returning PCRE_ERROR_NOMATCH\n")); +else + { + DPRINTF((">>>> returning PCRE_ERROR_NOMATCH\n")); rc = PCRE_ERROR_NOMATCH; - } + } /* Return the MARK data if it has been requested. */ @@ -7174,6 +7174,6 @@ if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0) release_match_heapframes(&frame_zero); #endif return rc; -} - -/* End of pcre_exec.c */ +} + +/* End of pcre_exec.c */ diff --git a/contrib/libs/pcre/pcre_fullinfo.c b/contrib/libs/pcre/pcre_fullinfo.c index bfccc02598..ac066ecf20 100644 --- a/contrib/libs/pcre/pcre_fullinfo.c +++ b/contrib/libs/pcre/pcre_fullinfo.c @@ -1,70 +1,70 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + /* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2013 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains the external function pcre_fullinfo(), which returns -information about a compiled pattern. */ - - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains the external function pcre_fullinfo(), which returns +information about a compiled pattern. */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - - -/************************************************* -* Return info about compiled pattern * -*************************************************/ - -/* This is a newer "info" function which has an extensible interface so -that additional items can be added compatibly. - -Arguments: - argument_re points to compiled code - extra_data points extra data, or NULL - what what information is required - where where to put the information - -Returns: 0 if data returned, negative on error -*/ - +#endif + +#include "pcre_internal.h" + + +/************************************************* +* Return info about compiled pattern * +*************************************************/ + +/* This is a newer "info" function which has an extensible interface so +that additional items can be added compatibly. + +Arguments: + argument_re points to compiled code + extra_data points extra data, or NULL + what what information is required + where where to put the information + +Returns: 0 if data returned, negative on error +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre_fullinfo(const pcre *argument_re, const pcre_extra *extra_data, @@ -78,42 +78,42 @@ PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre32_fullinfo(const pcre32 *argument_re, const pcre32_extra *extra_data, int what, void *where) #endif -{ +{ const REAL_PCRE *re = (const REAL_PCRE *)argument_re; -const pcre_study_data *study = NULL; - -if (re == NULL || where == NULL) return PCRE_ERROR_NULL; - -if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_STUDY_DATA) != 0) - study = (const pcre_study_data *)extra_data->study_data; - +const pcre_study_data *study = NULL; + +if (re == NULL || where == NULL) return PCRE_ERROR_NULL; + +if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_STUDY_DATA) != 0) + study = (const pcre_study_data *)extra_data->study_data; + /* Check that the first field in the block is the magic number. If it is not, return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which means that the pattern is likely compiled with different endianness. */ -if (re->magic_number != MAGIC_NUMBER) +if (re->magic_number != MAGIC_NUMBER) return re->magic_number == REVERSED_MAGIC_NUMBER? PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC; - + /* Check that this pattern was compiled in the correct bit mode */ if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE; -switch (what) - { - case PCRE_INFO_OPTIONS: +switch (what) + { + case PCRE_INFO_OPTIONS: *((unsigned long int *)where) = re->options & PUBLIC_COMPILE_OPTIONS; - break; - - case PCRE_INFO_SIZE: - *((size_t *)where) = re->size; - break; - - case PCRE_INFO_STUDYSIZE: - *((size_t *)where) = (study == NULL)? 0 : study->size; - break; - + break; + + case PCRE_INFO_SIZE: + *((size_t *)where) = re->size; + break; + + case PCRE_INFO_STUDYSIZE: + *((size_t *)where) = (study == NULL)? 0 : study->size; + break; + case PCRE_INFO_JITSIZE: #ifdef SUPPORT_JIT *((size_t *)where) = @@ -126,20 +126,20 @@ switch (what) #endif break; - case PCRE_INFO_CAPTURECOUNT: - *((int *)where) = re->top_bracket; - break; - - case PCRE_INFO_BACKREFMAX: - *((int *)where) = re->top_backref; - break; - - case PCRE_INFO_FIRSTBYTE: - *((int *)where) = + case PCRE_INFO_CAPTURECOUNT: + *((int *)where) = re->top_bracket; + break; + + case PCRE_INFO_BACKREFMAX: + *((int *)where) = re->top_backref; + break; + + case PCRE_INFO_FIRSTBYTE: + *((int *)where) = ((re->flags & PCRE_FIRSTSET) != 0)? (int)re->first_char : - ((re->flags & PCRE_STARTLINE) != 0)? -1 : -2; - break; - + ((re->flags & PCRE_STARTLINE) != 0)? -1 : -2; + break; + case PCRE_INFO_FIRSTCHARACTER: *((pcre_uint32 *)where) = (re->flags & PCRE_FIRSTSET) != 0 ? re->first_char : 0; @@ -151,15 +151,15 @@ switch (what) ((re->flags & PCRE_STARTLINE) != 0) ? 2 : 0; break; - /* Make sure we pass back the pointer to the bit vector in the external - block, not the internal copy (with flipped integer fields). */ - - case PCRE_INFO_FIRSTTABLE: + /* Make sure we pass back the pointer to the bit vector in the external + block, not the internal copy (with flipped integer fields). */ + + case PCRE_INFO_FIRSTTABLE: *((const pcre_uint8 **)where) = (study != NULL && (study->flags & PCRE_STUDY_MAPPED) != 0)? - ((const pcre_study_data *)extra_data->study_data)->start_bits : NULL; - break; - + ((const pcre_study_data *)extra_data->study_data)->start_bits : NULL; + break; + case PCRE_INFO_MINLENGTH: *((int *)where) = (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0)? @@ -172,11 +172,11 @@ switch (what) extra_data->executable_jit != NULL; break; - case PCRE_INFO_LASTLITERAL: - *((int *)where) = + case PCRE_INFO_LASTLITERAL: + *((int *)where) = ((re->flags & PCRE_REQCHSET) != 0)? (int)re->req_char : -1; - break; - + break; + case PCRE_INFO_REQUIREDCHAR: *((pcre_uint32 *)where) = ((re->flags & PCRE_REQCHSET) != 0) ? re->req_char : 0; @@ -187,37 +187,37 @@ switch (what) ((re->flags & PCRE_REQCHSET) != 0); break; - case PCRE_INFO_NAMEENTRYSIZE: - *((int *)where) = re->name_entry_size; - break; - - case PCRE_INFO_NAMECOUNT: - *((int *)where) = re->name_count; - break; - - case PCRE_INFO_NAMETABLE: + case PCRE_INFO_NAMEENTRYSIZE: + *((int *)where) = re->name_entry_size; + break; + + case PCRE_INFO_NAMECOUNT: + *((int *)where) = re->name_count; + break; + + case PCRE_INFO_NAMETABLE: *((const pcre_uchar **)where) = (const pcre_uchar *)re + re->name_table_offset; - break; - - case PCRE_INFO_DEFAULT_TABLES: + break; + + case PCRE_INFO_DEFAULT_TABLES: *((const pcre_uint8 **)where) = (const pcre_uint8 *)(PRIV(default_tables)); - break; - + break; + /* From release 8.00 this will always return TRUE because NOPARTIAL is no longer ever set (the restrictions have been removed). */ - case PCRE_INFO_OKPARTIAL: - *((int *)where) = (re->flags & PCRE_NOPARTIAL) == 0; - break; - - case PCRE_INFO_JCHANGED: - *((int *)where) = (re->flags & PCRE_JCHANGED) != 0; - break; - - case PCRE_INFO_HASCRORLF: - *((int *)where) = (re->flags & PCRE_HASCRORLF) != 0; - break; - + case PCRE_INFO_OKPARTIAL: + *((int *)where) = (re->flags & PCRE_NOPARTIAL) == 0; + break; + + case PCRE_INFO_JCHANGED: + *((int *)where) = (re->flags & PCRE_JCHANGED) != 0; + break; + + case PCRE_INFO_HASCRORLF: + *((int *)where) = (re->flags & PCRE_HASCRORLF) != 0; + break; + case PCRE_INFO_MAXLOOKBEHIND: *((int *)where) = re->max_lookbehind; break; @@ -236,10 +236,10 @@ switch (what) *((int *)where) = (re->flags & PCRE_MATCH_EMPTY) != 0; break; - default: return PCRE_ERROR_BADOPTION; - } - -return 0; -} - -/* End of pcre_fullinfo.c */ + default: return PCRE_ERROR_BADOPTION; + } + +return 0; +} + +/* End of pcre_fullinfo.c */ diff --git a/contrib/libs/pcre/pcre_get.c b/contrib/libs/pcre/pcre_get.c index 11392db08e..c6b1e97536 100644 --- a/contrib/libs/pcre/pcre_get.c +++ b/contrib/libs/pcre/pcre_get.c @@ -1,73 +1,73 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2012 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains some convenience functions for extracting substrings -from the subject string after a regex match has succeeded. The original idea -for these functions came from Scott Wimer. */ - - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains some convenience functions for extracting substrings +from the subject string after a regex match has succeeded. The original idea +for these functions came from Scott Wimer. */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - - -/************************************************* -* Find number for named string * -*************************************************/ - -/* This function is used by the get_first_set() function below, as well -as being generally available. It assumes that names are unique. - -Arguments: - code the compiled regex - stringname the name whose number is required - -Returns: the number of the named parentheses, or a negative number - (PCRE_ERROR_NOSUBSTRING) if not found -*/ - +#endif + +#include "pcre_internal.h" + + +/************************************************* +* Find number for named string * +*************************************************/ + +/* This function is used by the get_first_set() function below, as well +as being generally available. It assumes that names are unique. + +Arguments: + code the compiled regex + stringname the name whose number is required + +Returns: the number of the named parentheses, or a negative number + (PCRE_ERROR_NOSUBSTRING) if not found +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION -pcre_get_stringnumber(const pcre *code, const char *stringname) +pcre_get_stringnumber(const pcre *code, const char *stringname) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre16_get_stringnumber(const pcre16 *code, PCRE_SPTR16 stringname) @@ -75,27 +75,27 @@ pcre16_get_stringnumber(const pcre16 *code, PCRE_SPTR16 stringname) PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre32_get_stringnumber(const pcre32 *code, PCRE_SPTR32 stringname) #endif -{ -int rc; -int entrysize; -int top, bot; +{ +int rc; +int entrysize; +int top, bot; pcre_uchar *nametable; - + #ifdef COMPILE_PCRE8 -if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0) - return rc; -if (top <= 0) return PCRE_ERROR_NOSUBSTRING; - -if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0) - return rc; -if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0) - return rc; +if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0) + return rc; +if (top <= 0) return PCRE_ERROR_NOSUBSTRING; + +if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0) + return rc; +if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0) + return rc; #endif #ifdef COMPILE_PCRE16 if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0) return rc; if (top <= 0) return PCRE_ERROR_NOSUBSTRING; - + if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0) return rc; if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0) @@ -112,43 +112,43 @@ if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0) return rc; #endif -bot = 0; -while (top > bot) - { - int mid = (top + bot) / 2; +bot = 0; +while (top > bot) + { + int mid = (top + bot) / 2; pcre_uchar *entry = nametable + entrysize*mid; int c = STRCMP_UC_UC((pcre_uchar *)stringname, (pcre_uchar *)(entry + IMM2_SIZE)); if (c == 0) return GET2(entry, 0); - if (c > 0) bot = mid + 1; else top = mid; - } - -return PCRE_ERROR_NOSUBSTRING; -} - - - -/************************************************* -* Find (multiple) entries for named string * -*************************************************/ - -/* This is used by the get_first_set() function below, as well as being -generally available. It is used when duplicated names are permitted. - -Arguments: - code the compiled regex - stringname the name whose entries required - firstptr where to put the pointer to the first entry - lastptr where to put the pointer to the last entry - -Returns: the length of each entry, or a negative number - (PCRE_ERROR_NOSUBSTRING) if not found -*/ - + if (c > 0) bot = mid + 1; else top = mid; + } + +return PCRE_ERROR_NOSUBSTRING; +} + + + +/************************************************* +* Find (multiple) entries for named string * +*************************************************/ + +/* This is used by the get_first_set() function below, as well as being +generally available. It is used when duplicated names are permitted. + +Arguments: + code the compiled regex + stringname the name whose entries required + firstptr where to put the pointer to the first entry + lastptr where to put the pointer to the last entry + +Returns: the length of each entry, or a negative number + (PCRE_ERROR_NOSUBSTRING) if not found +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION -pcre_get_stringtable_entries(const pcre *code, const char *stringname, - char **firstptr, char **lastptr) +pcre_get_stringtable_entries(const pcre *code, const char *stringname, + char **firstptr, char **lastptr) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre16_get_stringtable_entries(const pcre16 *code, PCRE_SPTR16 stringname, @@ -158,27 +158,27 @@ PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre32_get_stringtable_entries(const pcre32 *code, PCRE_SPTR32 stringname, PCRE_UCHAR32 **firstptr, PCRE_UCHAR32 **lastptr) #endif -{ -int rc; -int entrysize; -int top, bot; +{ +int rc; +int entrysize; +int top, bot; pcre_uchar *nametable, *lastentry; - + #ifdef COMPILE_PCRE8 -if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0) - return rc; -if (top <= 0) return PCRE_ERROR_NOSUBSTRING; - -if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0) - return rc; -if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0) - return rc; +if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0) + return rc; +if (top <= 0) return PCRE_ERROR_NOSUBSTRING; + +if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0) + return rc; +if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0) + return rc; #endif #ifdef COMPILE_PCRE16 if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0) return rc; if (top <= 0) return PCRE_ERROR_NOSUBSTRING; - + if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0) return rc; if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0) @@ -195,33 +195,33 @@ if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0) return rc; #endif -lastentry = nametable + entrysize * (top - 1); -bot = 0; -while (top > bot) - { - int mid = (top + bot) / 2; +lastentry = nametable + entrysize * (top - 1); +bot = 0; +while (top > bot) + { + int mid = (top + bot) / 2; pcre_uchar *entry = nametable + entrysize*mid; int c = STRCMP_UC_UC((pcre_uchar *)stringname, (pcre_uchar *)(entry + IMM2_SIZE)); - if (c == 0) - { + if (c == 0) + { pcre_uchar *first = entry; pcre_uchar *last = entry; - while (first > nametable) - { + while (first > nametable) + { if (STRCMP_UC_UC((pcre_uchar *)stringname, (pcre_uchar *)(first - entrysize + IMM2_SIZE)) != 0) break; - first -= entrysize; - } - while (last < lastentry) - { + first -= entrysize; + } + while (last < lastentry) + { if (STRCMP_UC_UC((pcre_uchar *)stringname, (pcre_uchar *)(last + entrysize + IMM2_SIZE)) != 0) break; - last += entrysize; - } + last += entrysize; + } #if defined COMPILE_PCRE8 - *firstptr = (char *)first; - *lastptr = (char *)last; + *firstptr = (char *)first; + *lastptr = (char *)last; #elif defined COMPILE_PCRE16 *firstptr = (PCRE_UCHAR16 *)first; *lastptr = (PCRE_UCHAR16 *)last; @@ -229,36 +229,36 @@ while (top > bot) *firstptr = (PCRE_UCHAR32 *)first; *lastptr = (PCRE_UCHAR32 *)last; #endif - return entrysize; - } - if (c > 0) bot = mid + 1; else top = mid; - } - -return PCRE_ERROR_NOSUBSTRING; -} - - - -/************************************************* -* Find first set of multiple named strings * -*************************************************/ - -/* This function allows for duplicate names in the table of named substrings. -It returns the number of the first one that was set in a pattern match. - -Arguments: - code the compiled regex - stringname the name of the capturing substring - ovector the vector of matched substrings + return entrysize; + } + if (c > 0) bot = mid + 1; else top = mid; + } + +return PCRE_ERROR_NOSUBSTRING; +} + + + +/************************************************* +* Find first set of multiple named strings * +*************************************************/ + +/* This function allows for duplicate names in the table of named substrings. +It returns the number of the first one that was set in a pattern match. + +Arguments: + code the compiled regex + stringname the name of the capturing substring + ovector the vector of matched substrings stringcount number of captured substrings - -Returns: the number of the first that is set, - or the number of the last one if none are set, - or a negative number on error -*/ - + +Returns: the number of the first that is set, + or the number of the last one if none are set, + or a negative number on error +*/ + #if defined COMPILE_PCRE8 -static int +static int get_first_set(const pcre *code, const char *stringname, int *ovector, int stringcount) #elif defined COMPILE_PCRE16 @@ -270,12 +270,12 @@ static int get_first_set(const pcre32 *code, PCRE_SPTR32 stringname, int *ovector, int stringcount) #endif -{ +{ const REAL_PCRE *re = (const REAL_PCRE *)code; -int entrysize; +int entrysize; pcre_uchar *entry; #if defined COMPILE_PCRE8 -char *first, *last; +char *first, *last; #elif defined COMPILE_PCRE16 PCRE_UCHAR16 *first, *last; #elif defined COMPILE_PCRE32 @@ -283,9 +283,9 @@ PCRE_UCHAR32 *first, *last; #endif #if defined COMPILE_PCRE8 -if ((re->options & PCRE_DUPNAMES) == 0 && (re->flags & PCRE_JCHANGED) == 0) - return pcre_get_stringnumber(code, stringname); -entrysize = pcre_get_stringtable_entries(code, stringname, &first, &last); +if ((re->options & PCRE_DUPNAMES) == 0 && (re->flags & PCRE_JCHANGED) == 0) + return pcre_get_stringnumber(code, stringname); +entrysize = pcre_get_stringtable_entries(code, stringname, &first, &last); #elif defined COMPILE_PCRE16 if ((re->options & PCRE_DUPNAMES) == 0 && (re->flags & PCRE_JCHANGED) == 0) return pcre16_get_stringnumber(code, stringname); @@ -295,49 +295,49 @@ if ((re->options & PCRE_DUPNAMES) == 0 && (re->flags & PCRE_JCHANGED) == 0) return pcre32_get_stringnumber(code, stringname); entrysize = pcre32_get_stringtable_entries(code, stringname, &first, &last); #endif -if (entrysize <= 0) return entrysize; +if (entrysize <= 0) return entrysize; for (entry = (pcre_uchar *)first; entry <= (pcre_uchar *)last; entry += entrysize) - { + { int n = GET2(entry, 0); if (n < stringcount && ovector[n*2] >= 0) return n; - } + } return GET2(entry, 0); -} - - - - -/************************************************* -* Copy captured string to given buffer * -*************************************************/ - -/* This function copies a single captured substring into a given buffer. -Note that we use memcpy() rather than strncpy() in case there are binary zeros -in the string. - -Arguments: - subject the subject string that was matched - ovector pointer to the offsets table - stringcount the number of substrings that were captured - (i.e. the yield of the pcre_exec call, unless - that was zero, in which case it should be 1/3 - of the offset table size) - stringnumber the number of the required substring - buffer where to put the substring - size the size of the buffer - -Returns: if successful: - the length of the copied string, not including the zero - that is put on the end; can be zero - if not successful: - PCRE_ERROR_NOMEMORY (-6) buffer too small - PCRE_ERROR_NOSUBSTRING (-7) no such captured substring -*/ - +} + + + + +/************************************************* +* Copy captured string to given buffer * +*************************************************/ + +/* This function copies a single captured substring into a given buffer. +Note that we use memcpy() rather than strncpy() in case there are binary zeros +in the string. + +Arguments: + subject the subject string that was matched + ovector pointer to the offsets table + stringcount the number of substrings that were captured + (i.e. the yield of the pcre_exec call, unless + that was zero, in which case it should be 1/3 + of the offset table size) + stringnumber the number of the required substring + buffer where to put the substring + size the size of the buffer + +Returns: if successful: + the length of the copied string, not including the zero + that is put on the end; can be zero + if not successful: + PCRE_ERROR_NOMEMORY (-6) buffer too small + PCRE_ERROR_NOSUBSTRING (-7) no such captured substring +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION -pcre_copy_substring(const char *subject, int *ovector, int stringcount, - int stringnumber, char *buffer, int size) +pcre_copy_substring(const char *subject, int *ovector, int stringcount, + int stringnumber, char *buffer, int size) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre16_copy_substring(PCRE_SPTR16 subject, int *ovector, int stringcount, @@ -347,48 +347,48 @@ PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre32_copy_substring(PCRE_SPTR32 subject, int *ovector, int stringcount, int stringnumber, PCRE_UCHAR32 *buffer, int size) #endif -{ -int yield; -if (stringnumber < 0 || stringnumber >= stringcount) - return PCRE_ERROR_NOSUBSTRING; -stringnumber *= 2; -yield = ovector[stringnumber+1] - ovector[stringnumber]; -if (size < yield + 1) return PCRE_ERROR_NOMEMORY; +{ +int yield; +if (stringnumber < 0 || stringnumber >= stringcount) + return PCRE_ERROR_NOSUBSTRING; +stringnumber *= 2; +yield = ovector[stringnumber+1] - ovector[stringnumber]; +if (size < yield + 1) return PCRE_ERROR_NOMEMORY; memcpy(buffer, subject + ovector[stringnumber], IN_UCHARS(yield)); -buffer[yield] = 0; -return yield; -} - - - -/************************************************* -* Copy named captured string to given buffer * -*************************************************/ - -/* This function copies a single captured substring into a given buffer, -identifying it by name. If the regex permits duplicate names, the first -substring that is set is chosen. - -Arguments: - code the compiled regex - subject the subject string that was matched - ovector pointer to the offsets table - stringcount the number of substrings that were captured - (i.e. the yield of the pcre_exec call, unless - that was zero, in which case it should be 1/3 - of the offset table size) - stringname the name of the required substring - buffer where to put the substring - size the size of the buffer - -Returns: if successful: - the length of the copied string, not including the zero - that is put on the end; can be zero - if not successful: - PCRE_ERROR_NOMEMORY (-6) buffer too small - PCRE_ERROR_NOSUBSTRING (-7) no such captured substring -*/ - +buffer[yield] = 0; +return yield; +} + + + +/************************************************* +* Copy named captured string to given buffer * +*************************************************/ + +/* This function copies a single captured substring into a given buffer, +identifying it by name. If the regex permits duplicate names, the first +substring that is set is chosen. + +Arguments: + code the compiled regex + subject the subject string that was matched + ovector pointer to the offsets table + stringcount the number of substrings that were captured + (i.e. the yield of the pcre_exec call, unless + that was zero, in which case it should be 1/3 + of the offset table size) + stringname the name of the required substring + buffer where to put the substring + size the size of the buffer + +Returns: if successful: + the length of the copied string, not including the zero + that is put on the end; can be zero + if not successful: + PCRE_ERROR_NOMEMORY (-6) buffer too small + PCRE_ERROR_NOSUBSTRING (-7) no such captured substring +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre_copy_named_substring(const pcre *code, const char *subject, @@ -405,45 +405,45 @@ pcre32_copy_named_substring(const pcre32 *code, PCRE_SPTR32 subject, int *ovector, int stringcount, PCRE_SPTR32 stringname, PCRE_UCHAR32 *buffer, int size) #endif -{ +{ int n = get_first_set(code, stringname, ovector, stringcount); -if (n <= 0) return n; +if (n <= 0) return n; #if defined COMPILE_PCRE8 -return pcre_copy_substring(subject, ovector, stringcount, n, buffer, size); +return pcre_copy_substring(subject, ovector, stringcount, n, buffer, size); #elif defined COMPILE_PCRE16 return pcre16_copy_substring(subject, ovector, stringcount, n, buffer, size); #elif defined COMPILE_PCRE32 return pcre32_copy_substring(subject, ovector, stringcount, n, buffer, size); #endif -} - - - -/************************************************* -* Copy all captured strings to new store * -*************************************************/ - -/* This function gets one chunk of store and builds a list of pointers and all -of the captured substrings in it. A NULL pointer is put on the end of the list. - -Arguments: - subject the subject string that was matched - ovector pointer to the offsets table - stringcount the number of substrings that were captured - (i.e. the yield of the pcre_exec call, unless - that was zero, in which case it should be 1/3 - of the offset table size) - listptr set to point to the list of pointers - -Returns: if successful: 0 - if not successful: - PCRE_ERROR_NOMEMORY (-6) failed to get store -*/ - +} + + + +/************************************************* +* Copy all captured strings to new store * +*************************************************/ + +/* This function gets one chunk of store and builds a list of pointers and all +of the captured substrings in it. A NULL pointer is put on the end of the list. + +Arguments: + subject the subject string that was matched + ovector pointer to the offsets table + stringcount the number of substrings that were captured + (i.e. the yield of the pcre_exec call, unless + that was zero, in which case it should be 1/3 + of the offset table size) + listptr set to point to the list of pointers + +Returns: if successful: 0 + if not successful: + PCRE_ERROR_NOMEMORY (-6) failed to get store +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION -pcre_get_substring_list(const char *subject, int *ovector, int stringcount, - const char ***listptr) +pcre_get_substring_list(const char *subject, int *ovector, int stringcount, + const char ***listptr) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre16_get_substring_list(PCRE_SPTR16 subject, int *ovector, int stringcount, @@ -453,61 +453,61 @@ PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre32_get_substring_list(PCRE_SPTR32 subject, int *ovector, int stringcount, PCRE_SPTR32 **listptr) #endif -{ -int i; +{ +int i; int size = sizeof(pcre_uchar *); -int double_count = stringcount * 2; +int double_count = stringcount * 2; pcre_uchar **stringlist; pcre_uchar *p; - -for (i = 0; i < double_count; i += 2) + +for (i = 0; i < double_count; i += 2) { size += sizeof(pcre_uchar *) + IN_UCHARS(1); if (ovector[i+1] > ovector[i]) size += IN_UCHARS(ovector[i+1] - ovector[i]); } - + stringlist = (pcre_uchar **)(PUBL(malloc))(size); -if (stringlist == NULL) return PCRE_ERROR_NOMEMORY; - +if (stringlist == NULL) return PCRE_ERROR_NOMEMORY; + #if defined COMPILE_PCRE8 -*listptr = (const char **)stringlist; +*listptr = (const char **)stringlist; #elif defined COMPILE_PCRE16 *listptr = (PCRE_SPTR16 *)stringlist; #elif defined COMPILE_PCRE32 *listptr = (PCRE_SPTR32 *)stringlist; #endif p = (pcre_uchar *)(stringlist + stringcount + 1); - -for (i = 0; i < double_count; i += 2) - { + +for (i = 0; i < double_count; i += 2) + { int len = (ovector[i+1] > ovector[i])? (ovector[i+1] - ovector[i]) : 0; memcpy(p, subject + ovector[i], IN_UCHARS(len)); - *stringlist++ = p; - p += len; - *p++ = 0; - } - -*stringlist = NULL; -return 0; -} - - - -/************************************************* -* Free store obtained by get_substring_list * -*************************************************/ - -/* This function exists for the benefit of people calling PCRE from non-C + *stringlist++ = p; + p += len; + *p++ = 0; + } + +*stringlist = NULL; +return 0; +} + + + +/************************************************* +* Free store obtained by get_substring_list * +*************************************************/ + +/* This function exists for the benefit of people calling PCRE from non-C programs that can call its functions, but not free() or (PUBL(free))() directly. - -Argument: the result of a previous pcre_get_substring_list() -Returns: nothing -*/ - + +Argument: the result of a previous pcre_get_substring_list() +Returns: nothing +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN void PCRE_CALL_CONVENTION -pcre_free_substring_list(const char **pointer) +pcre_free_substring_list(const char **pointer) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN void PCRE_CALL_CONVENTION pcre16_free_substring_list(PCRE_SPTR16 *pointer) @@ -515,41 +515,41 @@ pcre16_free_substring_list(PCRE_SPTR16 *pointer) PCRE_EXP_DEFN void PCRE_CALL_CONVENTION pcre32_free_substring_list(PCRE_SPTR32 *pointer) #endif -{ +{ (PUBL(free))((void *)pointer); -} - - - -/************************************************* -* Copy captured string to new store * -*************************************************/ - -/* This function copies a single captured substring into a piece of new -store - -Arguments: - subject the subject string that was matched - ovector pointer to the offsets table - stringcount the number of substrings that were captured - (i.e. the yield of the pcre_exec call, unless - that was zero, in which case it should be 1/3 - of the offset table size) - stringnumber the number of the required substring - stringptr where to put a pointer to the substring - -Returns: if successful: - the length of the string, not including the zero that - is put on the end; can be zero - if not successful: - PCRE_ERROR_NOMEMORY (-6) failed to get store - PCRE_ERROR_NOSUBSTRING (-7) substring not present -*/ - +} + + + +/************************************************* +* Copy captured string to new store * +*************************************************/ + +/* This function copies a single captured substring into a piece of new +store + +Arguments: + subject the subject string that was matched + ovector pointer to the offsets table + stringcount the number of substrings that were captured + (i.e. the yield of the pcre_exec call, unless + that was zero, in which case it should be 1/3 + of the offset table size) + stringnumber the number of the required substring + stringptr where to put a pointer to the substring + +Returns: if successful: + the length of the string, not including the zero that + is put on the end; can be zero + if not successful: + PCRE_ERROR_NOMEMORY (-6) failed to get store + PCRE_ERROR_NOSUBSTRING (-7) substring not present +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION -pcre_get_substring(const char *subject, int *ovector, int stringcount, - int stringnumber, const char **stringptr) +pcre_get_substring(const char *subject, int *ovector, int stringcount, + int stringnumber, const char **stringptr) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre16_get_substring(PCRE_SPTR16 subject, int *ovector, int stringcount, @@ -559,17 +559,17 @@ PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre32_get_substring(PCRE_SPTR32 subject, int *ovector, int stringcount, int stringnumber, PCRE_SPTR32 *stringptr) #endif -{ -int yield; +{ +int yield; pcre_uchar *substring; -if (stringnumber < 0 || stringnumber >= stringcount) - return PCRE_ERROR_NOSUBSTRING; -stringnumber *= 2; -yield = ovector[stringnumber+1] - ovector[stringnumber]; +if (stringnumber < 0 || stringnumber >= stringcount) + return PCRE_ERROR_NOSUBSTRING; +stringnumber *= 2; +yield = ovector[stringnumber+1] - ovector[stringnumber]; substring = (pcre_uchar *)(PUBL(malloc))(IN_UCHARS(yield + 1)); -if (substring == NULL) return PCRE_ERROR_NOMEMORY; +if (substring == NULL) return PCRE_ERROR_NOMEMORY; memcpy(substring, subject + ovector[stringnumber], IN_UCHARS(yield)); -substring[yield] = 0; +substring[yield] = 0; #if defined COMPILE_PCRE8 *stringptr = (const char *)substring; #elif defined COMPILE_PCRE16 @@ -577,38 +577,38 @@ substring[yield] = 0; #elif defined COMPILE_PCRE32 *stringptr = (PCRE_SPTR32)substring; #endif -return yield; -} - - - -/************************************************* -* Copy named captured string to new store * -*************************************************/ - -/* This function copies a single captured substring, identified by name, into -new store. If the regex permits duplicate names, the first substring that is -set is chosen. - -Arguments: - code the compiled regex - subject the subject string that was matched - ovector pointer to the offsets table - stringcount the number of substrings that were captured - (i.e. the yield of the pcre_exec call, unless - that was zero, in which case it should be 1/3 - of the offset table size) - stringname the name of the required substring - stringptr where to put the pointer - -Returns: if successful: - the length of the copied string, not including the zero - that is put on the end; can be zero - if not successful: - PCRE_ERROR_NOMEMORY (-6) couldn't get memory - PCRE_ERROR_NOSUBSTRING (-7) no such captured substring -*/ - +return yield; +} + + + +/************************************************* +* Copy named captured string to new store * +*************************************************/ + +/* This function copies a single captured substring, identified by name, into +new store. If the regex permits duplicate names, the first substring that is +set is chosen. + +Arguments: + code the compiled regex + subject the subject string that was matched + ovector pointer to the offsets table + stringcount the number of substrings that were captured + (i.e. the yield of the pcre_exec call, unless + that was zero, in which case it should be 1/3 + of the offset table size) + stringname the name of the required substring + stringptr where to put the pointer + +Returns: if successful: + the length of the copied string, not including the zero + that is put on the end; can be zero + if not successful: + PCRE_ERROR_NOMEMORY (-6) couldn't get memory + PCRE_ERROR_NOSUBSTRING (-7) no such captured substring +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre_get_named_substring(const pcre *code, const char *subject, @@ -625,36 +625,36 @@ pcre32_get_named_substring(const pcre32 *code, PCRE_SPTR32 subject, int *ovector, int stringcount, PCRE_SPTR32 stringname, PCRE_SPTR32 *stringptr) #endif -{ +{ int n = get_first_set(code, stringname, ovector, stringcount); -if (n <= 0) return n; +if (n <= 0) return n; #if defined COMPILE_PCRE8 -return pcre_get_substring(subject, ovector, stringcount, n, stringptr); +return pcre_get_substring(subject, ovector, stringcount, n, stringptr); #elif defined COMPILE_PCRE16 return pcre16_get_substring(subject, ovector, stringcount, n, stringptr); #elif defined COMPILE_PCRE32 return pcre32_get_substring(subject, ovector, stringcount, n, stringptr); #endif -} - - - - -/************************************************* -* Free store obtained by get_substring * -*************************************************/ - -/* This function exists for the benefit of people calling PCRE from non-C +} + + + + +/************************************************* +* Free store obtained by get_substring * +*************************************************/ + +/* This function exists for the benefit of people calling PCRE from non-C programs that can call its functions, but not free() or (PUBL(free))() directly. - -Argument: the result of a previous pcre_get_substring() -Returns: nothing -*/ - + +Argument: the result of a previous pcre_get_substring() +Returns: nothing +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN void PCRE_CALL_CONVENTION -pcre_free_substring(const char *pointer) +pcre_free_substring(const char *pointer) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN void PCRE_CALL_CONVENTION pcre16_free_substring(PCRE_SPTR16 pointer) @@ -662,8 +662,8 @@ pcre16_free_substring(PCRE_SPTR16 pointer) PCRE_EXP_DEFN void PCRE_CALL_CONVENTION pcre32_free_substring(PCRE_SPTR32 pointer) #endif -{ +{ (PUBL(free))((void *)pointer); -} - -/* End of pcre_get.c */ +} + +/* End of pcre_get.c */ diff --git a/contrib/libs/pcre/pcre_globals.c b/contrib/libs/pcre/pcre_globals.c index b0418be597..21a2fe9de9 100644 --- a/contrib/libs/pcre/pcre_globals.c +++ b/contrib/libs/pcre/pcre_globals.c @@ -1,50 +1,50 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2014 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains global variables that are exported by the PCRE library. -PCRE is thread-clean and doesn't use any global variables in the normal sense. -However, it calls memory allocation and freeing functions via the four -indirections below, and it can optionally do callouts, using the fifth -indirection. These values can be changed by the caller, but are shared between + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains global variables that are exported by the PCRE library. +PCRE is thread-clean and doesn't use any global variables in the normal sense. +However, it calls memory allocation and freeing functions via the four +indirections below, and it can optionally do callouts, using the fifth +indirection. These values can be changed by the caller, but are shared between all threads. - + For MS Visual Studio and Symbian OS, there are problems in initializing these variables to non-local functions. In these cases, therefore, an indirection via a local function is used. @@ -52,12 +52,12 @@ a local function is used. Also, when compiling for Virtual Pascal, things are done differently, and global variables are not used. */ -#ifdef HAVE_CONFIG_H +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - +#endif + +#include "pcre_internal.h" + #if defined _MSC_VER || defined __SYMBIAN32__ static void* LocalPcreMalloc(size_t aSize) { @@ -81,6 +81,6 @@ PCRE_EXP_DATA_DEFN void *(*PUBL(stack_malloc))(size_t) = malloc; PCRE_EXP_DATA_DEFN void (*PUBL(stack_free))(void *) = free; PCRE_EXP_DATA_DEFN int (*PUBL(callout))(PUBL(callout_block) *) = NULL; PCRE_EXP_DATA_DEFN int (*PUBL(stack_guard))(void) = NULL; -#endif - -/* End of pcre_globals.c */ +#endif + +/* End of pcre_globals.c */ diff --git a/contrib/libs/pcre/pcre_internal.h b/contrib/libs/pcre/pcre_internal.h index 97ff55d03b..7d7d2c4bac 100644 --- a/contrib/libs/pcre/pcre_internal.h +++ b/contrib/libs/pcre/pcre_internal.h @@ -1,57 +1,57 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2016 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - -/* This header contains definitions that are shared between the different -modules, but which are not relevant to the exported API. This includes some + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + +/* This header contains definitions that are shared between the different +modules, but which are not relevant to the exported API. This includes some functions whose names all begin with "_pcre_", "_pcre16_" or "_pcre32_" depending on the PRIV macro. */ - -#ifndef PCRE_INTERNAL_H -#define PCRE_INTERNAL_H - + +#ifndef PCRE_INTERNAL_H +#define PCRE_INTERNAL_H + /* Define PCRE_DEBUG to get debugging output on stdout. */ - -#if 0 + +#if 0 #define PCRE_DEBUG -#endif - +#endif + /* PCRE is compiled as an 8 bit library if it is not requested otherwise. */ #if !defined COMPILE_PCRE16 && !defined COMPILE_PCRE32 @@ -86,66 +86,66 @@ script prevents both being selected, but not everybody uses "configure". */ #error The use of both EBCDIC and SUPPORT_UTF is not supported. #endif -/* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef -inline, and there are *still* stupid compilers about that don't like indented -pre-processor statements, or at least there were when I first wrote this. After -all, it had only been about 10 years then... - -It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so -be absolutely sure we get our version. */ - -#undef DPRINTF +/* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef +inline, and there are *still* stupid compilers about that don't like indented +pre-processor statements, or at least there were when I first wrote this. After +all, it had only been about 10 years then... + +It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so +be absolutely sure we get our version. */ + +#undef DPRINTF #ifdef PCRE_DEBUG -#define DPRINTF(p) printf p -#else -#define DPRINTF(p) /* Nothing */ -#endif - - -/* Standard C headers plus the external interface definition. The only time -setjmp and stdarg are used is when NO_RECURSE is set. */ - -#include <ctype.h> -#include <limits.h> -#include <stddef.h> -#include <stdio.h> -#include <stdlib.h> -#include <string.h> - +#define DPRINTF(p) printf p +#else +#define DPRINTF(p) /* Nothing */ +#endif + + +/* Standard C headers plus the external interface definition. The only time +setjmp and stdarg are used is when NO_RECURSE is set. */ + +#include <ctype.h> +#include <limits.h> +#include <stddef.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + /* Valgrind (memcheck) support */ #ifdef SUPPORT_VALGRIND #include <valgrind/memcheck.h> #endif -/* When compiling a DLL for Windows, the exported symbols have to be declared -using some MS magic. I found some useful information on this web page: -http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the -information there, using __declspec(dllexport) without "extern" we have a -definition; with "extern" we have a declaration. The settings here override the -setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL, -which is all that is needed for applications (they just import the symbols). We -use: - - PCRE_EXP_DECL for declarations - PCRE_EXP_DEFN for definitions of exported functions - PCRE_EXP_DATA_DEFN for definitions of exported variables - -The reason for the two DEFN macros is that in non-Windows environments, one -does not want to have "extern" before variable definitions because it leads to -compiler warnings. So we distinguish between functions and variables. In -Windows, the two should always be the same. - -The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest, -which is an application, but needs to import this file in order to "peek" at -internals, can #include pcre.h first to get an application's-eye view. - -In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon, -special-purpose environments) might want to stick other stuff in front of -exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and -PCRE_EXP_DATA_DEFN only if they are not already set. */ - -#ifndef PCRE_EXP_DECL +/* When compiling a DLL for Windows, the exported symbols have to be declared +using some MS magic. I found some useful information on this web page: +http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the +information there, using __declspec(dllexport) without "extern" we have a +definition; with "extern" we have a declaration. The settings here override the +setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL, +which is all that is needed for applications (they just import the symbols). We +use: + + PCRE_EXP_DECL for declarations + PCRE_EXP_DEFN for definitions of exported functions + PCRE_EXP_DATA_DEFN for definitions of exported variables + +The reason for the two DEFN macros is that in non-Windows environments, one +does not want to have "extern" before variable definitions because it leads to +compiler warnings. So we distinguish between functions and variables. In +Windows, the two should always be the same. + +The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest, +which is an application, but needs to import this file in order to "peek" at +internals, can #include pcre.h first to get an application's-eye view. + +In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon, +special-purpose environments) might want to stick other stuff in front of +exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and +PCRE_EXP_DATA_DEFN only if they are not already set. */ + +#ifndef PCRE_EXP_DECL # ifdef _WIN32 # ifndef PCRE_STATIC # define PCRE_EXP_DECL extern __declspec(dllexport) @@ -169,8 +169,8 @@ PCRE_EXP_DATA_DEFN only if they are not already set. */ # define PCRE_EXP_DATA_DEFN # endif # endif -#endif - +#endif + /* When compiling with the MSVC compiler, it is sometimes necessary to include a "calling convention" before exported function names. (This is secondhand information; I know nothing about MSVC myself). For example, something like @@ -186,27 +186,27 @@ set, we ensure here that it has no effect. */ #endif /* We need to have types that specify unsigned 8, 16 and 32-bit integers. We -cannot determine these outside the compilation (e.g. by running a program as -part of "configure") because PCRE is often cross-compiled for use on other -systems. Instead we make use of the maximum sizes that are available at -preprocessor time in standard C environments. */ - +cannot determine these outside the compilation (e.g. by running a program as +part of "configure") because PCRE is often cross-compiled for use on other +systems. Instead we make use of the maximum sizes that are available at +preprocessor time in standard C environments. */ + typedef unsigned char pcre_uint8; -#if USHRT_MAX == 65535 +#if USHRT_MAX == 65535 typedef unsigned short pcre_uint16; typedef short pcre_int16; #define PCRE_UINT16_MAX USHRT_MAX #define PCRE_INT16_MAX SHRT_MAX -#elif UINT_MAX == 65535 +#elif UINT_MAX == 65535 typedef unsigned int pcre_uint16; typedef int pcre_int16; #define PCRE_UINT16_MAX UINT_MAX #define PCRE_INT16_MAX INT_MAX -#else +#else #error Cannot determine a type for 16-bit integers -#endif - +#endif + #if UINT_MAX == 4294967295U typedef unsigned int pcre_uint32; typedef int pcre_int32; @@ -217,10 +217,10 @@ typedef unsigned long int pcre_uint32; typedef long int pcre_int32; #define PCRE_UINT32_MAX ULONG_MAX #define PCRE_INT32_MAX LONG_MAX -#else +#else #error Cannot determine a type for 32-bit integers -#endif - +#endif + /* When checking for integer overflow in pcre_compile(), we need to handle large integers. If a 64-bit integer type is available, we can use that. Otherwise we have to cast to double, which of course requires floating point @@ -241,16 +241,16 @@ by "configure". */ #define INT64_OR_DOUBLE double #endif -/* All character handling must be done as unsigned characters. Otherwise there -are problems with top-bit-set characters and functions such as isspace(). +/* All character handling must be done as unsigned characters. Otherwise there +are problems with top-bit-set characters and functions such as isspace(). However, we leave the interface to the outside world as char * or short *, because that should make things easier for callers. This character type is called pcre_uchar. - + The IN_UCHARS macro multiply its argument with the byte size of the current pcre_uchar type. Useful for memcpy and such operations, whose require the byte size of their input/output buffers. - + The MAX_255 macro checks whether its pcre_uchar input is less than 256. The TABLE_GET macro is designed for accessing elements of tables whose contain @@ -291,182 +291,182 @@ typedef pcre_uint32 pcre_uchar; #error Unsupported compiling mode #endif /* COMPILE_PCRE[8|16|32] */ -/* This is an unsigned int value that no character can ever have. UTF-8 -characters only go up to 0x7fffffff (though Unicode doesn't go beyond -0x0010ffff). */ - -#define NOTACHAR 0xffffffff - -/* PCRE is able to support several different kinds of newline (CR, LF, CRLF, -"any" and "anycrlf" at present). The following macros are used to package up -testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various -modules to indicate in which datablock the parameters exist, and what the -start/end of string field names are. */ - -#define NLTYPE_FIXED 0 /* Newline is a fixed length string */ -#define NLTYPE_ANY 1 /* Newline is any Unicode line ending */ -#define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */ - -/* This macro checks for a newline at the given position */ - -#define IS_NEWLINE(p) \ - ((NLBLOCK->nltype != NLTYPE_FIXED)? \ - ((p) < NLBLOCK->PSEND && \ +/* This is an unsigned int value that no character can ever have. UTF-8 +characters only go up to 0x7fffffff (though Unicode doesn't go beyond +0x0010ffff). */ + +#define NOTACHAR 0xffffffff + +/* PCRE is able to support several different kinds of newline (CR, LF, CRLF, +"any" and "anycrlf" at present). The following macros are used to package up +testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various +modules to indicate in which datablock the parameters exist, and what the +start/end of string field names are. */ + +#define NLTYPE_FIXED 0 /* Newline is a fixed length string */ +#define NLTYPE_ANY 1 /* Newline is any Unicode line ending */ +#define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */ + +/* This macro checks for a newline at the given position */ + +#define IS_NEWLINE(p) \ + ((NLBLOCK->nltype != NLTYPE_FIXED)? \ + ((p) < NLBLOCK->PSEND && \ PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \ &(NLBLOCK->nllen), utf)) \ - : \ - ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \ + : \ + ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \ UCHAR21TEST(p) == NLBLOCK->nl[0] && \ (NLBLOCK->nllen == 1 || UCHAR21TEST(p+1) == NLBLOCK->nl[1]) \ - ) \ - ) - -/* This macro checks for a newline immediately preceding the given position */ - -#define WAS_NEWLINE(p) \ - ((NLBLOCK->nltype != NLTYPE_FIXED)? \ - ((p) > NLBLOCK->PSSTART && \ + ) \ + ) + +/* This macro checks for a newline immediately preceding the given position */ + +#define WAS_NEWLINE(p) \ + ((NLBLOCK->nltype != NLTYPE_FIXED)? \ + ((p) > NLBLOCK->PSSTART && \ PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \ &(NLBLOCK->nllen), utf)) \ - : \ - ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \ + : \ + ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \ UCHAR21TEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] && \ (NLBLOCK->nllen == 1 || UCHAR21TEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \ - ) \ - ) - -/* When PCRE is compiled as a C++ library, the subject pointer can be replaced -with a custom type. This makes it possible, for example, to allow pcre_exec() -to process subject strings that are discontinuous by using a smart pointer -class. It must always be possible to inspect all of the subject string in -pcre_exec() because of the way it backtracks. Two macros are required in the -normal case, for sign-unspecified and unsigned char pointers. The former is -used for the external interface and appears in pcre.h, which is why its name -must begin with PCRE_. */ - -#ifdef CUSTOM_SUBJECT_PTR + ) \ + ) + +/* When PCRE is compiled as a C++ library, the subject pointer can be replaced +with a custom type. This makes it possible, for example, to allow pcre_exec() +to process subject strings that are discontinuous by using a smart pointer +class. It must always be possible to inspect all of the subject string in +pcre_exec() because of the way it backtracks. Two macros are required in the +normal case, for sign-unspecified and unsigned char pointers. The former is +used for the external interface and appears in pcre.h, which is why its name +must begin with PCRE_. */ + +#ifdef CUSTOM_SUBJECT_PTR #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR -#else +#else #define PCRE_PUCHAR const pcre_uchar * -#endif - -/* Include the public PCRE header and the definitions of UCP character property -values. */ - -#include "pcre.h" -#include "ucp.h" - +#endif + +/* Include the public PCRE header and the definitions of UCP character property +values. */ + +#include "pcre.h" +#include "ucp.h" + #ifdef COMPILE_PCRE32 /* Assert that the public PCRE_UCHAR32 is a 32-bit type */ typedef int __assert_pcre_uchar32_size[sizeof(PCRE_UCHAR32) == 4 ? 1 : -1]; #endif -/* When compiling for use with the Virtual Pascal compiler, these functions -need to have their names changed. PCRE must be compiled with the -DVPCOMPAT -option on the command line. */ - -#ifdef VPCOMPAT -#define strlen(s) _strlen(s) -#define strncmp(s1,s2,m) _strncmp(s1,s2,m) -#define memcmp(s,c,n) _memcmp(s,c,n) -#define memcpy(d,s,n) _memcpy(d,s,n) -#define memmove(d,s,n) _memmove(d,s,n) -#define memset(s,c,n) _memset(s,c,n) -#else /* VPCOMPAT */ - -/* To cope with SunOS4 and other systems that lack memmove() but have bcopy(), -define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY -is set. Otherwise, include an emulating function for those systems that have -neither (there some non-Unix environments where this is the case). */ - -#ifndef HAVE_MEMMOVE -#undef memmove /* some systems may have a macro */ -#ifdef HAVE_BCOPY -#define memmove(a, b, c) bcopy(b, a, c) -#else /* HAVE_BCOPY */ -static void * -pcre_memmove(void *d, const void *s, size_t n) -{ -size_t i; -unsigned char *dest = (unsigned char *)d; -const unsigned char *src = (const unsigned char *)s; -if (dest > src) - { - dest += n; - src += n; - for (i = 0; i < n; ++i) *(--dest) = *(--src); - return (void *)dest; - } -else - { - for (i = 0; i < n; ++i) *dest++ = *src++; - return (void *)(dest - n); - } -} -#define memmove(a, b, c) pcre_memmove(a, b, c) -#endif /* not HAVE_BCOPY */ -#endif /* not HAVE_MEMMOVE */ -#endif /* not VPCOMPAT */ - - -/* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored -in big-endian order) by default. These are used, for example, to link from the -start of a subpattern to its alternatives and its end. The use of 2 bytes per -offset limits the size of the compiled regex to around 64K, which is big enough -for almost everybody. However, I received a request for an even bigger limit. -For this reason, and also to make the code easier to maintain, the storing and -loading of offsets from the byte string is now handled by the macros that are -defined here. - -The macros are controlled by the value of LINK_SIZE. This defaults to 2 in -the config.h file, but can be overridden by using -D on the command line. This -is automated on Unix systems via the "configure" command. */ - +/* When compiling for use with the Virtual Pascal compiler, these functions +need to have their names changed. PCRE must be compiled with the -DVPCOMPAT +option on the command line. */ + +#ifdef VPCOMPAT +#define strlen(s) _strlen(s) +#define strncmp(s1,s2,m) _strncmp(s1,s2,m) +#define memcmp(s,c,n) _memcmp(s,c,n) +#define memcpy(d,s,n) _memcpy(d,s,n) +#define memmove(d,s,n) _memmove(d,s,n) +#define memset(s,c,n) _memset(s,c,n) +#else /* VPCOMPAT */ + +/* To cope with SunOS4 and other systems that lack memmove() but have bcopy(), +define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY +is set. Otherwise, include an emulating function for those systems that have +neither (there some non-Unix environments where this is the case). */ + +#ifndef HAVE_MEMMOVE +#undef memmove /* some systems may have a macro */ +#ifdef HAVE_BCOPY +#define memmove(a, b, c) bcopy(b, a, c) +#else /* HAVE_BCOPY */ +static void * +pcre_memmove(void *d, const void *s, size_t n) +{ +size_t i; +unsigned char *dest = (unsigned char *)d; +const unsigned char *src = (const unsigned char *)s; +if (dest > src) + { + dest += n; + src += n; + for (i = 0; i < n; ++i) *(--dest) = *(--src); + return (void *)dest; + } +else + { + for (i = 0; i < n; ++i) *dest++ = *src++; + return (void *)(dest - n); + } +} +#define memmove(a, b, c) pcre_memmove(a, b, c) +#endif /* not HAVE_BCOPY */ +#endif /* not HAVE_MEMMOVE */ +#endif /* not VPCOMPAT */ + + +/* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored +in big-endian order) by default. These are used, for example, to link from the +start of a subpattern to its alternatives and its end. The use of 2 bytes per +offset limits the size of the compiled regex to around 64K, which is big enough +for almost everybody. However, I received a request for an even bigger limit. +For this reason, and also to make the code easier to maintain, the storing and +loading of offsets from the byte string is now handled by the macros that are +defined here. + +The macros are controlled by the value of LINK_SIZE. This defaults to 2 in +the config.h file, but can be overridden by using -D on the command line. This +is automated on Unix systems via the "configure" command. */ + #if defined COMPILE_PCRE8 -#if LINK_SIZE == 2 - -#define PUT(a,n,d) \ - (a[n] = (d) >> 8), \ - (a[(n)+1] = (d) & 255) - -#define GET(a,n) \ - (((a)[n] << 8) | (a)[(n)+1]) - -#define MAX_PATTERN_SIZE (1 << 16) - - -#elif LINK_SIZE == 3 - -#define PUT(a,n,d) \ - (a[n] = (d) >> 16), \ - (a[(n)+1] = (d) >> 8), \ - (a[(n)+2] = (d) & 255) - -#define GET(a,n) \ - (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2]) - -#define MAX_PATTERN_SIZE (1 << 24) - - -#elif LINK_SIZE == 4 - -#define PUT(a,n,d) \ - (a[n] = (d) >> 24), \ - (a[(n)+1] = (d) >> 16), \ - (a[(n)+2] = (d) >> 8), \ - (a[(n)+3] = (d) & 255) - -#define GET(a,n) \ - (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3]) - +#if LINK_SIZE == 2 + +#define PUT(a,n,d) \ + (a[n] = (d) >> 8), \ + (a[(n)+1] = (d) & 255) + +#define GET(a,n) \ + (((a)[n] << 8) | (a)[(n)+1]) + +#define MAX_PATTERN_SIZE (1 << 16) + + +#elif LINK_SIZE == 3 + +#define PUT(a,n,d) \ + (a[n] = (d) >> 16), \ + (a[(n)+1] = (d) >> 8), \ + (a[(n)+2] = (d) & 255) + +#define GET(a,n) \ + (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2]) + +#define MAX_PATTERN_SIZE (1 << 24) + + +#elif LINK_SIZE == 4 + +#define PUT(a,n,d) \ + (a[n] = (d) >> 24), \ + (a[(n)+1] = (d) >> 16), \ + (a[(n)+2] = (d) >> 8), \ + (a[(n)+3] = (d) & 255) + +#define GET(a,n) \ + (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3]) + /* Keep it positive */ #define MAX_PATTERN_SIZE (1 << 30) - + #else #error LINK_SIZE must be either 2, 3, or 4 #endif - + #elif defined COMPILE_PCRE16 #if LINK_SIZE == 2 @@ -499,12 +499,12 @@ is automated on Unix systems via the "configure" command. */ /* Keep it positive */ #define MAX_PATTERN_SIZE (1 << 30) -#else -#error LINK_SIZE must be either 2, 3, or 4 -#endif - +#else +#error LINK_SIZE must be either 2, 3, or 4 +#endif + #elif defined COMPILE_PCRE32 - + /* Only supported LINK_SIZE is 4 */ /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */ #undef LINK_SIZE @@ -523,37 +523,37 @@ is automated on Unix systems via the "configure" command. */ #error Unsupported compiling mode #endif /* COMPILE_PCRE[8|16|32] */ -/* Convenience macro defined in terms of the others */ - -#define PUTINC(a,n,d) PUT(a,n,d), a += LINK_SIZE - - -/* PCRE uses some other 2-byte quantities that do not change when the size of -offsets changes. There are used for repeat counts and for other things such as -capturing parenthesis numbers in back references. */ - +/* Convenience macro defined in terms of the others */ + +#define PUTINC(a,n,d) PUT(a,n,d), a += LINK_SIZE + + +/* PCRE uses some other 2-byte quantities that do not change when the size of +offsets changes. There are used for repeat counts and for other things such as +capturing parenthesis numbers in back references. */ + #if defined COMPILE_PCRE8 #define IMM2_SIZE 2 -#define PUT2(a,n,d) \ - a[n] = (d) >> 8; \ - a[(n)+1] = (d) & 255 - +#define PUT2(a,n,d) \ + a[n] = (d) >> 8; \ + a[(n)+1] = (d) & 255 + /* For reasons that I do not understand, the expression in this GET2 macro is treated by gcc as a signed expression, even when a is declared as unsigned. It seems that any kind of arithmetic results in a signed value. */ -#define GET2(a,n) \ +#define GET2(a,n) \ (unsigned int)(((a)[n] << 8) | (a)[(n)+1]) - + #elif defined COMPILE_PCRE16 - + #define IMM2_SIZE 1 - + #define PUT2(a,n,d) \ a[n] = d - + #define GET2(a,n) \ a[n] @@ -610,27 +610,27 @@ UTF support is omitted, we don't even define them. */ /* #define HAS_EXTRALEN(c) */ /* #define GET_EXTRALEN(c) */ /* #define NOT_FIRSTCHAR(c) */ -#define GETCHAR(c, eptr) c = *eptr; -#define GETCHARTEST(c, eptr) c = *eptr; -#define GETCHARINC(c, eptr) c = *eptr++; -#define GETCHARINCTEST(c, eptr) c = *eptr++; -#define GETCHARLEN(c, eptr, len) c = *eptr; +#define GETCHAR(c, eptr) c = *eptr; +#define GETCHARTEST(c, eptr) c = *eptr; +#define GETCHARINC(c, eptr) c = *eptr++; +#define GETCHARINCTEST(c, eptr) c = *eptr++; +#define GETCHARLEN(c, eptr, len) c = *eptr; /* #define GETCHARLENTEST(c, eptr, len) */ -/* #define BACKCHAR(eptr) */ +/* #define BACKCHAR(eptr) */ /* #define FORWARDCHAR(eptr) */ /* #define ACROSSCHAR(condition, eptr, action) */ - + #else /* SUPPORT_UTF */ - + /* Tests whether the code point needs extra characters to decode. */ - + #define HASUTF8EXTRALEN(c) ((c) >= 0xc0) - + /* Base macro to pick up the remaining bytes of a UTF-8 character, not advancing the pointer. */ - + #define GETUTF8(c, eptr) \ - { \ + { \ if ((c & 0x20) == 0) \ c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \ else if ((c & 0x10) == 0) \ @@ -656,10 +656,10 @@ the pointer. */ if ((c & 0x20) == 0) \ c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \ else if ((c & 0x10) == 0) \ - { \ + { \ c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \ eptr += 2; \ - } \ + } \ else if ((c & 0x08) == 0) \ { \ c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \ @@ -680,8 +680,8 @@ the pointer. */ ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \ eptr += 5; \ } \ - } - + } + #if defined COMPILE_PCRE8 /* These macros were originally written in the form of loops that used data @@ -714,37 +714,37 @@ we know we are in UTF-8 mode. */ c = *eptr; \ if (c >= 0xc0) GETUTF8(c, eptr); -/* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the -pointer. */ - -#define GETCHARTEST(c, eptr) \ - c = *eptr; \ +/* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the +pointer. */ + +#define GETCHARTEST(c, eptr) \ + c = *eptr; \ if (utf && c >= 0xc0) GETUTF8(c, eptr); - -/* Get the next UTF-8 character, advancing the pointer. This is called when we -know we are in UTF-8 mode. */ - -#define GETCHARINC(c, eptr) \ - c = *eptr++; \ + +/* Get the next UTF-8 character, advancing the pointer. This is called when we +know we are in UTF-8 mode. */ + +#define GETCHARINC(c, eptr) \ + c = *eptr++; \ if (c >= 0xc0) GETUTF8INC(c, eptr); - + /* Get the next character, testing for UTF-8 mode, and advancing the pointer. This is called when we don't know if we are in UTF-8 mode. */ - -#define GETCHARINCTEST(c, eptr) \ - c = *eptr++; \ + +#define GETCHARINCTEST(c, eptr) \ + c = *eptr++; \ if (utf && c >= 0xc0) GETUTF8INC(c, eptr); /* Base macro to pick up the remaining bytes of a UTF-8 character, not advancing the pointer, incrementing the length. */ #define GETUTF8LEN(c, eptr, len) \ - { \ + { \ if ((c & 0x20) == 0) \ - { \ + { \ c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \ len++; \ - } \ + } \ else if ((c & 0x10) == 0) \ { \ c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \ @@ -770,15 +770,15 @@ advancing the pointer, incrementing the length. */ ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \ len += 5; \ } \ - } - -/* Get the next UTF-8 character, not advancing the pointer, incrementing length -if there are extra bytes. This is called when we know we are in UTF-8 mode. */ - -#define GETCHARLEN(c, eptr, len) \ - c = *eptr; \ + } + +/* Get the next UTF-8 character, not advancing the pointer, incrementing length +if there are extra bytes. This is called when we know we are in UTF-8 mode. */ + +#define GETCHARLEN(c, eptr, len) \ + c = *eptr; \ if (c >= 0xc0) GETUTF8LEN(c, eptr, len); - + /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the pointer, incrementing length if there are extra bytes. This is called when we do not know if we are in UTF-8 mode. */ @@ -787,21 +787,21 @@ do not know if we are in UTF-8 mode. */ c = *eptr; \ if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len); -/* If the pointer is not at the start of a character, move it back until -it is. This is called only in UTF-8 mode - we don't put a test within the macro -because almost all calls are already within a block of UTF-8 only code. */ - -#define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr-- - +/* If the pointer is not at the start of a character, move it back until +it is. This is called only in UTF-8 mode - we don't put a test within the macro +because almost all calls are already within a block of UTF-8 only code. */ + +#define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr-- + /* Same as above, just in the other direction. */ #define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++ - + /* Same as above, but it allows a fully customizable form. */ #define ACROSSCHAR(condition, eptr, action) \ while((condition) && ((eptr) & 0xc0) == 0x80) action - + #elif defined COMPILE_PCRE16 - + /* Tells the biggest code point which can be encoded as a single character. */ #define MAX_VALUE_FOR_SINGLE_CHAR 65535 @@ -1052,29 +1052,29 @@ other. NOTE: The values also appear in pcre_jit_compile.c. */ #else #define VSPACE_LIST \ CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR -#endif - +#endif + #define VSPACE_BYTE_CASES \ case CHAR_LF: \ case CHAR_VT: \ case CHAR_FF: \ case CHAR_CR: \ case CHAR_NEL - + #define VSPACE_CASES VSPACE_BYTE_CASES #endif /* EBCDIC */ - + /* ------ End of whitespace macros ------ */ + - -/* Private flags containing information about the compiled regex. They used to +/* Private flags containing information about the compiled regex. They used to live at the top end of the options word, but that got almost full, so they were moved to a 16-bit flags word - which got almost full, so now they are in a 32-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as the restrictions on partial matching have been lifted. It remains for backwards compatibility. */ - + #define PCRE_MODE8 0x00000001 /* compiled in 8 bit mode */ #define PCRE_MODE16 0x00000002 /* compiled in 16 bit mode */ #define PCRE_MODE32 0x00000004 /* compiled in 32 bit mode */ @@ -1090,7 +1090,7 @@ compatibility. */ #define PCRE_MLSET 0x00002000 /* match limit set by regex */ #define PCRE_RLSET 0x00004000 /* recursion limit set by regex */ #define PCRE_MATCH_EMPTY 0x00008000 /* pattern can match empty string */ - + #if defined COMPILE_PCRE8 #define PCRE_MODE PCRE_MODE8 #elif defined COMPILE_PCRE16 @@ -1099,70 +1099,70 @@ compatibility. */ #define PCRE_MODE PCRE_MODE32 #endif #define PCRE_MODE_MASK (PCRE_MODE8 | PCRE_MODE16 | PCRE_MODE32) - + /* Flags for the "extra" block produced by pcre_study(). */ - + #define PCRE_STUDY_MAPPED 0x0001 /* a map of starting chars exists */ #define PCRE_STUDY_MINLEN 0x0002 /* a minimum length field exists */ -/* Masks for identifying the public options that are permitted at compile -time, run time, or study time, respectively. */ - -#define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \ - PCRE_NEWLINE_ANYCRLF) - +/* Masks for identifying the public options that are permitted at compile +time, run time, or study time, respectively. */ + +#define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \ + PCRE_NEWLINE_ANYCRLF) + #define PUBLIC_COMPILE_OPTIONS \ - (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \ - PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \ + (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \ + PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \ PCRE_NO_AUTO_CAPTURE|PCRE_NO_AUTO_POSSESS| \ PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \ PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \ PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE|PCRE_NEVER_UTF) - -#define PUBLIC_EXEC_OPTIONS \ + +#define PUBLIC_EXEC_OPTIONS \ (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \ PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \ PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE) - -#define PUBLIC_DFA_EXEC_OPTIONS \ + +#define PUBLIC_DFA_EXEC_OPTIONS \ (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \ PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \ PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \ PCRE_NO_START_OPTIMIZE) - + #define PUBLIC_STUDY_OPTIONS \ (PCRE_STUDY_JIT_COMPILE|PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE| \ PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE|PCRE_STUDY_EXTRA_NEEDED) - + #define PUBLIC_JIT_EXEC_OPTIONS \ (PCRE_NO_UTF8_CHECK|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|\ PCRE_NOTEMPTY_ATSTART|PCRE_PARTIAL_SOFT|PCRE_PARTIAL_HARD) - + /* Magic number to provide a small check against being handed junk. */ -#define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */ - +#define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */ + /* This variable is used to detect a loaded regular expression in different endianness. */ - + #define REVERSED_MAGIC_NUMBER 0x45524350UL /* 'ERCP' */ - -/* The maximum remaining length of subject we are prepared to search for a -req_byte match. */ - -#define REQ_BYTE_MAX 1000 - + +/* The maximum remaining length of subject we are prepared to search for a +req_byte match. */ + +#define REQ_BYTE_MAX 1000 + /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in environments where these macros are defined elsewhere. Unfortunately, there is no way to do the same for the typedef. */ - -typedef int BOOL; - + +typedef int BOOL; + #ifndef FALSE -#define FALSE 0 -#define TRUE 1 +#define FALSE 0 +#define TRUE 1 #endif - + /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal character constants like '*' because the compiler would emit their EBCDIC code, which is different from their ASCII/UTF-8 code. Instead we define macros for @@ -1764,42 +1764,42 @@ only. */ #endif /* SUPPORT_UTF */ -/* Escape items that are just an encoding of a particular data value. */ - +/* Escape items that are just an encoding of a particular data value. */ + #ifndef ESC_a #define ESC_a CHAR_BEL #endif -#ifndef ESC_e +#ifndef ESC_e #define ESC_e CHAR_ESC -#endif - -#ifndef ESC_f +#endif + +#ifndef ESC_f #define ESC_f CHAR_FF -#endif - -#ifndef ESC_n +#endif + +#ifndef ESC_n #define ESC_n CHAR_LF -#endif - -#ifndef ESC_r +#endif + +#ifndef ESC_r #define ESC_r CHAR_CR -#endif - -/* We can't officially use ESC_t because it is a POSIX reserved identifier -(presumably because of all the others like size_t). */ - -#ifndef ESC_tee +#endif + +/* We can't officially use ESC_t because it is a POSIX reserved identifier +(presumably because of all the others like size_t). */ + +#ifndef ESC_tee #define ESC_tee CHAR_HT -#endif - -/* Codes for different types of Unicode property */ - -#define PT_ANY 0 /* Any property - matches all chars */ -#define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */ +#endif + +/* Codes for different types of Unicode property */ + +#define PT_ANY 0 /* Any property - matches all chars */ +#define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */ #define PT_GC 2 /* Specified general characteristic (e.g. L) */ #define PT_PC 3 /* Specified particular characteristic (e.g. Lu) */ -#define PT_SC 4 /* Script (e.g. Han) */ +#define PT_SC 4 /* Script (e.g. Han) */ #define PT_ALNUM 5 /* Alphanumeric - the union of L and N */ #define PT_SPACE 6 /* Perl space - Z plus 9,10,12,13 */ #define PT_PXSPACE 7 /* POSIX space - Z plus 9,10,11,12,13 */ @@ -1807,7 +1807,7 @@ only. */ #define PT_CLIST 9 /* Pseudo-property: match character list */ #define PT_UCNC 10 /* Universal Character nameable character */ #define PT_TABSIZE 11 /* Size of square table for autopossessify tests */ - + /* The following special properties are used only in XCLASS items, when POSIX classes are specified and PCRE_UCP is set - in other words, for Unicode handling of these classes. They are not available via the \p or \P escapes like @@ -1818,27 +1818,27 @@ table. */ #define PT_PXPRINT 12 /* [:print:] - [:graph:] plus non-control spaces */ #define PT_PXPUNCT 13 /* [:punct:] - punctuation characters */ -/* Flag bits and data types for the extended class (OP_XCLASS) for classes that +/* Flag bits and data types for the extended class (OP_XCLASS) for classes that contain characters with values greater than 255. */ - + #define XCL_NOT 0x01 /* Flag: this is a negative class */ #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */ #define XCL_HASPROP 0x04 /* Flag: property checks are present. */ - -#define XCL_END 0 /* Marks end of individual items */ -#define XCL_SINGLE 1 /* Single item (one multibyte char) follows */ -#define XCL_RANGE 2 /* A range (two multibyte chars) follows */ -#define XCL_PROP 3 /* Unicode property (2-byte property code follows) */ -#define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */ - -/* These are escaped items that aren't just an encoding of a particular data + +#define XCL_END 0 /* Marks end of individual items */ +#define XCL_SINGLE 1 /* Single item (one multibyte char) follows */ +#define XCL_RANGE 2 /* A range (two multibyte chars) follows */ +#define XCL_PROP 3 /* Unicode property (2-byte property code follows) */ +#define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */ + +/* These are escaped items that aren't just an encoding of a particular data value such as \n. They must have non-zero values, as check_escape() returns 0 for a data character. Also, they must appear in the same order as in the opcode definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it corresponds to "." in DOTALL mode rather than an escape sequence. It is also used for [^] in JavaScript compatibility mode, and for \C in non-utf mode. In non-DOTALL mode, "." behaves like \N. - + The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc. when PCRE_UCP is set and replacement of \d etc by \p sequences is required. They must be contiguous, and remain in order so that the replacements can be @@ -1851,17 +1851,17 @@ repeated. These are the types that consume characters. If any new escapes are put in between that don't consume a character, that code will have to change. */ -enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s, +enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s, ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H, ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, ESC_E, ESC_Q, ESC_g, ESC_k, ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu }; - - + + /********************** Opcode definitions ******************/ - + /****** NOTE NOTE NOTE ****** - + Starting from 1 (i.e. after OP_END), the values up to OP_EOD must correspond in order to the list of escapes immediately above. Furthermore, values up to OP_DOLLM must not be changed without adjusting the table called autoposstab in @@ -1883,23 +1883,23 @@ auto-possessified. */ #define LAST_AUTOTAB_LEFT_OP OP_EXTUNI #define LAST_AUTOTAB_RIGHT_OP OP_DOLLM -enum { - OP_END, /* 0 End of pattern */ - - /* Values corresponding to backslashed metacharacters */ - - OP_SOD, /* 1 Start of data: \A */ - OP_SOM, /* 2 Start of match (subject + offset): \G */ - OP_SET_SOM, /* 3 Set start of match (\K) */ - OP_NOT_WORD_BOUNDARY, /* 4 \B */ - OP_WORD_BOUNDARY, /* 5 \b */ - OP_NOT_DIGIT, /* 6 \D */ - OP_DIGIT, /* 7 \d */ - OP_NOT_WHITESPACE, /* 8 \S */ - OP_WHITESPACE, /* 9 \s */ - OP_NOT_WORDCHAR, /* 10 \W */ - OP_WORDCHAR, /* 11 \w */ - +enum { + OP_END, /* 0 End of pattern */ + + /* Values corresponding to backslashed metacharacters */ + + OP_SOD, /* 1 Start of data: \A */ + OP_SOM, /* 2 Start of match (subject + offset): \G */ + OP_SET_SOM, /* 3 Set start of match (\K) */ + OP_NOT_WORD_BOUNDARY, /* 4 \B */ + OP_WORD_BOUNDARY, /* 5 \b */ + OP_NOT_DIGIT, /* 6 \D */ + OP_DIGIT, /* 7 \d */ + OP_NOT_WHITESPACE, /* 8 \S */ + OP_WHITESPACE, /* 9 \s */ + OP_NOT_WORDCHAR, /* 10 \W */ + OP_WORDCHAR, /* 11 \w */ + OP_ANY, /* 12 Match any character except newline (\N) */ OP_ALLANY, /* 13 Match any character */ OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */ @@ -1913,81 +1913,81 @@ enum { OP_EXTUNI, /* 22 \X (extended Unicode sequence */ OP_EODN, /* 23 End of data or \n at end of data (\Z) */ OP_EOD, /* 24 End of data (\z) */ - + /* Line end assertions */ - + OP_DOLL, /* 25 End of line - not multiline */ OP_DOLLM, /* 26 End of line - multiline */ OP_CIRC, /* 27 Start of line - not multiline */ OP_CIRCM, /* 28 Start of line - multiline */ - + /* Single characters; caseful must precede the caseless ones */ - + OP_CHAR, /* 29 Match one character, casefully */ OP_CHARI, /* 30 Match one character, caselessly */ OP_NOT, /* 31 Match one character, not the given one, casefully */ OP_NOTI, /* 32 Match one character, not the given one, caselessly */ - + /* The following sets of 13 opcodes must always be kept in step because the offset from the first one is used to generate the others. */ - + /* Repeated characters; caseful must precede the caseless ones */ - + OP_STAR, /* 33 The maximizing and minimizing versions of */ OP_MINSTAR, /* 34 these six opcodes must come in pairs, with */ OP_PLUS, /* 35 the minimizing one second. */ OP_MINPLUS, /* 36 */ OP_QUERY, /* 37 */ OP_MINQUERY, /* 38 */ - + OP_UPTO, /* 39 From 0 to n matches of one character, caseful*/ OP_MINUPTO, /* 40 */ OP_EXACT, /* 41 Exactly n matches */ - + OP_POSSTAR, /* 42 Possessified star, caseful */ OP_POSPLUS, /* 43 Possessified plus, caseful */ OP_POSQUERY, /* 44 Posesssified query, caseful */ OP_POSUPTO, /* 45 Possessified upto, caseful */ - + /* Repeated characters; caseless must follow the caseful ones */ - + OP_STARI, /* 46 */ OP_MINSTARI, /* 47 */ OP_PLUSI, /* 48 */ OP_MINPLUSI, /* 49 */ OP_QUERYI, /* 50 */ OP_MINQUERYI, /* 51 */ - + OP_UPTOI, /* 52 From 0 to n matches of one character, caseless */ OP_MINUPTOI, /* 53 */ OP_EXACTI, /* 54 */ - + OP_POSSTARI, /* 55 Possessified star, caseless */ OP_POSPLUSI, /* 56 Possessified plus, caseless */ OP_POSQUERYI, /* 57 Posesssified query, caseless */ OP_POSUPTOI, /* 58 Possessified upto, caseless */ - + /* The negated ones must follow the non-negated ones, and match them */ /* Negated repeated character, caseful; must precede the caseless ones */ - + OP_NOTSTAR, /* 59 The maximizing and minimizing versions of */ OP_NOTMINSTAR, /* 60 these six opcodes must come in pairs, with */ OP_NOTPLUS, /* 61 the minimizing one second. They must be in */ OP_NOTMINPLUS, /* 62 exactly the same order as those above. */ OP_NOTQUERY, /* 63 */ OP_NOTMINQUERY, /* 64 */ - + OP_NOTUPTO, /* 65 From 0 to n matches, caseful */ OP_NOTMINUPTO, /* 66 */ OP_NOTEXACT, /* 67 Exactly n matches */ - + OP_NOTPOSSTAR, /* 68 Possessified versions, caseful */ OP_NOTPOSPLUS, /* 69 */ OP_NOTPOSQUERY, /* 70 */ OP_NOTPOSUPTO, /* 71 */ - + /* Negated repeated character, caseless; must follow the caseful ones */ - + OP_NOTSTARI, /* 72 */ OP_NOTMINSTARI, /* 73 */ OP_NOTPLUSI, /* 74 */ @@ -2084,28 +2084,28 @@ enum { OP_COND, /* 135 Conditional group */ /* These five must follow the previous five, in the same order. There's a - check for >= SBRA to distinguish the two sets. */ - + check for >= SBRA to distinguish the two sets. */ + OP_SBRA, /* 136 Start of non-capturing bracket, check empty */ OP_SBRAPOS, /* 137 Ditto, with unlimited, possessive repeat */ OP_SCBRA, /* 138 Start of capturing bracket, check empty */ OP_SCBRAPOS, /* 139 Ditto, with unlimited, possessive repeat */ OP_SCOND, /* 140 Conditional group, check empty */ - + /* The next two pairs must (respectively) be kept together. */ - + OP_CREF, /* 141 Used to hold a capture number as condition */ OP_DNCREF, /* 142 Used to point to duplicate names as a condition */ OP_RREF, /* 143 Used to hold a recursion number as condition */ OP_DNRREF, /* 144 Used to point to duplicate names as a condition */ OP_DEF, /* 145 The DEFINE condition */ - + OP_BRAZERO, /* 146 These two must remain together and in this */ OP_BRAMINZERO, /* 147 order. */ OP_BRAPOSZERO, /* 148 */ - /* These are backtracking control verbs */ - + /* These are backtracking control verbs */ + OP_MARK, /* 149 always has an argument */ OP_PRUNE, /* 150 */ OP_PRUNE_ARG, /* 151 same, but with argument */ @@ -2114,9 +2114,9 @@ enum { OP_THEN, /* 154 */ OP_THEN_ARG, /* 155 same, but with argument */ OP_COMMIT, /* 156 */ - - /* These are forced failure and success verbs */ - + + /* These are forced failure and success verbs */ + OP_FAIL, /* 157 */ OP_ACCEPT, /* 158 */ OP_ASSERT_ACCEPT, /* 159 Used inside assertions */ @@ -2131,40 +2131,40 @@ enum { some in the past. */ OP_TABLE_LENGTH -}; - +}; + /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro definitions that follow must also be updated to match. There are also tables called "opcode_possessify" in pcre_compile.c and "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */ + - -/* This macro defines textual names for all the opcodes. These are used only +/* This macro defines textual names for all the opcodes. These are used only for debugging, and some of them are only partial names. The macro is referenced only in pcre_printint.c, which fills out the full names in many cases (and in some cases doesn't actually use these names at all). */ - -#define OP_NAME_LIST \ - "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \ + +#define OP_NAME_LIST \ + "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \ "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \ - "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \ - "extuni", "\\Z", "\\z", \ + "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \ + "extuni", "\\Z", "\\z", \ "$", "$", "^", "^", "char", "chari", "not", "noti", \ "*", "*?", "+", "+?", "?", "??", \ "{", "{", "{", \ - "*+","++", "?+", "{", \ + "*+","++", "?+", "{", \ "*", "*?", "+", "+?", "?", "??", \ "{", "{", "{", \ - "*+","++", "?+", "{", \ + "*+","++", "?+", "{", \ "*", "*?", "+", "+?", "?", "??", \ "{", "{", "{", \ "*+","++", "?+", "{", \ "*", "*?", "+", "+?", "?", "??", \ "{", "{", "{", \ "*+","++", "?+", "{", \ - "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \ - "*+","++", "?+", "{", \ - "*", "*?", "+", "+?", "?", "??", "{", "{", \ + "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \ + "*+","++", "?+", "{", \ + "*", "*?", "+", "+?", "?", "??", "{", "{", \ "*+","++", "?+", "{", \ "class", "nclass", "xclass", "Ref", "Refi", "DnRef", "DnRefi", \ "Recurse", "Callout", \ @@ -2181,29 +2181,29 @@ some cases doesn't actually use these names at all). */ "*THEN", "*THEN", "*COMMIT", "*FAIL", \ "*ACCEPT", "*ASSERT_ACCEPT", \ "Close", "Skip zero" - - -/* This macro defines the length of fixed length operations in the compiled -regex. The lengths are used when searching for specific things, and also in the -debugging printing of a compiled regex. We use a macro so that it can be -defined close to the definitions of the opcodes themselves. - -As things have been extended, some of these are no longer fixed lenths, but are -minima instead. For example, the length of a single-character repeat may vary -in UTF-8 mode. The code that uses this table must know about such things. */ - -#define OP_LENGTHS \ - 1, /* End */ \ - 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \ - 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \ + + +/* This macro defines the length of fixed length operations in the compiled +regex. The lengths are used when searching for specific things, and also in the +debugging printing of a compiled regex. We use a macro so that it can be +defined close to the definitions of the opcodes themselves. + +As things have been extended, some of these are no longer fixed lenths, but are +minima instead. For example, the length of a single-character repeat may vary +in UTF-8 mode. The code that uses this table must know about such things. */ + +#define OP_LENGTHS \ + 1, /* End */ \ + 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \ + 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \ 1, 1, 1, /* Any, AllAny, Anybyte */ \ 3, 3, /* \P, \p */ \ - 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \ + 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \ 1, /* \X */ \ 1, 1, 1, 1, 1, 1, /* \Z, \z, $, $M ^, ^M */ \ - 2, /* Char - the minimum length */ \ + 2, /* Char - the minimum length */ \ 2, /* Chari - the minimum length */ \ - 2, /* not */ \ + 2, /* not */ \ 2, /* noti */ \ /* Positive single-char repeats ** These are */ \ 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \ @@ -2214,8 +2214,8 @@ in UTF-8 mode. The code that uses this table must know about such things. */ 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto I, minupto I */ \ 2+IMM2_SIZE, /* exact I */ \ 2, 2, 2, 2+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ \ - /* Negative single-char repeats - only for chars < 256 */ \ - 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \ + /* Negative single-char repeats - only for chars < 256 */ \ + 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \ 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto, minupto */ \ 2+IMM2_SIZE, /* NOT exact */ \ 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *, +, ?, upto */ \ @@ -2223,85 +2223,85 @@ in UTF-8 mode. The code that uses this table must know about such things. */ 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto I, minupto I */ \ 2+IMM2_SIZE, /* NOT exact I */ \ 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *I, +I, ?I, upto I */ \ - /* Positive type repeats */ \ - 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \ + /* Positive type repeats */ \ + 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \ 2+IMM2_SIZE, 2+IMM2_SIZE, /* Type upto, minupto */ \ 2+IMM2_SIZE, /* Type exact */ \ 2, 2, 2, 2+IMM2_SIZE, /* Possessive *+, ++, ?+, upto+ */ \ - /* Character class & ref repeats */ \ - 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \ + /* Character class & ref repeats */ \ + 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \ 1+2*IMM2_SIZE, 1+2*IMM2_SIZE, /* CRRANGE, CRMINRANGE */ \ 1, 1, 1, 1+2*IMM2_SIZE, /* Possessive *+, ++, ?+, CRPOSRANGE */ \ 1+(32/sizeof(pcre_uchar)), /* CLASS */ \ 1+(32/sizeof(pcre_uchar)), /* NCLASS */ \ - 0, /* XCLASS - variable length */ \ + 0, /* XCLASS - variable length */ \ 1+IMM2_SIZE, /* REF */ \ 1+IMM2_SIZE, /* REFI */ \ 1+2*IMM2_SIZE, /* DNREF */ \ 1+2*IMM2_SIZE, /* DNREFI */ \ - 1+LINK_SIZE, /* RECURSE */ \ - 2+2*LINK_SIZE, /* CALLOUT */ \ - 1+LINK_SIZE, /* Alt */ \ - 1+LINK_SIZE, /* Ket */ \ - 1+LINK_SIZE, /* KetRmax */ \ - 1+LINK_SIZE, /* KetRmin */ \ + 1+LINK_SIZE, /* RECURSE */ \ + 2+2*LINK_SIZE, /* CALLOUT */ \ + 1+LINK_SIZE, /* Alt */ \ + 1+LINK_SIZE, /* Ket */ \ + 1+LINK_SIZE, /* KetRmax */ \ + 1+LINK_SIZE, /* KetRmin */ \ 1+LINK_SIZE, /* KetRpos */ \ 1+LINK_SIZE, /* Reverse */ \ - 1+LINK_SIZE, /* Assert */ \ - 1+LINK_SIZE, /* Assert not */ \ - 1+LINK_SIZE, /* Assert behind */ \ - 1+LINK_SIZE, /* Assert behind not */ \ - 1+LINK_SIZE, /* ONCE */ \ + 1+LINK_SIZE, /* Assert */ \ + 1+LINK_SIZE, /* Assert not */ \ + 1+LINK_SIZE, /* Assert behind */ \ + 1+LINK_SIZE, /* Assert behind not */ \ + 1+LINK_SIZE, /* ONCE */ \ 1+LINK_SIZE, /* ONCE_NC */ \ - 1+LINK_SIZE, /* BRA */ \ + 1+LINK_SIZE, /* BRA */ \ 1+LINK_SIZE, /* BRAPOS */ \ 1+LINK_SIZE+IMM2_SIZE, /* CBRA */ \ 1+LINK_SIZE+IMM2_SIZE, /* CBRAPOS */ \ - 1+LINK_SIZE, /* COND */ \ - 1+LINK_SIZE, /* SBRA */ \ + 1+LINK_SIZE, /* COND */ \ + 1+LINK_SIZE, /* SBRA */ \ 1+LINK_SIZE, /* SBRAPOS */ \ 1+LINK_SIZE+IMM2_SIZE, /* SCBRA */ \ 1+LINK_SIZE+IMM2_SIZE, /* SCBRAPOS */ \ - 1+LINK_SIZE, /* SCOND */ \ + 1+LINK_SIZE, /* SCOND */ \ 1+IMM2_SIZE, 1+2*IMM2_SIZE, /* CREF, DNCREF */ \ 1+IMM2_SIZE, 1+2*IMM2_SIZE, /* RREF, DNRREF */ \ - 1, /* DEF */ \ + 1, /* DEF */ \ 1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \ 3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \ 1, 3, /* SKIP, SKIP_ARG */ \ 1, 3, /* THEN, THEN_ARG */ \ 1, 1, 1, 1, /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT */ \ 1+IMM2_SIZE, 1 /* CLOSE, SKIPZERO */ - -/* A magic value for OP_RREF to indicate the "any recursion" condition. */ - -#define RREF_ANY 0xffff - + +/* A magic value for OP_RREF to indicate the "any recursion" condition. */ + +#define RREF_ANY 0xffff + /* Compile time error code numbers. They are given names so that they can more easily be tracked. When a new number is added, the table called eint in pcreposix.c must be updated. */ - -enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9, - ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19, - ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29, - ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39, - ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49, - ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59, + +enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9, + ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19, + ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29, + ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39, + ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49, + ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59, ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69, ERR70, ERR71, ERR72, ERR73, ERR74, ERR75, ERR76, ERR77, ERR78, ERR79, ERR80, ERR81, ERR82, ERR83, ERR84, ERR85, ERR86, ERR87, ERRCOUNT }; - + /* JIT compiling modes. The function list is indexed by them. */ enum { JIT_COMPILE, JIT_PARTIAL_SOFT_COMPILE, JIT_PARTIAL_HARD_COMPILE, JIT_NUMBER_OF_COMPILE_MODES }; -/* The real format of the start of the pcre block; the index of names and the -code vector run on as long as necessary after the end. We store an explicit -offset to the name table so that if a regex is compiled on one host, saved, and -then run on another where the size of pointers is different, all might still +/* The real format of the start of the pcre block; the index of names and the +code vector run on as long as necessary after the end. We store an explicit +offset to the name table so that if a regex is compiled on one host, saved, and +then run on another where the size of pointers is different, all might still be well. - + The size of the structure must be a multiple of 8 bytes. For the case of compiled-on-4 and run-on-8, we include an extra pointer that is always NULL so that there are an even number of pointers which therefore are a multiple of 8 @@ -2320,12 +2320,12 @@ when a compiled regex is reloaded on a host with different endianness. There is also similar byte-flipping code in pcretest.c, which is used for testing the byte-flipping features. It must also be kept in step. *** WARNING *** -*/ - +*/ + typedef struct real_pcre8_or_16 { - pcre_uint32 magic_number; - pcre_uint32 size; /* Total that was malloced */ - pcre_uint32 options; /* Public options */ + pcre_uint32 magic_number; + pcre_uint32 size; /* Total that was malloced */ + pcre_uint32 options; /* Public options */ pcre_uint32 flags; /* Private flags */ pcre_uint32 limit_match; /* Limit set from regex */ pcre_uint32 limit_recursion; /* Limit set from regex */ @@ -2334,20 +2334,20 @@ typedef struct real_pcre8_or_16 { pcre_uint16 max_lookbehind; /* Longest lookbehind (characters) */ pcre_uint16 top_bracket; /* Highest numbered group */ pcre_uint16 top_backref; /* Highest numbered back reference */ - pcre_uint16 name_table_offset; /* Offset to name table that follows */ - pcre_uint16 name_entry_size; /* Size of any name items */ - pcre_uint16 name_count; /* Number of name items */ - pcre_uint16 ref_count; /* Reference count */ + pcre_uint16 name_table_offset; /* Offset to name table that follows */ + pcre_uint16 name_entry_size; /* Size of any name items */ + pcre_uint16 name_count; /* Number of name items */ + pcre_uint16 ref_count; /* Reference count */ pcre_uint16 dummy1; /* To ensure size is a multiple of 8 */ pcre_uint16 dummy2; /* To ensure size is a multiple of 8 */ pcre_uint16 dummy3; /* To ensure size is a multiple of 8 */ const pcre_uint8 *tables; /* Pointer to tables or NULL for std */ void *nullpad; /* NULL padding */ } real_pcre8_or_16; - + typedef struct real_pcre8_or_16 real_pcre; typedef struct real_pcre8_or_16 real_pcre16; - + typedef struct real_pcre32 { pcre_uint32 magic_number; pcre_uint32 size; /* Total that was malloced */ @@ -2389,16 +2389,16 @@ typedef int __assert_real_pcre_size_divisible_8[(sizeof(REAL_PCRE) % 8) == 0 ? 1 #define REAL_PCRE_OPTIONS(re) (((REAL_PCRE*)re)->options) #define REAL_PCRE_FLAGS(re) (((REAL_PCRE*)re)->flags) -/* The format of the block used to store data from pcre_study(). The same -remark (see NOTE above) about extending this structure applies. */ - -typedef struct pcre_study_data { - pcre_uint32 size; /* Total that was malloced */ +/* The format of the block used to store data from pcre_study(). The same +remark (see NOTE above) about extending this structure applies. */ + +typedef struct pcre_study_data { + pcre_uint32 size; /* Total that was malloced */ pcre_uint32 flags; /* Private flags */ pcre_uint8 start_bits[32]; /* Starting char bits */ pcre_uint32 minlength; /* Minimum subject length */ -} pcre_study_data; - +} pcre_study_data; + /* Structure for building a chain of open capturing subpatterns during compiling, so that instructions to close them can be compiled when (*ACCEPT) is encountered. This is also used to identify subpatterns that contain recursive @@ -2419,10 +2419,10 @@ typedef struct named_group { pcre_uint32 number; /* Group number */ } named_group; -/* Structure for passing "static" information around between the functions -doing the compiling, so that they are thread-safe. */ - -typedef struct compile_data { +/* Structure for passing "static" information around between the functions +doing the compiling, so that they are thread-safe. */ + +typedef struct compile_data { const pcre_uint8 *lcc; /* Points to lower casing table */ const pcre_uint8 *fcc; /* Points to case-flipping table */ const pcre_uint8 *cbits; /* Points to character type table */ @@ -2459,16 +2459,16 @@ typedef struct compile_data { int nltype; /* Newline type */ int nllen; /* Newline string length */ pcre_uchar nl[4]; /* Newline string when fixed length */ -} compile_data; - -/* Structure for maintaining a chain of pointers to the currently incomplete +} compile_data; + +/* Structure for maintaining a chain of pointers to the currently incomplete branches, for testing for left recursion while compiling. */ - -typedef struct branch_chain { - struct branch_chain *outer; + +typedef struct branch_chain { + struct branch_chain *outer; pcre_uchar *current_branch; -} branch_chain; - +} branch_chain; + /* Structure for mutual recursion detection. */ typedef struct recurse_check { @@ -2476,18 +2476,18 @@ typedef struct recurse_check { const pcre_uchar *group; } recurse_check; -/* Structure for items in a linked list that represents an explicit recursive +/* Structure for items in a linked list that represents an explicit recursive call within the pattern; used by pcre_exec(). */ - -typedef struct recursion_info { - struct recursion_info *prevrec; /* Previous recursion record (or NULL) */ + +typedef struct recursion_info { + struct recursion_info *prevrec; /* Previous recursion record (or NULL) */ unsigned int group_num; /* Number of group that was called */ int *offset_save; /* Pointer to start of saved offsets */ int saved_max; /* Number of saved offsets */ int saved_capture_last; /* Last capture number */ PCRE_PUCHAR subject_position; /* Position at start of recursion */ -} recursion_info; - +} recursion_info; + /* A similar structure for pcre_dfa_exec(). */ typedef struct dfa_recursion_info { @@ -2496,24 +2496,24 @@ typedef struct dfa_recursion_info { PCRE_PUCHAR subject_position; } dfa_recursion_info; -/* Structure for building a chain of data for holding the values of the subject -pointer at the start of each subpattern, so as to detect when an empty string +/* Structure for building a chain of data for holding the values of the subject +pointer at the start of each subpattern, so as to detect when an empty string has been matched by a subpattern - to break infinite loops; used by pcre_exec(). */ - -typedef struct eptrblock { - struct eptrblock *epb_prev; + +typedef struct eptrblock { + struct eptrblock *epb_prev; PCRE_PUCHAR epb_saved_eptr; -} eptrblock; - - -/* Structure for passing "static" information around between the functions -doing traditional NFA matching, so that they are thread-safe. */ - -typedef struct match_data { - unsigned long int match_call_count; /* As it says */ - unsigned long int match_limit; /* As it says */ - unsigned long int match_limit_recursion; /* As it says */ +} eptrblock; + + +/* Structure for passing "static" information around between the functions +doing traditional NFA matching, so that they are thread-safe. */ + +typedef struct match_data { + unsigned long int match_call_count; /* As it says */ + unsigned long int match_limit; /* As it says */ + unsigned long int match_limit_recursion; /* As it says */ int *offset_vector; /* Offset vector */ int offset_end; /* One past the end */ int offset_max; /* The maximum usable for return data */ @@ -2560,12 +2560,12 @@ typedef struct match_data { #ifdef NO_RECURSE void *match_frames_base; /* For remembering malloc'd frames */ #endif -} match_data; - -/* A similar structure is used for the same purpose by the DFA matching -functions. */ - -typedef struct dfa_match_data { +} match_data; + +/* A similar structure is used for the same purpose by the DFA matching +functions. */ + +typedef struct dfa_match_data { const pcre_uchar *start_code; /* Start of the compiled pattern */ const pcre_uchar *start_subject ; /* Start of the subject string */ const pcre_uchar *end_subject; /* End of subject string */ @@ -2579,41 +2579,41 @@ typedef struct dfa_match_data { pcre_uchar nl[4]; /* Newline string when fixed */ void *callout_data; /* To pass back to callouts */ dfa_recursion_info *recursive; /* Linked list of recursion data */ -} dfa_match_data; - -/* Bit definitions for entries in the pcre_ctypes table. */ - -#define ctype_space 0x01 -#define ctype_letter 0x02 -#define ctype_digit 0x04 -#define ctype_xdigit 0x08 -#define ctype_word 0x10 /* alphanumeric or '_' */ -#define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */ - -/* Offsets for the bitmap tables in pcre_cbits. Each table contains a set -of bits for a class map. Some classes are built by combining these tables. */ - -#define cbit_space 0 /* [:space:] or \s */ -#define cbit_xdigit 32 /* [:xdigit:] */ -#define cbit_digit 64 /* [:digit:] or \d */ -#define cbit_upper 96 /* [:upper:] */ -#define cbit_lower 128 /* [:lower:] */ -#define cbit_word 160 /* [:word:] or \w */ -#define cbit_graph 192 /* [:graph:] */ -#define cbit_print 224 /* [:print:] */ -#define cbit_punct 256 /* [:punct:] */ -#define cbit_cntrl 288 /* [:cntrl:] */ -#define cbit_length 320 /* Length of the cbits table */ - -/* Offsets of the various tables from the base tables pointer, and -total length. */ - -#define lcc_offset 0 -#define fcc_offset 256 -#define cbits_offset 512 -#define ctypes_offset (cbits_offset + cbit_length) -#define tables_length (ctypes_offset + 256) - +} dfa_match_data; + +/* Bit definitions for entries in the pcre_ctypes table. */ + +#define ctype_space 0x01 +#define ctype_letter 0x02 +#define ctype_digit 0x04 +#define ctype_xdigit 0x08 +#define ctype_word 0x10 /* alphanumeric or '_' */ +#define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */ + +/* Offsets for the bitmap tables in pcre_cbits. Each table contains a set +of bits for a class map. Some classes are built by combining these tables. */ + +#define cbit_space 0 /* [:space:] or \s */ +#define cbit_xdigit 32 /* [:xdigit:] */ +#define cbit_digit 64 /* [:digit:] or \d */ +#define cbit_upper 96 /* [:upper:] */ +#define cbit_lower 128 /* [:lower:] */ +#define cbit_word 160 /* [:word:] or \w */ +#define cbit_graph 192 /* [:graph:] */ +#define cbit_print 224 /* [:print:] */ +#define cbit_punct 256 /* [:punct:] */ +#define cbit_cntrl 288 /* [:cntrl:] */ +#define cbit_length 320 /* Length of the cbits table */ + +/* Offsets of the various tables from the base tables pointer, and +total length. */ + +#define lcc_offset 0 +#define fcc_offset 256 +#define cbits_offset 512 +#define ctypes_offset (cbits_offset + cbit_length) +#define tables_length (ctypes_offset + 256) + /* Internal function and data prefixes. */ #if defined COMPILE_PCRE8 @@ -2641,23 +2641,23 @@ total length. */ #error Unsupported compiling mode #endif /* COMPILE_PCRE[8|16|32] */ -/* Layout of the UCP type table that translates property names into types and -codes. Each entry used to point directly to a name, but to reduce the number of -relocations in shared libraries, it now has an offset into a single string -instead. */ - -typedef struct { - pcre_uint16 name_offset; - pcre_uint16 type; - pcre_uint16 value; -} ucp_type_table; - - -/* Internal shared data tables. These are tables that are used by more than one -of the exported public functions. They have to be "external" in the C sense, -but are not part of the PCRE public API. The data for these tables is in the -pcre_tables.c module. */ - +/* Layout of the UCP type table that translates property names into types and +codes. Each entry used to point directly to a name, but to reduce the number of +relocations in shared libraries, it now has an offset into a single string +instead. */ + +typedef struct { + pcre_uint16 name_offset; + pcre_uint16 type; + pcre_uint16 value; +} ucp_type_table; + + +/* Internal shared data tables. These are tables that are used by more than one +of the exported public functions. They have to be "external" in the C sense, +but are not part of the PCRE public API. The data for these tables is in the +pcre_tables.c module. */ + #ifdef COMPILE_PCRE8 extern const int PRIV(utf8_table1)[]; extern const int PRIV(utf8_table1_size); @@ -2665,25 +2665,25 @@ extern const int PRIV(utf8_table2)[]; extern const int PRIV(utf8_table3)[]; extern const pcre_uint8 PRIV(utf8_table4)[]; #endif /* COMPILE_PCRE8 */ - + extern const char PRIV(utt_names)[]; extern const ucp_type_table PRIV(utt)[]; extern const int PRIV(utt_size); - + extern const pcre_uint8 PRIV(OP_lengths)[]; extern const pcre_uint8 PRIV(default_tables)[]; - + extern const pcre_uint32 PRIV(hspace_list)[]; extern const pcre_uint32 PRIV(vspace_list)[]; - - -/* Internal shared functions. These are functions that are used by more than -one of the exported public functions. They have to be "external" in the C -sense, but are not part of the PCRE public API. */ - + + +/* Internal shared functions. These are functions that are used by more than +one of the exported public functions. They have to be "external" in the C +sense, but are not part of the PCRE public API. */ + /* String comparison functions. */ #if defined COMPILE_PCRE8 - + #define STRCMP_UC_UC(str1, str2) \ strcmp((char *)(str1), (char *)(str2)) #define STRCMP_UC_C8(str1, str2) \ @@ -2754,8 +2754,8 @@ extern int PRIV(jit_exec)(const PUBL(extra) *, extern void PRIV(jit_free)(void *); extern int PRIV(jit_get_size)(void *); extern const char* PRIV(jit_get_target)(void); -#endif - +#endif + /* Unicode character database (UCD) */ typedef struct { @@ -2804,4 +2804,4 @@ extern const int PRIV(ucp_typerange)[]; #endif -/* End of pcre_internal.h */ +/* End of pcre_internal.h */ diff --git a/contrib/libs/pcre/pcre_maketables.c b/contrib/libs/pcre/pcre_maketables.c index 873b46aa91..5328e96673 100644 --- a/contrib/libs/pcre/pcre_maketables.c +++ b/contrib/libs/pcre/pcre_maketables.c @@ -1,74 +1,74 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2012 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains the external function pcre_maketables(), which builds -character tables for PCRE in the current locale. The file is compiled on its -own as part of the PCRE library. However, it is also included in the -compilation of dftables.c, in which case the macro DFTABLES is defined. */ - - -#ifndef DFTABLES -# ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains the external function pcre_maketables(), which builds +character tables for PCRE in the current locale. The file is compiled on its +own as part of the PCRE library. However, it is also included in the +compilation of dftables.c, in which case the macro DFTABLES is defined. */ + + +#ifndef DFTABLES +# ifdef HAVE_CONFIG_H # include "pcre_config.h" -# endif -# include "pcre_internal.h" -#endif - - -/************************************************* -* Create PCRE character tables * -*************************************************/ - -/* This function builds a set of character tables for use by PCRE and returns -a pointer to them. They are build using the ctype functions, and consequently -their contents will depend upon the current locale setting. When compiled as +# endif +# include "pcre_internal.h" +#endif + + +/************************************************* +* Create PCRE character tables * +*************************************************/ + +/* This function builds a set of character tables for use by PCRE and returns +a pointer to them. They are build using the ctype functions, and consequently +their contents will depend upon the current locale setting. When compiled as part of the library, the store is obtained via PUBL(malloc)(), but when compiled inside dftables, use malloc(). - -Arguments: none -Returns: pointer to the contiguous block of data -*/ - + +Arguments: none +Returns: pointer to the contiguous block of data +*/ + #if defined COMPILE_PCRE8 -const unsigned char * -pcre_maketables(void) +const unsigned char * +pcre_maketables(void) #elif defined COMPILE_PCRE16 const unsigned char * pcre16_maketables(void) @@ -76,30 +76,30 @@ pcre16_maketables(void) const unsigned char * pcre32_maketables(void) #endif -{ -unsigned char *yield, *p; -int i; - -#ifndef DFTABLES +{ +unsigned char *yield, *p; +int i; + +#ifndef DFTABLES yield = (unsigned char*)(PUBL(malloc))(tables_length); -#else -yield = (unsigned char*)malloc(tables_length); -#endif - -if (yield == NULL) return NULL; -p = yield; - -/* First comes the lower casing table */ - -for (i = 0; i < 256; i++) *p++ = tolower(i); - -/* Next the case-flipping table */ - -for (i = 0; i < 256; i++) *p++ = islower(i)? toupper(i) : tolower(i); - -/* Then the character class tables. Don't try to be clever and save effort on +#else +yield = (unsigned char*)malloc(tables_length); +#endif + +if (yield == NULL) return NULL; +p = yield; + +/* First comes the lower casing table */ + +for (i = 0; i < 256; i++) *p++ = tolower(i); + +/* Next the case-flipping table */ + +for (i = 0; i < 256; i++) *p++ = islower(i)? toupper(i) : tolower(i); + +/* Then the character class tables. Don't try to be clever and save effort on exclusive ones - in some locales things may be different. - + Note that the table for "space" includes everything "isspace" gives, including VT in the default locale. This makes it work for the POSIX class [:space:]. From release 8.34 is is also correct for Perl space, because Perl added VT at @@ -110,47 +110,47 @@ being lower or upper, such as "male and female ordinals" (\xAA and \xBA) in the fr_FR locale (at least under Debian Linux's locales as of 12/2005). So we must test for alnum specially. */ -memset(p, 0, cbit_length); -for (i = 0; i < 256; i++) - { - if (isdigit(i)) p[cbit_digit + i/8] |= 1 << (i&7); - if (isupper(i)) p[cbit_upper + i/8] |= 1 << (i&7); - if (islower(i)) p[cbit_lower + i/8] |= 1 << (i&7); - if (isalnum(i)) p[cbit_word + i/8] |= 1 << (i&7); - if (i == '_') p[cbit_word + i/8] |= 1 << (i&7); - if (isspace(i)) p[cbit_space + i/8] |= 1 << (i&7); - if (isxdigit(i))p[cbit_xdigit + i/8] |= 1 << (i&7); - if (isgraph(i)) p[cbit_graph + i/8] |= 1 << (i&7); - if (isprint(i)) p[cbit_print + i/8] |= 1 << (i&7); - if (ispunct(i)) p[cbit_punct + i/8] |= 1 << (i&7); - if (iscntrl(i)) p[cbit_cntrl + i/8] |= 1 << (i&7); - } -p += cbit_length; - +memset(p, 0, cbit_length); +for (i = 0; i < 256; i++) + { + if (isdigit(i)) p[cbit_digit + i/8] |= 1 << (i&7); + if (isupper(i)) p[cbit_upper + i/8] |= 1 << (i&7); + if (islower(i)) p[cbit_lower + i/8] |= 1 << (i&7); + if (isalnum(i)) p[cbit_word + i/8] |= 1 << (i&7); + if (i == '_') p[cbit_word + i/8] |= 1 << (i&7); + if (isspace(i)) p[cbit_space + i/8] |= 1 << (i&7); + if (isxdigit(i))p[cbit_xdigit + i/8] |= 1 << (i&7); + if (isgraph(i)) p[cbit_graph + i/8] |= 1 << (i&7); + if (isprint(i)) p[cbit_print + i/8] |= 1 << (i&7); + if (ispunct(i)) p[cbit_punct + i/8] |= 1 << (i&7); + if (iscntrl(i)) p[cbit_cntrl + i/8] |= 1 << (i&7); + } +p += cbit_length; + /* Finally, the character type table. In this, we used to exclude VT from the white space chars, because Perl didn't recognize it as such for \s and for comments within regexes. However, Perl changed at release 5.18, so PCRE changed at release 8.34. */ - -for (i = 0; i < 256; i++) - { - int x = 0; + +for (i = 0; i < 256; i++) + { + int x = 0; if (isspace(i)) x += ctype_space; - if (isalpha(i)) x += ctype_letter; - if (isdigit(i)) x += ctype_digit; - if (isxdigit(i)) x += ctype_xdigit; - if (isalnum(i) || i == '_') x += ctype_word; - - /* Note: strchr includes the terminating zero in the characters it considers. - In this instance, that is ok because we want binary zero to be flagged as a - meta-character, which in this sense is any character that terminates a run - of data characters. */ - - if (strchr("\\*+?{^.$|()[", i) != 0) x += ctype_meta; - *p++ = x; - } - -return yield; -} - -/* End of pcre_maketables.c */ + if (isalpha(i)) x += ctype_letter; + if (isdigit(i)) x += ctype_digit; + if (isxdigit(i)) x += ctype_xdigit; + if (isalnum(i) || i == '_') x += ctype_word; + + /* Note: strchr includes the terminating zero in the characters it considers. + In this instance, that is ok because we want binary zero to be flagged as a + meta-character, which in this sense is any character that terminates a run + of data characters. */ + + if (strchr("\\*+?{^.$|()[", i) != 0) x += ctype_meta; + *p++ = x; + } + +return yield; +} + +/* End of pcre_maketables.c */ diff --git a/contrib/libs/pcre/pcre_newline.c b/contrib/libs/pcre/pcre_newline.c index 252cad9c9e..c8b5e374ae 100644 --- a/contrib/libs/pcre/pcre_newline.c +++ b/contrib/libs/pcre/pcre_newline.c @@ -1,81 +1,81 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2012 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains internal functions for testing newlines when more than -one kind of newline is to be recognized. When a newline is found, its length is -returned. In principle, we could implement several newline "types", each -referring to a different set of newline characters. At present, PCRE supports -only NLTYPE_FIXED, which gets handled without these functions, NLTYPE_ANYCRLF, -and NLTYPE_ANY. The full list of Unicode newline characters is taken from -http://unicode.org/unicode/reports/tr18/. */ - - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains internal functions for testing newlines when more than +one kind of newline is to be recognized. When a newline is found, its length is +returned. In principle, we could implement several newline "types", each +referring to a different set of newline characters. At present, PCRE supports +only NLTYPE_FIXED, which gets handled without these functions, NLTYPE_ANYCRLF, +and NLTYPE_ANY. The full list of Unicode newline characters is taken from +http://unicode.org/unicode/reports/tr18/. */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - - - -/************************************************* -* Check for newline at given position * -*************************************************/ - -/* It is guaranteed that the initial value of ptr is less than the end of the -string that is being processed. - -Arguments: - ptr pointer to possible newline - type the newline type - endptr pointer to the end of the string - lenptr where to return the length +#endif + +#include "pcre_internal.h" + + + +/************************************************* +* Check for newline at given position * +*************************************************/ + +/* It is guaranteed that the initial value of ptr is less than the end of the +string that is being processed. + +Arguments: + ptr pointer to possible newline + type the newline type + endptr pointer to the end of the string + lenptr where to return the length utf TRUE if in utf mode - -Returns: TRUE or FALSE -*/ - -BOOL + +Returns: TRUE or FALSE +*/ + +BOOL PRIV(is_newline)(PCRE_PUCHAR ptr, int type, PCRE_PUCHAR endptr, int *lenptr, BOOL utf) -{ +{ pcre_uint32 c; (void)utf; #ifdef SUPPORT_UTF @@ -86,21 +86,21 @@ if (utf) else #endif /* SUPPORT_UTF */ c = *ptr; - + /* Note that this function is called only for ANY or ANYCRLF. */ -if (type == NLTYPE_ANYCRLF) switch(c) - { +if (type == NLTYPE_ANYCRLF) switch(c) + { case CHAR_LF: *lenptr = 1; return TRUE; case CHAR_CR: *lenptr = (ptr < endptr - 1 && ptr[1] == CHAR_LF)? 2 : 1; return TRUE; - default: return FALSE; - } - -/* NLTYPE_ANY */ - -else switch(c) - { + default: return FALSE; + } + +/* NLTYPE_ANY */ + +else switch(c) + { #ifdef EBCDIC case CHAR_NEL: #endif @@ -115,8 +115,8 @@ else switch(c) #ifndef EBCDIC #ifdef COMPILE_PCRE8 case CHAR_NEL: *lenptr = utf? 2 : 1; return TRUE; - case 0x2028: /* LS */ - case 0x2029: *lenptr = 3; return TRUE; /* PS */ + case 0x2028: /* LS */ + case 0x2029: *lenptr = 3; return TRUE; /* PS */ #else /* COMPILE_PCRE16 || COMPILE_PCRE32 */ case CHAR_NEL: case 0x2028: /* LS */ @@ -124,62 +124,62 @@ else switch(c) #endif /* COMPILE_PCRE8 */ #endif /* Not EBCDIC */ - default: return FALSE; - } -} - - - -/************************************************* -* Check for newline at previous position * -*************************************************/ - -/* It is guaranteed that the initial value of ptr is greater than the start of -the string that is being processed. - -Arguments: - ptr pointer to possible newline - type the newline type - startptr pointer to the start of the string - lenptr where to return the length + default: return FALSE; + } +} + + + +/************************************************* +* Check for newline at previous position * +*************************************************/ + +/* It is guaranteed that the initial value of ptr is greater than the start of +the string that is being processed. + +Arguments: + ptr pointer to possible newline + type the newline type + startptr pointer to the start of the string + lenptr where to return the length utf TRUE if in utf mode - -Returns: TRUE or FALSE -*/ - -BOOL + +Returns: TRUE or FALSE +*/ + +BOOL PRIV(was_newline)(PCRE_PUCHAR ptr, int type, PCRE_PUCHAR startptr, int *lenptr, BOOL utf) -{ +{ pcre_uint32 c; (void)utf; -ptr--; +ptr--; #ifdef SUPPORT_UTF if (utf) - { - BACKCHAR(ptr); - GETCHAR(c, ptr); - } + { + BACKCHAR(ptr); + GETCHAR(c, ptr); + } else #endif /* SUPPORT_UTF */ c = *ptr; - + /* Note that this function is called only for ANY or ANYCRLF. */ -if (type == NLTYPE_ANYCRLF) switch(c) - { +if (type == NLTYPE_ANYCRLF) switch(c) + { case CHAR_LF: *lenptr = (ptr > startptr && ptr[-1] == CHAR_CR)? 2 : 1; return TRUE; case CHAR_CR: *lenptr = 1; return TRUE; - default: return FALSE; - } - + default: return FALSE; + } + /* NLTYPE_ANY */ -else switch(c) - { +else switch(c) + { case CHAR_LF: *lenptr = (ptr > startptr && ptr[-1] == CHAR_CR)? 2 : 1; return TRUE; @@ -203,8 +203,8 @@ else switch(c) #endif /* COMPILE_PCRE8 */ #endif /* NotEBCDIC */ - default: return FALSE; - } -} - -/* End of pcre_newline.c */ + default: return FALSE; + } +} + +/* End of pcre_newline.c */ diff --git a/contrib/libs/pcre/pcre_ord2utf8.c b/contrib/libs/pcre/pcre_ord2utf8.c index e608a29a30..827a9fa22a 100644 --- a/contrib/libs/pcre/pcre_ord2utf8.c +++ b/contrib/libs/pcre/pcre_ord2utf8.c @@ -1,94 +1,94 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2012 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This file contains a private PCRE function that converts an ordinal -character value into a UTF8 string. */ - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This file contains a private PCRE function that converts an ordinal +character value into a UTF8 string. */ + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - +#endif + #define COMPILE_PCRE8 -#include "pcre_internal.h" - -/************************************************* -* Convert character value to UTF-8 * -*************************************************/ - +#include "pcre_internal.h" + +/************************************************* +* Convert character value to UTF-8 * +*************************************************/ + /* This function takes an integer value in the range 0 - 0x10ffff and encodes it as a UTF-8 character in 1 to 4 pcre_uchars. - -Arguments: - cvalue the character value + +Arguments: + cvalue the character value buffer pointer to buffer for result - at least 6 pcre_uchars long - -Returns: number of characters placed in the buffer -*/ - + +Returns: number of characters placed in the buffer +*/ + unsigned -int +int PRIV(ord2utf)(pcre_uint32 cvalue, pcre_uchar *buffer) -{ +{ #ifdef SUPPORT_UTF -register int i, j; +register int i, j; for (i = 0; i < PRIV(utf8_table1_size); i++) if ((int)cvalue <= PRIV(utf8_table1)[i]) break; -buffer += i; -for (j = i; j > 0; j--) - { - *buffer-- = 0x80 | (cvalue & 0x3f); - cvalue >>= 6; - } +buffer += i; +for (j = i; j > 0; j--) + { + *buffer-- = 0x80 | (cvalue & 0x3f); + cvalue >>= 6; + } *buffer = PRIV(utf8_table2)[i] | cvalue; -return i + 1; +return i + 1; -#else +#else (void)(cvalue); /* Keep compiler happy; this function won't ever be */ (void)(buffer); /* called when SUPPORT_UTF is not defined. */ return 0; #endif -} - -/* End of pcre_ord2utf8.c */ +} + +/* End of pcre_ord2utf8.c */ diff --git a/contrib/libs/pcre/pcre_refcount.c b/contrib/libs/pcre/pcre_refcount.c index 65a3c23a8f..5de3422a14 100644 --- a/contrib/libs/pcre/pcre_refcount.c +++ b/contrib/libs/pcre/pcre_refcount.c @@ -1,76 +1,76 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2012 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains the external function pcre_refcount(), which is an -auxiliary function that can be used to maintain a reference count in a compiled -pattern data block. This might be helpful in applications where the block is -shared by different users. */ - - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains the external function pcre_refcount(), which is an +auxiliary function that can be used to maintain a reference count in a compiled +pattern data block. This might be helpful in applications where the block is +shared by different users. */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - - -/************************************************* -* Maintain reference count * -*************************************************/ - -/* The reference count is a 16-bit field, initialized to zero. It is not -possible to transfer a non-zero count from one host to a different host that -has a different byte order - though I can't see why anyone in their right mind -would ever want to do that! - -Arguments: - argument_re points to compiled code - adjust value to add to the count - -Returns: the (possibly updated) count value (a non-negative number), or - a negative error number -*/ - +#endif + +#include "pcre_internal.h" + + +/************************************************* +* Maintain reference count * +*************************************************/ + +/* The reference count is a 16-bit field, initialized to zero. It is not +possible to transfer a non-zero count from one host to a different host that +has a different byte order - though I can't see why anyone in their right mind +would ever want to do that! + +Arguments: + argument_re points to compiled code + adjust value to add to the count + +Returns: the (possibly updated) count value (a non-negative number), or + a negative error number +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION -pcre_refcount(pcre *argument_re, int adjust) +pcre_refcount(pcre *argument_re, int adjust) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre16_refcount(pcre16 *argument_re, int adjust) @@ -78,15 +78,15 @@ pcre16_refcount(pcre16 *argument_re, int adjust) PCRE_EXP_DEFN int PCRE_CALL_CONVENTION pcre32_refcount(pcre32 *argument_re, int adjust) #endif -{ +{ REAL_PCRE *re = (REAL_PCRE *)argument_re; -if (re == NULL) return PCRE_ERROR_NULL; +if (re == NULL) return PCRE_ERROR_NULL; if (re->magic_number != MAGIC_NUMBER) return PCRE_ERROR_BADMAGIC; if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE; -re->ref_count = (-adjust > re->ref_count)? 0 : - (adjust + re->ref_count > 65535)? 65535 : - re->ref_count + adjust; -return re->ref_count; -} - -/* End of pcre_refcount.c */ +re->ref_count = (-adjust > re->ref_count)? 0 : + (adjust + re->ref_count > 65535)? 65535 : + re->ref_count + adjust; +return re->ref_count; +} + +/* End of pcre_refcount.c */ diff --git a/contrib/libs/pcre/pcre_study.c b/contrib/libs/pcre/pcre_study.c index b6088fe882..d4ee0295d4 100644 --- a/contrib/libs/pcre/pcre_study.c +++ b/contrib/libs/pcre/pcre_study.c @@ -1,62 +1,62 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2012 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains the external function pcre_study(), along with local -supporting functions. */ - - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains the external function pcre_study(), along with local +supporting functions. */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - +#endif + +#include "pcre_internal.h" + #define SET_BIT(c) start_bits[c/8] |= (1 << (c&7)) - -/* Returns from set_start_bits() */ - + +/* Returns from set_start_bits() */ + enum { SSB_FAIL, SSB_DONE, SSB_CONTINUE, SSB_UNKNOWN }; + + - - -/************************************************* +/************************************************* * Find the minimum subject length for a group * *************************************************/ @@ -613,24 +613,24 @@ for (;;) /************************************************* -* Set a bit and maybe its alternate case * -*************************************************/ - +* Set a bit and maybe its alternate case * +*************************************************/ + /* Given a character, set its first byte's bit in the table, and also the corresponding bit for the other version of a letter if we are caseless. In UTF-8 mode, for characters greater than 127, we can only do the caseless thing when Unicode property support is available. - -Arguments: - start_bits points to the bit map + +Arguments: + start_bits points to the bit map p points to the character - caseless the caseless flag - cd the block with char table pointers + caseless the caseless flag + cd the block with char table pointers utf TRUE for UTF-8 / UTF-16 / UTF-32 mode - + Returns: pointer after the character -*/ - +*/ + static const pcre_uchar * set_table_bit(pcre_uint8 *start_bits, const pcre_uchar *p, BOOL caseless, compile_data *cd, BOOL utf) @@ -719,10 +719,10 @@ Arguments: Returns: nothing */ -static void +static void set_type_bits(pcre_uint8 *start_bits, int cbit_type, unsigned int table_limit, compile_data *cd) -{ +{ register pcre_uint32 c; for (c = 0; c < table_limit; c++) start_bits[c] |= cd->cbits[c+cbit_type]; #if defined SUPPORT_UTF && defined COMPILE_PCRE8 @@ -737,13 +737,13 @@ for (c = 128; c < 256; c++) } } #endif -} - - +} + + /************************************************* * Set bits for a negative character type * *************************************************/ - + /* This function sets starting bits for a negative character type such as \D. In UTF-8 mode, we can only do a direct setting for bytes less than 128, as otherwise there can be confusion with bytes in the middle of UTF-8 characters. @@ -774,78 +774,78 @@ if (table_limit != 32) for (c = 24; c < 32; c++) start_bits[c] = 0xff; -/************************************************* -* Create bitmap of starting bytes * -*************************************************/ - -/* This function scans a compiled unanchored expression recursively and -attempts to build a bitmap of the set of possible starting bytes. As time goes -by, we may be able to get more clever at doing this. The SSB_CONTINUE return is -useful for parenthesized groups in patterns such as (a*)b where the group -provides some optional starting bytes but scanning must continue at the outer -level to find at least one mandatory byte. At the outermost level, this -function fails unless the result is SSB_DONE. - -Arguments: - code points to an expression - start_bits points to a 32-byte table, initialized to 0 +/************************************************* +* Create bitmap of starting bytes * +*************************************************/ + +/* This function scans a compiled unanchored expression recursively and +attempts to build a bitmap of the set of possible starting bytes. As time goes +by, we may be able to get more clever at doing this. The SSB_CONTINUE return is +useful for parenthesized groups in patterns such as (a*)b where the group +provides some optional starting bytes but scanning must continue at the outer +level to find at least one mandatory byte. At the outermost level, this +function fails unless the result is SSB_DONE. + +Arguments: + code points to an expression + start_bits points to a 32-byte table, initialized to 0 utf TRUE if in UTF-8 / UTF-16 / UTF-32 mode - cd the block with char table pointers - -Returns: SSB_FAIL => Failed to find any starting bytes - SSB_DONE => Found mandatory starting bytes - SSB_CONTINUE => Found optional starting bytes + cd the block with char table pointers + +Returns: SSB_FAIL => Failed to find any starting bytes + SSB_DONE => Found mandatory starting bytes + SSB_CONTINUE => Found optional starting bytes SSB_UNKNOWN => Hit an unrecognized opcode -*/ - -static int +*/ + +static int set_start_bits(const pcre_uchar *code, pcre_uint8 *start_bits, BOOL utf, compile_data *cd) -{ +{ register pcre_uint32 c; -int yield = SSB_DONE; +int yield = SSB_DONE; #if defined SUPPORT_UTF && defined COMPILE_PCRE8 int table_limit = utf? 16:32; #else int table_limit = 32; #endif - -#if 0 -/* ========================================================================= */ -/* The following comment and code was inserted in January 1999. In May 2006, -when it was observed to cause compiler warnings about unused values, I took it -out again. If anybody is still using OS/2, they will have to put it back -manually. */ - -/* This next statement and the later reference to dummy are here in order to -trick the optimizer of the IBM C compiler for OS/2 into generating correct -code. Apparently IBM isn't going to fix the problem, and we would rather not -disable optimization (in this module it actually makes a big difference, and -the pcre module can use all the optimization it can get). */ - -volatile int dummy; -/* ========================================================================= */ -#endif - -do - { - BOOL try_next = TRUE; + +#if 0 +/* ========================================================================= */ +/* The following comment and code was inserted in January 1999. In May 2006, +when it was observed to cause compiler warnings about unused values, I took it +out again. If anybody is still using OS/2, they will have to put it back +manually. */ + +/* This next statement and the later reference to dummy are here in order to +trick the optimizer of the IBM C compiler for OS/2 into generating correct +code. Apparently IBM isn't going to fix the problem, and we would rather not +disable optimization (in this module it actually makes a big difference, and +the pcre module can use all the optimization it can get). */ + +volatile int dummy; +/* ========================================================================= */ +#endif + +do + { + BOOL try_next = TRUE; const pcre_uchar *tcode = code + 1 + LINK_SIZE; - + if (*code == OP_CBRA || *code == OP_SCBRA || *code == OP_CBRAPOS || *code == OP_SCBRAPOS) tcode += IMM2_SIZE; - while (try_next) /* Loop for items in this branch */ - { - int rc; + while (try_next) /* Loop for items in this branch */ + { + int rc; - switch(*tcode) - { + switch(*tcode) + { /* If we reach something we don't understand, it means a new opcode has been created that hasn't been added to this code. Hopefully this problem will be discovered during testing. */ - - default: + + default: return SSB_UNKNOWN; /* Fail for a valid opcode that implies no starting bits. */ @@ -920,8 +920,8 @@ do case OP_SOM: case OP_THEN: case OP_THEN_ARG: - return SSB_FAIL; - + return SSB_FAIL; + /* A "real" property test implies no starting bits, but the fake property PT_CLIST identifies a list of characters. These lists are short, as they are used for characters with more than one "other case", so there is no @@ -954,80 +954,80 @@ do tcode++; break; - /* If we hit a bracket or a positive lookahead assertion, recurse to set - bits from within the subpattern. If it can't find anything, we have to - give up. If it finds some mandatory character(s), we are done for this - branch. Otherwise, carry on scanning after the subpattern. */ - - case OP_BRA: - case OP_SBRA: - case OP_CBRA: - case OP_SCBRA: + /* If we hit a bracket or a positive lookahead assertion, recurse to set + bits from within the subpattern. If it can't find anything, we have to + give up. If it finds some mandatory character(s), we are done for this + branch. Otherwise, carry on scanning after the subpattern. */ + + case OP_BRA: + case OP_SBRA: + case OP_CBRA: + case OP_SCBRA: case OP_BRAPOS: case OP_SBRAPOS: case OP_CBRAPOS: case OP_SCBRAPOS: - case OP_ONCE: + case OP_ONCE: case OP_ONCE_NC: - case OP_ASSERT: + case OP_ASSERT: rc = set_start_bits(tcode, start_bits, utf, cd); if (rc == SSB_FAIL || rc == SSB_UNKNOWN) return rc; - if (rc == SSB_DONE) try_next = FALSE; else - { - do tcode += GET(tcode, 1); while (*tcode == OP_ALT); - tcode += 1 + LINK_SIZE; - } - break; - - /* If we hit ALT or KET, it means we haven't found anything mandatory in - this branch, though we might have found something optional. For ALT, we - continue with the next alternative, but we have to arrange that the final - result from subpattern is SSB_CONTINUE rather than SSB_DONE. For KET, - return SSB_CONTINUE: if this is the top level, that indicates failure, - but after a nested subpattern, it causes scanning to continue. */ - - case OP_ALT: - yield = SSB_CONTINUE; - try_next = FALSE; - break; - - case OP_KET: - case OP_KETRMAX: - case OP_KETRMIN: + if (rc == SSB_DONE) try_next = FALSE; else + { + do tcode += GET(tcode, 1); while (*tcode == OP_ALT); + tcode += 1 + LINK_SIZE; + } + break; + + /* If we hit ALT or KET, it means we haven't found anything mandatory in + this branch, though we might have found something optional. For ALT, we + continue with the next alternative, but we have to arrange that the final + result from subpattern is SSB_CONTINUE rather than SSB_DONE. For KET, + return SSB_CONTINUE: if this is the top level, that indicates failure, + but after a nested subpattern, it causes scanning to continue. */ + + case OP_ALT: + yield = SSB_CONTINUE; + try_next = FALSE; + break; + + case OP_KET: + case OP_KETRMAX: + case OP_KETRMIN: case OP_KETRPOS: - return SSB_CONTINUE; - - /* Skip over callout */ - - case OP_CALLOUT: - tcode += 2 + 2*LINK_SIZE; - break; - - /* Skip over lookbehind and negative lookahead assertions */ - - case OP_ASSERT_NOT: - case OP_ASSERTBACK: - case OP_ASSERTBACK_NOT: - do tcode += GET(tcode, 1); while (*tcode == OP_ALT); - tcode += 1 + LINK_SIZE; - break; - - /* BRAZERO does the bracket, but carries on. */ - - case OP_BRAZERO: - case OP_BRAMINZERO: + return SSB_CONTINUE; + + /* Skip over callout */ + + case OP_CALLOUT: + tcode += 2 + 2*LINK_SIZE; + break; + + /* Skip over lookbehind and negative lookahead assertions */ + + case OP_ASSERT_NOT: + case OP_ASSERTBACK: + case OP_ASSERTBACK_NOT: + do tcode += GET(tcode, 1); while (*tcode == OP_ALT); + tcode += 1 + LINK_SIZE; + break; + + /* BRAZERO does the bracket, but carries on. */ + + case OP_BRAZERO: + case OP_BRAMINZERO: case OP_BRAPOSZERO: rc = set_start_bits(++tcode, start_bits, utf, cd); if (rc == SSB_FAIL || rc == SSB_UNKNOWN) return rc; -/* ========================================================================= - See the comment at the head of this function concerning the next line, - which was an old fudge for the benefit of OS/2. - dummy = 1; - ========================================================================= */ - do tcode += GET(tcode,1); while (*tcode == OP_ALT); - tcode += 1 + LINK_SIZE; - break; - +/* ========================================================================= + See the comment at the head of this function concerning the next line, + which was an old fudge for the benefit of OS/2. + dummy = 1; + ========================================================================= */ + do tcode += GET(tcode,1); while (*tcode == OP_ALT); + tcode += 1 + LINK_SIZE; + break; + /* SKIPZERO skips the bracket. */ case OP_SKIPZERO: @@ -1036,17 +1036,17 @@ do tcode += 1 + LINK_SIZE; break; - /* Single-char * or ? sets the bit and tries the next item */ - - case OP_STAR: - case OP_MINSTAR: - case OP_POSSTAR: - case OP_QUERY: - case OP_MINQUERY: - case OP_POSQUERY: + /* Single-char * or ? sets the bit and tries the next item */ + + case OP_STAR: + case OP_MINSTAR: + case OP_POSSTAR: + case OP_QUERY: + case OP_MINQUERY: + case OP_POSQUERY: tcode = set_table_bit(start_bits, tcode + 1, FALSE, cd, utf); - break; - + break; + case OP_STARI: case OP_MINSTARI: case OP_POSSTARI: @@ -1056,33 +1056,33 @@ do tcode = set_table_bit(start_bits, tcode + 1, TRUE, cd, utf); break; - /* Single-char upto sets the bit and tries the next */ - - case OP_UPTO: - case OP_MINUPTO: - case OP_POSUPTO: + /* Single-char upto sets the bit and tries the next */ + + case OP_UPTO: + case OP_MINUPTO: + case OP_POSUPTO: tcode = set_table_bit(start_bits, tcode + 1 + IMM2_SIZE, FALSE, cd, utf); - break; - + break; + case OP_UPTOI: case OP_MINUPTOI: case OP_POSUPTOI: tcode = set_table_bit(start_bits, tcode + 1 + IMM2_SIZE, TRUE, cd, utf); break; - /* At least one single char sets the bit and stops */ - + /* At least one single char sets the bit and stops */ + case OP_EXACT: tcode += IMM2_SIZE; /* Fall through */ - case OP_CHAR: - case OP_PLUS: - case OP_MINPLUS: - case OP_POSPLUS: + case OP_CHAR: + case OP_PLUS: + case OP_MINPLUS: + case OP_POSPLUS: (void)set_table_bit(start_bits, tcode + 1, FALSE, cd, utf); - try_next = FALSE; - break; - + try_next = FALSE; + break; + case OP_EXACTI: tcode += IMM2_SIZE; /* Fall through */ @@ -1093,7 +1093,7 @@ do (void)set_table_bit(start_bits, tcode + 1, TRUE, cd, utf); try_next = FALSE; break; - + /* Special spacing and line-terminating items. These recognize specific lists of characters. The difference between VSPACE and ANYNL is that the latter can match the two-character CRLF sequence, but that is not @@ -1162,74 +1162,74 @@ do properties. Therefore, these apply in the case when only characters less than 256 are recognized to match the types. */ - case OP_NOT_DIGIT: + case OP_NOT_DIGIT: set_nottype_bits(start_bits, cbit_digit, table_limit, cd); - try_next = FALSE; - break; - - case OP_DIGIT: + try_next = FALSE; + break; + + case OP_DIGIT: set_type_bits(start_bits, cbit_digit, table_limit, cd); - try_next = FALSE; - break; - + try_next = FALSE; + break; + /* The cbit_space table has vertical tab as whitespace; we no longer have to play fancy tricks because Perl added VT to its whitespace at release 5.18. PCRE added it at release 8.34. */ - - case OP_NOT_WHITESPACE: + + case OP_NOT_WHITESPACE: set_nottype_bits(start_bits, cbit_space, table_limit, cd); - try_next = FALSE; - break; - - case OP_WHITESPACE: + try_next = FALSE; + break; + + case OP_WHITESPACE: set_type_bits(start_bits, cbit_space, table_limit, cd); - try_next = FALSE; - break; - - case OP_NOT_WORDCHAR: + try_next = FALSE; + break; + + case OP_NOT_WORDCHAR: set_nottype_bits(start_bits, cbit_word, table_limit, cd); - try_next = FALSE; - break; - - case OP_WORDCHAR: + try_next = FALSE; + break; + + case OP_WORDCHAR: set_type_bits(start_bits, cbit_word, table_limit, cd); - try_next = FALSE; - break; - - /* One or more character type fudges the pointer and restarts, knowing - it will hit a single character type and stop there. */ - - case OP_TYPEPLUS: - case OP_TYPEMINPLUS: + try_next = FALSE; + break; + + /* One or more character type fudges the pointer and restarts, knowing + it will hit a single character type and stop there. */ + + case OP_TYPEPLUS: + case OP_TYPEMINPLUS: case OP_TYPEPOSPLUS: - tcode++; - break; - - case OP_TYPEEXACT: + tcode++; + break; + + case OP_TYPEEXACT: tcode += 1 + IMM2_SIZE; - break; - - /* Zero or more repeats of character types set the bits and then - try again. */ - - case OP_TYPEUPTO: - case OP_TYPEMINUPTO: - case OP_TYPEPOSUPTO: + break; + + /* Zero or more repeats of character types set the bits and then + try again. */ + + case OP_TYPEUPTO: + case OP_TYPEMINUPTO: + case OP_TYPEPOSUPTO: tcode += IMM2_SIZE; /* Fall through */ - - case OP_TYPESTAR: - case OP_TYPEMINSTAR: - case OP_TYPEPOSSTAR: - case OP_TYPEQUERY: - case OP_TYPEMINQUERY: - case OP_TYPEPOSQUERY: - switch(tcode[1]) - { + + case OP_TYPESTAR: + case OP_TYPEMINSTAR: + case OP_TYPEPOSSTAR: + case OP_TYPEQUERY: + case OP_TYPEMINQUERY: + case OP_TYPEPOSQUERY: + switch(tcode[1]) + { default: - case OP_ANY: + case OP_ANY: case OP_ALLANY: - return SSB_FAIL; - + return SSB_FAIL; + case OP_HSPACE: SET_BIT(CHAR_HT); SET_BIT(CHAR_SPACE); @@ -1275,44 +1275,44 @@ do SET_BIT(CHAR_NEL); break; - case OP_NOT_DIGIT: + case OP_NOT_DIGIT: set_nottype_bits(start_bits, cbit_digit, table_limit, cd); - break; - - case OP_DIGIT: + break; + + case OP_DIGIT: set_type_bits(start_bits, cbit_digit, table_limit, cd); - break; - + break; + /* The cbit_space table has vertical tab as whitespace; we no longer have to play fancy tricks because Perl added VT to its whitespace at release 5.18. PCRE added it at release 8.34. */ - - case OP_NOT_WHITESPACE: + + case OP_NOT_WHITESPACE: set_nottype_bits(start_bits, cbit_space, table_limit, cd); - break; - - case OP_WHITESPACE: + break; + + case OP_WHITESPACE: set_type_bits(start_bits, cbit_space, table_limit, cd); - break; - - case OP_NOT_WORDCHAR: + break; + + case OP_NOT_WORDCHAR: set_nottype_bits(start_bits, cbit_word, table_limit, cd); - break; - - case OP_WORDCHAR: + break; + + case OP_WORDCHAR: set_type_bits(start_bits, cbit_word, table_limit, cd); - break; - } - - tcode += 2; - break; - - /* Character class where all the information is in a bit map: set the - bits and either carry on or not, according to the repeat count. If it was - a negative class, and we are operating with UTF-8 characters, any byte - with a value >= 0xc4 is a potentially valid starter because it starts a - character with a value > 255. */ - + break; + } + + tcode += 2; + break; + + /* Character class where all the information is in a bit map: set the + bits and either carry on or not, according to the repeat count. If it was + a negative class, and we are operating with UTF-8 characters, any byte + with a value >= 0xc4 is a potentially valid starter because it starts a + character with a value > 255. */ + #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 case OP_XCLASS: if ((tcode[1 + LINK_SIZE] & XCL_HASPROP) != 0) @@ -1323,21 +1323,21 @@ do #endif /* Fall through */ - case OP_NCLASS: + case OP_NCLASS: #if defined SUPPORT_UTF && defined COMPILE_PCRE8 if (utf) - { - start_bits[24] |= 0xf0; /* Bits for 0xc4 - 0xc8 */ - memset(start_bits+25, 0xff, 7); /* Bits for 0xc9 - 0xff */ - } -#endif + { + start_bits[24] |= 0xf0; /* Bits for 0xc4 - 0xc8 */ + memset(start_bits+25, 0xff, 7); /* Bits for 0xc9 - 0xff */ + } +#endif #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32 SET_BIT(0xFF); /* For characters > 255 */ #endif - /* Fall through */ - - case OP_CLASS: - { + /* Fall through */ + + case OP_CLASS: + { pcre_uint8 *map; #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 map = NULL; @@ -1354,102 +1354,102 @@ do map = (pcre_uint8 *)tcode; tcode += 32 / sizeof(pcre_uchar); } - - /* In UTF-8 mode, the bits in a bit map correspond to character - values, not to byte values. However, the bit map we are constructing is - for byte values. So we have to do a conversion for characters whose - value is > 127. In fact, there are only two possible starting bytes for - characters in the range 128 - 255. */ - + + /* In UTF-8 mode, the bits in a bit map correspond to character + values, not to byte values. However, the bit map we are constructing is + for byte values. So we have to do a conversion for characters whose + value is > 127. In fact, there are only two possible starting bytes for + characters in the range 128 - 255. */ + #if defined SUPPORT_UTF || !defined COMPILE_PCRE8 if (map != NULL) #endif - { + { #if defined SUPPORT_UTF && defined COMPILE_PCRE8 if (utf) - { + { for (c = 0; c < 16; c++) start_bits[c] |= map[c]; for (c = 128; c < 256; c++) - { + { if ((map[c/8] & (1 << (c&7))) != 0) { int d = (c >> 6) | 0xc0; /* Set bit for this starter */ start_bits[d/8] |= (1 << (d&7)); /* and then skip on to the */ c = (c & 0xc0) + 0x40 - 1; /* next relevant character. */ } - } - } + } + } else -#endif +#endif { /* In non-UTF-8 mode, the two bit maps are completely compatible. */ for (c = 0; c < 32; c++) start_bits[c] |= map[c]; } - } - + } + /* Advance past the bit map, and act on what follows. For a zero minimum repeat, continue; otherwise stop processing. */ - - switch (*tcode) - { - case OP_CRSTAR: - case OP_CRMINSTAR: - case OP_CRQUERY: - case OP_CRMINQUERY: + + switch (*tcode) + { + case OP_CRSTAR: + case OP_CRMINSTAR: + case OP_CRQUERY: + case OP_CRMINQUERY: case OP_CRPOSSTAR: case OP_CRPOSQUERY: - tcode++; - break; - - case OP_CRRANGE: - case OP_CRMINRANGE: + tcode++; + break; + + case OP_CRRANGE: + case OP_CRMINRANGE: case OP_CRPOSRANGE: if (GET2(tcode, 1) == 0) tcode += 1 + 2 * IMM2_SIZE; - else try_next = FALSE; - break; - - default: - try_next = FALSE; - break; - } - } - break; /* End of bitmap class handling */ - - } /* End of switch */ - } /* End of try_next loop */ - - code += GET(code, 1); /* Advance to next branch */ - } -while (*code == OP_ALT); -return yield; -} - - - - - -/************************************************* -* Study a compiled expression * -*************************************************/ - -/* This function is handed a compiled expression that it must study to produce + else try_next = FALSE; + break; + + default: + try_next = FALSE; + break; + } + } + break; /* End of bitmap class handling */ + + } /* End of switch */ + } /* End of try_next loop */ + + code += GET(code, 1); /* Advance to next branch */ + } +while (*code == OP_ALT); +return yield; +} + + + + + +/************************************************* +* Study a compiled expression * +*************************************************/ + +/* This function is handed a compiled expression that it must study to produce information that will speed up the matching. It returns a pcre[16]_extra block -which then gets handed back to pcre_exec(). - -Arguments: - re points to the compiled expression - options contains option bits - errorptr points to where to place error messages; - set NULL unless error - +which then gets handed back to pcre_exec(). + +Arguments: + re points to the compiled expression + options contains option bits + errorptr points to where to place error messages; + set NULL unless error + Returns: pointer to a pcre[16]_extra block, with study_data filled in and the appropriate flags set; - NULL on error or if no optimization possible -*/ - + NULL on error or if no optimization possible +*/ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN pcre_extra * PCRE_CALL_CONVENTION -pcre_study(const pcre *external_re, int options, const char **errorptr) +pcre_study(const pcre *external_re, int options, const char **errorptr) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN pcre16_extra * PCRE_CALL_CONVENTION pcre16_study(const pcre16 *external_re, int options, const char **errorptr) @@ -1457,27 +1457,27 @@ pcre16_study(const pcre16 *external_re, int options, const char **errorptr) PCRE_EXP_DEFN pcre32_extra * PCRE_CALL_CONVENTION pcre32_study(const pcre32 *external_re, int options, const char **errorptr) #endif -{ +{ int min; int count = 0; BOOL bits_set = FALSE; pcre_uint8 start_bits[32]; PUBL(extra) *extra = NULL; -pcre_study_data *study; +pcre_study_data *study; const pcre_uint8 *tables; pcre_uchar *code; -compile_data compile_block; +compile_data compile_block; const REAL_PCRE *re = (const REAL_PCRE *)external_re; - - -*errorptr = NULL; - -if (re == NULL || re->magic_number != MAGIC_NUMBER) - { - *errorptr = "argument is not a compiled regular expression"; - return NULL; - } - + + +*errorptr = NULL; + +if (re == NULL || re->magic_number != MAGIC_NUMBER) + { + *errorptr = "argument is not a compiled regular expression"; + return NULL; + } + if ((re->flags & PCRE_MODE) == 0) { #if defined COMPILE_PCRE8 @@ -1490,28 +1490,28 @@ if ((re->flags & PCRE_MODE) == 0) return NULL; } -if ((options & ~PUBLIC_STUDY_OPTIONS) != 0) - { - *errorptr = "unknown or incorrect option bit(s) set"; - return NULL; - } - +if ((options & ~PUBLIC_STUDY_OPTIONS) != 0) + { + *errorptr = "unknown or incorrect option bit(s) set"; + return NULL; + } + code = (pcre_uchar *)re + re->name_table_offset + - (re->name_count * re->name_entry_size); - -/* For an anchored pattern, or an unanchored pattern that has a first char, or + (re->name_count * re->name_entry_size); + +/* For an anchored pattern, or an unanchored pattern that has a first char, or a multiline pattern that matches only at "line starts", there is no point in seeking a list of starting bytes. */ - + if ((re->options & PCRE_ANCHORED) == 0 && (re->flags & (PCRE_FIRSTSET|PCRE_STARTLINE)) == 0) { int rc; - + /* Set the character tables in the block that is passed around */ - + tables = re->tables; - + #if defined COMPILE_PCRE8 if (tables == NULL) (void)pcre_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES, @@ -1525,14 +1525,14 @@ if ((re->options & PCRE_ANCHORED) == 0 && (void)pcre32_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES, (void *)(&tables)); #endif - + compile_block.lcc = tables + lcc_offset; compile_block.fcc = tables + fcc_offset; compile_block.cbits = tables + cbits_offset; compile_block.ctypes = tables + ctypes_offset; - + /* See if we can find a fixed set of initial characters for the pattern. */ - + memset(start_bits, 0, 32 * sizeof(pcre_uint8)); rc = set_start_bits(code, start_bits, (re->options & PCRE_UTF8) != 0, &compile_block); @@ -1543,16 +1543,16 @@ if ((re->options & PCRE_ANCHORED) == 0 && return NULL; } } - + /* Find the minimum length of subject string. */ - + switch(min = find_minlength(re, code, code, re->options, NULL, &count)) - { + { case -2: *errorptr = "internal error: missing capturing bracket"; return NULL; case -3: *errorptr = "internal error: opcode not recognized"; return NULL; default: break; - } - + } + /* If a set of starting bytes has been identified, or if the minimum length is greater than zero, or if JIT optimization has been requested, or if PCRE_STUDY_EXTRA_NEEDED is set, get a pcre[16]_extra block and a @@ -1561,7 +1561,7 @@ by the former, which may also get additional data set later by the calling program. At the moment, the size of pcre_study_data is fixed. We nevertheless save it in a field for returning via the pcre_fullinfo() function so that if it becomes variable in the future, we don't have to change that code. */ - + if (bits_set || min > 0 || (options & ( #ifdef SUPPORT_JIT PCRE_STUDY_JIT_COMPILE | PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE | @@ -1576,7 +1576,7 @@ if (bits_set || min > 0 || (options & ( *errorptr = "failed to get memory"; return NULL; } - + study = (pcre_study_data *)((char *)extra + sizeof(PUBL(extra))); extra->flags = PCRE_EXTRA_STUDY_DATA; extra->study_data = study; @@ -1648,9 +1648,9 @@ if (bits_set || min > 0 || (options & ( #endif } -return extra; -} - +return extra; +} + /************************************************* * Free the study data * @@ -1683,4 +1683,4 @@ if ((extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 && PUBL(free)(extra); } -/* End of pcre_study.c */ +/* End of pcre_study.c */ diff --git a/contrib/libs/pcre/pcre_tables.c b/contrib/libs/pcre/pcre_tables.c index 179038d025..00abeff330 100644 --- a/contrib/libs/pcre/pcre_tables.c +++ b/contrib/libs/pcre/pcre_tables.c @@ -1,103 +1,103 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2017 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + #ifndef PCRE_INCLUDED - -/* This module contains some fixed tables that are used by more than one of the -PCRE code modules. The tables are also #included by the pcretest program, which -uses macros to change their names from _pcre_xxx to xxxx, thereby avoiding name -clashes with the library. */ - - -#ifdef HAVE_CONFIG_H + +/* This module contains some fixed tables that are used by more than one of the +PCRE code modules. The tables are also #included by the pcretest program, which +uses macros to change their names from _pcre_xxx to xxxx, thereby avoiding name +clashes with the library. */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - +#endif + +#include "pcre_internal.h" + #endif /* PCRE_INCLUDED */ - -/* Table of sizes for the fixed-length opcodes. It's defined in a macro so that -the definition is next to the definition of the opcodes in pcre_internal.h. */ - + +/* Table of sizes for the fixed-length opcodes. It's defined in a macro so that +the definition is next to the definition of the opcodes in pcre_internal.h. */ + const pcre_uint8 PRIV(OP_lengths)[] = { OP_LENGTHS }; - + /* Tables of horizontal and vertical whitespace characters, suitable for adding to classes. */ - + const pcre_uint32 PRIV(hspace_list)[] = { HSPACE_LIST }; const pcre_uint32 PRIV(vspace_list)[] = { VSPACE_LIST }; + - -/************************************************* -* Tables for UTF-8 support * -*************************************************/ - -/* These are the breakpoints for different numbers of bytes in a UTF-8 -character. */ - +/************************************************* +* Tables for UTF-8 support * +*************************************************/ + +/* These are the breakpoints for different numbers of bytes in a UTF-8 +character. */ + #if (defined SUPPORT_UTF && defined COMPILE_PCRE8) \ || (defined PCRE_INCLUDED && (defined SUPPORT_PCRE16 || defined SUPPORT_PCRE32)) - + /* These tables are also required by pcretest in 16- or 32-bit mode. */ const int PRIV(utf8_table1)[] = - { 0x7f, 0x7ff, 0xffff, 0x1fffff, 0x3ffffff, 0x7fffffff}; - + { 0x7f, 0x7ff, 0xffff, 0x1fffff, 0x3ffffff, 0x7fffffff}; + const int PRIV(utf8_table1_size) = sizeof(PRIV(utf8_table1)) / sizeof(int); - -/* These are the indicator bits and the mask for the data bits to set in the -first byte of a character, indexed by the number of additional bytes. */ - + +/* These are the indicator bits and the mask for the data bits to set in the +first byte of a character, indexed by the number of additional bytes. */ + const int PRIV(utf8_table2)[] = { 0, 0xc0, 0xe0, 0xf0, 0xf8, 0xfc}; const int PRIV(utf8_table3)[] = { 0xff, 0x1f, 0x0f, 0x07, 0x03, 0x01}; - -/* Table of the number of extra bytes, indexed by the first byte masked with -0x3f. The highest number for a valid UTF-8 first byte is in fact 0x3d. */ - + +/* Table of the number of extra bytes, indexed by the first byte masked with +0x3f. The highest number for a valid UTF-8 first byte is in fact 0x3d. */ + const pcre_uint8 PRIV(utf8_table4)[] = { - 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, - 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, - 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, - 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 }; - + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, + 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 }; + #endif /* (SUPPORT_UTF && COMPILE_PCRE8) || (PCRE_INCLUDED && SUPPORT_PCRE[16|32])*/ #ifdef SUPPORT_UTF @@ -190,18 +190,18 @@ const int PRIV(ucp_typerange)[] = { }; #endif /* SUPPORT_JIT */ -/* The pcre_utt[] table below translates Unicode property names into type and -code values. It is searched by binary chop, so must be in collating sequence of -name. Originally, the table contained pointers to the name strings in the first -field of each entry. However, that leads to a large number of relocations when -a shared library is dynamically loaded. A significant reduction is made by -putting all the names into a single, large string and then using offsets in the -table itself. Maintenance is more error-prone, but frequent changes to this +/* The pcre_utt[] table below translates Unicode property names into type and +code values. It is searched by binary chop, so must be in collating sequence of +name. Originally, the table contained pointers to the name strings in the first +field of each entry. However, that leads to a large number of relocations when +a shared library is dynamically loaded. A significant reduction is made by +putting all the names into a single, large string and then using offsets in the +table itself. Maintenance is more error-prone, but frequent changes to this data are unlikely. - + July 2008: There is now a script called maint/GenerateUtt.py that can be used to generate this data automatically instead of maintaining it by hand. - + The script was updated in March 2009 to generate a new EBCDIC-compliant version. Like all other character and string literals that are compared against the regular expression pattern, we must use STR_ macros instead of literal @@ -718,10 +718,10 @@ const ucp_type_table PRIV(utt)[] = { { 1277, PT_PC, ucp_Zl }, { 1280, PT_PC, ucp_Zp }, { 1283, PT_PC, ucp_Zs } -}; - +}; + const int PRIV(utt_size) = sizeof(PRIV(utt)) / sizeof(ucp_type_table); - + #endif /* SUPPORT_UTF */ - -/* End of pcre_tables.c */ + +/* End of pcre_tables.c */ diff --git a/contrib/libs/pcre/pcre_valid_utf8.c b/contrib/libs/pcre/pcre_valid_utf8.c index 3983ed1d68..d81291b7f2 100644 --- a/contrib/libs/pcre/pcre_valid_utf8.c +++ b/contrib/libs/pcre/pcre_valid_utf8.c @@ -1,72 +1,72 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2013 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains an internal function for validating UTF-8 character -strings. */ - - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains an internal function for validating UTF-8 character +strings. */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - - -/************************************************* -* Validate a UTF-8 string * -*************************************************/ - -/* This function is called (optionally) at the start of compile or match, to +#endif + +#include "pcre_internal.h" + + +/************************************************* +* Validate a UTF-8 string * +*************************************************/ + +/* This function is called (optionally) at the start of compile or match, to check that a supposed UTF-8 string is actually valid. The early check means -that subsequent code can assume it is dealing with a valid string. The check +that subsequent code can assume it is dealing with a valid string. The check can be turned off for maximum performance, but the consequences of supplying an invalid string are then undefined. - -Originally, this function checked according to RFC 2279, allowing for values in -the range 0 to 0x7fffffff, up to 6 bytes long, but ensuring that they were in -the canonical format. Once somebody had pointed out RFC 3629 to me (it -obsoletes 2279), additional restrictions were applied. The values are now -limited to be between 0 and 0x0010ffff, no more than 4 bytes long, and the + +Originally, this function checked according to RFC 2279, allowing for values in +the range 0 to 0x7fffffff, up to 6 bytes long, but ensuring that they were in +the canonical format. Once somebody had pointed out RFC 3629 to me (it +obsoletes 2279), additional restrictions were applied. The values are now +limited to be between 0 and 0x0010ffff, no more than 4 bytes long, and the subrange 0xd000 to 0xdfff is excluded. However, the format of 5-byte and 6-byte characters is still checked. - + From release 8.13 more information about the details of the error are passed back in the returned value: @@ -94,31 +94,31 @@ PCRE_UTF8_ERR20 Isolated 0x80 byte (not within UTF-8 character) PCRE_UTF8_ERR21 Byte with the illegal value 0xfe or 0xff PCRE_UTF8_ERR22 Unused (was non-character) -Arguments: - string points to the string - length length of string, or -1 if the string is zero-terminated +Arguments: + string points to the string + length length of string, or -1 if the string is zero-terminated errp pointer to an error position offset variable - + Returns: = 0 if the string is a valid UTF-8 string > 0 otherwise, setting the offset of the bad character -*/ - -int +*/ + +int PRIV(valid_utf)(PCRE_PUCHAR string, int length, int *erroroffset) -{ +{ #ifdef SUPPORT_UTF register PCRE_PUCHAR p; - -if (length < 0) - { - for (p = string; *p != 0; p++); + +if (length < 0) + { + for (p = string; *p != 0; p++); length = (int)(p - string); - } - -for (p = string; length-- > 0; p++) - { + } + +for (p = string; length-- > 0; p++) + { register pcre_uchar ab, c, d; - + c = *p; if (c < 128) continue; /* ASCII character */ @@ -142,35 +142,35 @@ for (p = string; length-- > 0; p++) } length -= ab; /* Length remaining */ - /* Check top bits in the second byte */ - + /* Check top bits in the second byte */ + if (((d = *(++p)) & 0xc0) != 0x80) { *erroroffset = (int)(p - string) - 1; return PCRE_UTF8_ERR6; } - + /* For each length, check that the remaining bytes start with the 0x80 bit set and not the 0x40 bit. Then check for an overlong sequence, and for the excluded range 0xd800 to 0xdfff. */ - switch (ab) - { + switch (ab) + { /* 2-byte character. No further bytes to check for 0x80. Check first byte for for xx00 000x (overlong sequence). */ - + case 1: if ((c & 0x3e) == 0) { *erroroffset = (int)(p - string) - 1; return PCRE_UTF8_ERR15; } break; - + /* 3-byte character. Check third byte for 0x80. Then check first 2 bytes for 1110 0000, xx0x xxxx (overlong sequence) or 1110 1101, 1010 xxxx (0xd800 - 0xdfff) */ - - case 2: + + case 2: if ((*(++p) & 0xc0) != 0x80) /* Third byte */ { *erroroffset = (int)(p - string) - 2; @@ -186,13 +186,13 @@ for (p = string; length-- > 0; p++) *erroroffset = (int)(p - string) - 2; return PCRE_UTF8_ERR14; } - break; - + break; + /* 4-byte character. Check 3rd and 4th bytes for 0x80. Then check first 2 bytes for for 1111 0000, xx00 xxxx (overlong sequence), then check for a character greater than 0x0010ffff (f4 8f bf bf) */ - - case 3: + + case 3: if ((*(++p) & 0xc0) != 0x80) /* Third byte */ { *erroroffset = (int)(p - string) - 2; @@ -213,17 +213,17 @@ for (p = string; length-- > 0; p++) *erroroffset = (int)(p - string) - 3; return PCRE_UTF8_ERR13; } - break; - + break; + /* 5-byte and 6-byte characters are not allowed by RFC 3629, and will be rejected by the length test below. However, we do the appropriate tests here so that overlong sequences get diagnosed, and also in case there is ever an option for handling these larger code points. */ - + /* 5-byte character. Check 3rd, 4th, and 5th bytes for 0x80. Then check for 1111 1000, xx00 0xxx */ - case 4: + case 4: if ((*(++p) & 0xc0) != 0x80) /* Third byte */ { *erroroffset = (int)(p - string) - 2; @@ -244,12 +244,12 @@ for (p = string; length-- > 0; p++) *erroroffset = (int)(p - string) - 4; return PCRE_UTF8_ERR18; } - break; - + break; + /* 6-byte character. Check 3rd-6th bytes for 0x80. Then check for 1111 1100, xx00 00xx. */ - case 5: + case 5: if ((*(++p) & 0xc0) != 0x80) /* Third byte */ { *erroroffset = (int)(p - string) - 2; @@ -275,27 +275,27 @@ for (p = string; length-- > 0; p++) *erroroffset = (int)(p - string) - 5; return PCRE_UTF8_ERR19; } - break; - } - + break; + } + /* Character is valid under RFC 2279, but 4-byte and 5-byte characters are excluded by RFC 3629. The pointer p is currently at the last byte of the character. */ if (ab > 3) - { + { *erroroffset = (int)(p - string) - ab; return (ab == 4)? PCRE_UTF8_ERR11 : PCRE_UTF8_ERR12; - } - } + } + } #else /* Not SUPPORT_UTF */ (void)(string); /* Keep picky compilers happy */ (void)(length); (void)(erroroffset); -#endif - +#endif + return PCRE_UTF8_ERR0; /* This indicates success */ -} - -/* End of pcre_valid_utf8.c */ +} + +/* End of pcre_valid_utf8.c */ diff --git a/contrib/libs/pcre/pcre_version.c b/contrib/libs/pcre/pcre_version.c index 2ff2b79b8c..ea896e1d80 100644 --- a/contrib/libs/pcre/pcre_version.c +++ b/contrib/libs/pcre/pcre_version.c @@ -1,87 +1,87 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2012 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains the external function pcre_version(), which returns a -string that identifies the PCRE version that is in use. */ - - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains the external function pcre_version(), which returns a +string that identifies the PCRE version that is in use. */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - - -/************************************************* -* Return version string * -*************************************************/ - -/* These macros are the standard way of turning unquoted text into C strings. -They allow macros like PCRE_MAJOR to be defined without quotes, which is -convenient for user programs that want to test its value. */ - -#define STRING(a) # a -#define XSTRING(s) STRING(s) - -/* A problem turned up with PCRE_PRERELEASE, which is defined empty for -production releases. Originally, it was used naively in this code: - - return XSTRING(PCRE_MAJOR) - "." XSTRING(PCRE_MINOR) - XSTRING(PCRE_PRERELEASE) - " " XSTRING(PCRE_DATE); - -However, when PCRE_PRERELEASE is empty, this leads to an attempted expansion of -STRING(). The C standard states: "If (before argument substitution) any -argument consists of no preprocessing tokens, the behavior is undefined." It -turns out the gcc treats this case as a single empty string - which is what we -really want - but Visual C grumbles about the lack of an argument for the -macro. Unfortunately, both are within their rights. To cope with both ways of -handling this, I had resort to some messy hackery that does a test at run time. -I could find no way of detecting that a macro is defined as an empty string at -pre-processor time. This hack uses a standard trick for avoiding calling -the STRING macro with an empty argument when doing the test. */ - +#endif + +#include "pcre_internal.h" + + +/************************************************* +* Return version string * +*************************************************/ + +/* These macros are the standard way of turning unquoted text into C strings. +They allow macros like PCRE_MAJOR to be defined without quotes, which is +convenient for user programs that want to test its value. */ + +#define STRING(a) # a +#define XSTRING(s) STRING(s) + +/* A problem turned up with PCRE_PRERELEASE, which is defined empty for +production releases. Originally, it was used naively in this code: + + return XSTRING(PCRE_MAJOR) + "." XSTRING(PCRE_MINOR) + XSTRING(PCRE_PRERELEASE) + " " XSTRING(PCRE_DATE); + +However, when PCRE_PRERELEASE is empty, this leads to an attempted expansion of +STRING(). The C standard states: "If (before argument substitution) any +argument consists of no preprocessing tokens, the behavior is undefined." It +turns out the gcc treats this case as a single empty string - which is what we +really want - but Visual C grumbles about the lack of an argument for the +macro. Unfortunately, both are within their rights. To cope with both ways of +handling this, I had resort to some messy hackery that does a test at run time. +I could find no way of detecting that a macro is defined as an empty string at +pre-processor time. This hack uses a standard trick for avoiding calling +the STRING macro with an empty argument when doing the test. */ + #if defined COMPILE_PCRE8 PCRE_EXP_DEFN const char * PCRE_CALL_CONVENTION -pcre_version(void) +pcre_version(void) #elif defined COMPILE_PCRE16 PCRE_EXP_DEFN const char * PCRE_CALL_CONVENTION pcre16_version(void) @@ -89,10 +89,10 @@ pcre16_version(void) PCRE_EXP_DEFN const char * PCRE_CALL_CONVENTION pcre32_version(void) #endif -{ -return (XSTRING(Z PCRE_PRERELEASE)[1] == 0)? - XSTRING(PCRE_MAJOR.PCRE_MINOR PCRE_DATE) : - XSTRING(PCRE_MAJOR.PCRE_MINOR) XSTRING(PCRE_PRERELEASE PCRE_DATE); -} - -/* End of pcre_version.c */ +{ +return (XSTRING(Z PCRE_PRERELEASE)[1] == 0)? + XSTRING(PCRE_MAJOR.PCRE_MINOR PCRE_DATE) : + XSTRING(PCRE_MAJOR.PCRE_MINOR) XSTRING(PCRE_PRERELEASE PCRE_DATE); +} + +/* End of pcre_version.c */ diff --git a/contrib/libs/pcre/pcre_xclass.c b/contrib/libs/pcre/pcre_xclass.c index 942696ed7c..f42b4a1a9c 100644 --- a/contrib/libs/pcre/pcre_xclass.c +++ b/contrib/libs/pcre/pcre_xclass.c @@ -1,86 +1,86 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2013 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module contains an internal function that is used to match an extended + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains an internal function that is used to match an extended class. It is used by both pcre_exec() and pcre_def_exec(). */ - - -#ifdef HAVE_CONFIG_H + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - -#include "pcre_internal.h" - - -/************************************************* -* Match character against an XCLASS * -*************************************************/ - -/* This function is called to match a character against an extended class that +#endif + +#include "pcre_internal.h" + + +/************************************************* +* Match character against an XCLASS * +*************************************************/ + +/* This function is called to match a character against an extended class that might contain values > 255 and/or Unicode properties. - -Arguments: - c the character - data points to the flag byte of the XCLASS data - -Returns: TRUE if character matches, else FALSE -*/ - -BOOL + +Arguments: + c the character + data points to the flag byte of the XCLASS data + +Returns: TRUE if character matches, else FALSE +*/ + +BOOL PRIV(xclass)(pcre_uint32 c, const pcre_uchar *data, BOOL utf) -{ +{ pcre_uchar t; -BOOL negated = (*data & XCL_NOT) != 0; - +BOOL negated = (*data & XCL_NOT) != 0; + (void)utf; #ifdef COMPILE_PCRE8 /* In 8 bit mode, this must always be TRUE. Help the compiler to know that. */ utf = TRUE; #endif -/* Character values < 256 are matched against a bitmap, if one is present. If -not, we still carry on, because there may be ranges that start below 256 in the -additional data. */ - -if (c < 256) - { +/* Character values < 256 are matched against a bitmap, if one is present. If +not, we still carry on, because there may be ranges that start below 256 in the +additional data. */ + +if (c < 256) + { if ((*data & XCL_HASPROP) == 0) { if ((*data & XCL_MAP) == 0) return negated; @@ -89,19 +89,19 @@ if (c < 256) if ((*data & XCL_MAP) != 0 && (((pcre_uint8 *)(data + 1))[c/8] & (1 << (c&7))) != 0) return !negated; /* char found */ - } - -/* First skip the bit map if present. Then match against the list of Unicode -properties or large chars or ranges that end with a large char. We won't ever -encounter XCL_PROP or XCL_NOTPROP when UCP support is not compiled. */ - + } + +/* First skip the bit map if present. Then match against the list of Unicode +properties or large chars or ranges that end with a large char. We won't ever +encounter XCL_PROP or XCL_NOTPROP when UCP support is not compiled. */ + if ((*data++ & XCL_MAP) != 0) data += 32 / sizeof(pcre_uchar); - -while ((t = *data++) != XCL_END) - { + +while ((t = *data++) != XCL_END) + { pcre_uint32 x, y; - if (t == XCL_SINGLE) - { + if (t == XCL_SINGLE) + { #ifdef SUPPORT_UTF if (utf) { @@ -110,10 +110,10 @@ while ((t = *data++) != XCL_END) else #endif x = *data++; - if (c == x) return !negated; - } - else if (t == XCL_RANGE) - { + if (c == x) return !negated; + } + else if (t == XCL_RANGE) + { #ifdef SUPPORT_UTF if (utf) { @@ -126,39 +126,39 @@ while ((t = *data++) != XCL_END) x = *data++; y = *data++; } - if (c >= x && c <= y) return !negated; - } - -#ifdef SUPPORT_UCP - else /* XCL_PROP & XCL_NOTPROP */ - { + if (c >= x && c <= y) return !negated; + } + +#ifdef SUPPORT_UCP + else /* XCL_PROP & XCL_NOTPROP */ + { const ucd_record *prop = GET_UCD(c); BOOL isprop = t == XCL_PROP; - - switch(*data) - { - case PT_ANY: + + switch(*data) + { + case PT_ANY: if (isprop) return !negated; - break; - - case PT_LAMP: + break; + + case PT_LAMP: if ((prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt) == isprop) return !negated; - break; - - case PT_GC: + break; + + case PT_GC: if ((data[1] == PRIV(ucp_gentype)[prop->chartype]) == isprop) return !negated; - break; - - case PT_PC: + break; + + case PT_PC: if ((data[1] == prop->chartype) == isprop) return !negated; - break; - - case PT_SC: + break; + + case PT_SC: if ((data[1] == prop->script) == isprop) return !negated; - break; - + break; + case PT_ALNUM: if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L || PRIV(ucp_gentype)[prop->chartype] == ucp_N) == isprop) @@ -250,19 +250,19 @@ while ((t = *data++) != XCL_END) return !negated; break; - /* This should never occur, but compilers may mutter if there is no - default. */ - - default: - return FALSE; - } - - data += 2; - } -#endif /* SUPPORT_UCP */ - } - -return negated; /* char did not match */ -} - -/* End of pcre_xclass.c */ + /* This should never occur, but compilers may mutter if there is no + default. */ + + default: + return FALSE; + } + + data += 2; + } +#endif /* SUPPORT_UCP */ + } + +return negated; /* char did not match */ +} + +/* End of pcre_xclass.c */ diff --git a/contrib/libs/pcre/pcrecpp/ya.make b/contrib/libs/pcre/pcrecpp/ya.make index c832b9e56e..8eb2dacc7f 100644 --- a/contrib/libs/pcre/pcrecpp/ya.make +++ b/contrib/libs/pcre/pcrecpp/ya.make @@ -9,8 +9,8 @@ OWNER( g:cpp-contrib ) -LICENSE(BSD-3-Clause) - +LICENSE(BSD-3-Clause) + PEERDIR( contrib/libs/pcre ) diff --git a/contrib/libs/pcre/pcreposix.c b/contrib/libs/pcre/pcreposix.c index 55972c1c23..94a82336eb 100644 --- a/contrib/libs/pcre/pcreposix.c +++ b/contrib/libs/pcre/pcreposix.c @@ -1,57 +1,57 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -/* PCRE is a library of functions to support regular expressions whose syntax -and semantics are as close as possible to those of the Perl 5 language. - - Written by Philip Hazel +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel Copyright (c) 1997-2020 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - - -/* This module is a wrapper that provides a POSIX API to the underlying PCRE -functions. */ - - -#ifdef HAVE_CONFIG_H + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module is a wrapper that provides a POSIX API to the underlying PCRE +functions. */ + + +#ifdef HAVE_CONFIG_H #include "pcre_config.h" -#endif - - -/* Ensure that the PCREPOSIX_EXP_xxx macros are set appropriately for -compiling these functions. This must come before including pcreposix.h, where -they are set for an application (using these functions) if they have not -previously been set. */ - +#endif + + +/* Ensure that the PCREPOSIX_EXP_xxx macros are set appropriately for +compiling these functions. This must come before including pcreposix.h, where +they are set for an application (using these functions) if they have not +previously been set. */ + #if defined(_WIN32) && !defined(PCRE_STATIC) # define PCREPOSIX_EXP_DECL extern __declspec(dllexport) # define PCREPOSIX_EXP_DEFN __declspec(dllexport) @@ -62,89 +62,89 @@ are declared as "import" for Windows by defining PCRE_EXP_DECL as "import". This is needed even though pcre_internal.h itself includes pcre.h, because it does so after it has set PCRE_EXP_DECL to "export" if it is not already set. */ -#include "pcre.h" -#include "pcre_internal.h" -#include "pcreposix.h" - - -/* Table to translate PCRE compile time error codes into POSIX error codes. */ - -static const int eint[] = { - 0, /* no error */ - REG_EESCAPE, /* \ at end of pattern */ - REG_EESCAPE, /* \c at end of pattern */ - REG_EESCAPE, /* unrecognized character follows \ */ - REG_BADBR, /* numbers out of order in {} quantifier */ +#include "pcre.h" +#include "pcre_internal.h" +#include "pcreposix.h" + + +/* Table to translate PCRE compile time error codes into POSIX error codes. */ + +static const int eint[] = { + 0, /* no error */ + REG_EESCAPE, /* \ at end of pattern */ + REG_EESCAPE, /* \c at end of pattern */ + REG_EESCAPE, /* unrecognized character follows \ */ + REG_BADBR, /* numbers out of order in {} quantifier */ /* 5 */ - REG_BADBR, /* number too big in {} quantifier */ - REG_EBRACK, /* missing terminating ] for character class */ - REG_ECTYPE, /* invalid escape sequence in character class */ - REG_ERANGE, /* range out of order in character class */ - REG_BADRPT, /* nothing to repeat */ + REG_BADBR, /* number too big in {} quantifier */ + REG_EBRACK, /* missing terminating ] for character class */ + REG_ECTYPE, /* invalid escape sequence in character class */ + REG_ERANGE, /* range out of order in character class */ + REG_BADRPT, /* nothing to repeat */ /* 10 */ - REG_BADRPT, /* operand of unlimited repeat could match the empty string */ - REG_ASSERT, /* internal error: unexpected repeat */ - REG_BADPAT, /* unrecognized character after (? */ - REG_BADPAT, /* POSIX named classes are supported only within a class */ - REG_EPAREN, /* missing ) */ + REG_BADRPT, /* operand of unlimited repeat could match the empty string */ + REG_ASSERT, /* internal error: unexpected repeat */ + REG_BADPAT, /* unrecognized character after (? */ + REG_BADPAT, /* POSIX named classes are supported only within a class */ + REG_EPAREN, /* missing ) */ /* 15 */ - REG_ESUBREG, /* reference to non-existent subpattern */ - REG_INVARG, /* erroffset passed as NULL */ - REG_INVARG, /* unknown option bit(s) set */ - REG_EPAREN, /* missing ) after comment */ - REG_ESIZE, /* parentheses nested too deeply */ + REG_ESUBREG, /* reference to non-existent subpattern */ + REG_INVARG, /* erroffset passed as NULL */ + REG_INVARG, /* unknown option bit(s) set */ + REG_EPAREN, /* missing ) after comment */ + REG_ESIZE, /* parentheses nested too deeply */ /* 20 */ - REG_ESIZE, /* regular expression too large */ - REG_ESPACE, /* failed to get memory */ + REG_ESIZE, /* regular expression too large */ + REG_ESPACE, /* failed to get memory */ REG_EPAREN, /* unmatched parentheses */ - REG_ASSERT, /* internal error: code overflow */ - REG_BADPAT, /* unrecognized character after (?< */ + REG_ASSERT, /* internal error: code overflow */ + REG_BADPAT, /* unrecognized character after (?< */ /* 25 */ - REG_BADPAT, /* lookbehind assertion is not fixed length */ - REG_BADPAT, /* malformed number or name after (?( */ - REG_BADPAT, /* conditional group contains more than two branches */ - REG_BADPAT, /* assertion expected after (?( */ - REG_BADPAT, /* (?R or (?[+-]digits must be followed by ) */ + REG_BADPAT, /* lookbehind assertion is not fixed length */ + REG_BADPAT, /* malformed number or name after (?( */ + REG_BADPAT, /* conditional group contains more than two branches */ + REG_BADPAT, /* assertion expected after (?( */ + REG_BADPAT, /* (?R or (?[+-]digits must be followed by ) */ /* 30 */ - REG_ECTYPE, /* unknown POSIX class name */ - REG_BADPAT, /* POSIX collating elements are not supported */ - REG_INVARG, /* this version of PCRE is not compiled with PCRE_UTF8 support */ - REG_BADPAT, /* spare error */ + REG_ECTYPE, /* unknown POSIX class name */ + REG_BADPAT, /* POSIX collating elements are not supported */ + REG_INVARG, /* this version of PCRE is not compiled with PCRE_UTF8 support */ + REG_BADPAT, /* spare error */ REG_BADPAT, /* character value in \x{} or \o{} is too large */ /* 35 */ - REG_BADPAT, /* invalid condition (?(0) */ - REG_BADPAT, /* \C not allowed in lookbehind assertion */ - REG_EESCAPE, /* PCRE does not support \L, \l, \N, \U, or \u */ - REG_BADPAT, /* number after (?C is > 255 */ - REG_BADPAT, /* closing ) for (?C expected */ + REG_BADPAT, /* invalid condition (?(0) */ + REG_BADPAT, /* \C not allowed in lookbehind assertion */ + REG_EESCAPE, /* PCRE does not support \L, \l, \N, \U, or \u */ + REG_BADPAT, /* number after (?C is > 255 */ + REG_BADPAT, /* closing ) for (?C expected */ /* 40 */ - REG_BADPAT, /* recursive call could loop indefinitely */ - REG_BADPAT, /* unrecognized character after (?P */ - REG_BADPAT, /* syntax error in subpattern name (missing terminator) */ - REG_BADPAT, /* two named subpatterns have the same name */ - REG_BADPAT, /* invalid UTF-8 string */ + REG_BADPAT, /* recursive call could loop indefinitely */ + REG_BADPAT, /* unrecognized character after (?P */ + REG_BADPAT, /* syntax error in subpattern name (missing terminator) */ + REG_BADPAT, /* two named subpatterns have the same name */ + REG_BADPAT, /* invalid UTF-8 string */ /* 45 */ - REG_BADPAT, /* support for \P, \p, and \X has not been compiled */ - REG_BADPAT, /* malformed \P or \p sequence */ - REG_BADPAT, /* unknown property name after \P or \p */ - REG_BADPAT, /* subpattern name is too long (maximum 32 characters) */ - REG_BADPAT, /* too many named subpatterns (maximum 10,000) */ + REG_BADPAT, /* support for \P, \p, and \X has not been compiled */ + REG_BADPAT, /* malformed \P or \p sequence */ + REG_BADPAT, /* unknown property name after \P or \p */ + REG_BADPAT, /* subpattern name is too long (maximum 32 characters) */ + REG_BADPAT, /* too many named subpatterns (maximum 10,000) */ /* 50 */ - REG_BADPAT, /* repeated subpattern is too long */ - REG_BADPAT, /* octal value is greater than \377 (not in UTF-8 mode) */ - REG_BADPAT, /* internal error: overran compiling workspace */ - REG_BADPAT, /* internal error: previously-checked referenced subpattern not found */ - REG_BADPAT, /* DEFINE group contains more than one branch */ + REG_BADPAT, /* repeated subpattern is too long */ + REG_BADPAT, /* octal value is greater than \377 (not in UTF-8 mode) */ + REG_BADPAT, /* internal error: overran compiling workspace */ + REG_BADPAT, /* internal error: previously-checked referenced subpattern not found */ + REG_BADPAT, /* DEFINE group contains more than one branch */ /* 55 */ - REG_BADPAT, /* repeating a DEFINE group is not allowed */ - REG_INVARG, /* inconsistent NEWLINE options */ - REG_BADPAT, /* \g is not followed followed by an (optionally braced) non-zero number */ + REG_BADPAT, /* repeating a DEFINE group is not allowed */ + REG_INVARG, /* inconsistent NEWLINE options */ + REG_BADPAT, /* \g is not followed followed by an (optionally braced) non-zero number */ REG_BADPAT, /* a numbered reference must not be zero */ REG_BADPAT, /* an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT) */ /* 60 */ REG_BADPAT, /* (*VERB) not recognized */ - REG_BADPAT, /* number is too big */ - REG_BADPAT, /* subpattern name expected */ + REG_BADPAT, /* number is too big */ + REG_BADPAT, /* subpattern name expected */ REG_BADPAT, /* digit expected after (?+ */ REG_BADPAT, /* ] is an invalid data character in JavaScript compatibility mode */ /* 65 */ @@ -175,105 +175,105 @@ static const int eint[] = { REG_BADPAT, /* parentheses too deeply nested (stack check) */ REG_BADPAT, /* missing digits in \x{} or \o{} */ REG_BADPAT /* pattern too complicated */ -}; - -/* Table of texts corresponding to POSIX error codes */ - -static const char *const pstring[] = { - "", /* Dummy for value 0 */ - "internal error", /* REG_ASSERT */ - "invalid repeat counts in {}", /* BADBR */ - "pattern error", /* BADPAT */ - "? * + invalid", /* BADRPT */ - "unbalanced {}", /* EBRACE */ - "unbalanced []", /* EBRACK */ - "collation error - not relevant", /* ECOLLATE */ - "bad class", /* ECTYPE */ - "bad escape sequence", /* EESCAPE */ - "empty expression", /* EMPTY */ - "unbalanced ()", /* EPAREN */ - "bad range inside []", /* ERANGE */ - "expression too big", /* ESIZE */ - "failed to get memory", /* ESPACE */ - "bad back reference", /* ESUBREG */ - "bad argument", /* INVARG */ - "match failed" /* NOMATCH */ -}; - - - - -/************************************************* -* Translate error code to string * -*************************************************/ - +}; + +/* Table of texts corresponding to POSIX error codes */ + +static const char *const pstring[] = { + "", /* Dummy for value 0 */ + "internal error", /* REG_ASSERT */ + "invalid repeat counts in {}", /* BADBR */ + "pattern error", /* BADPAT */ + "? * + invalid", /* BADRPT */ + "unbalanced {}", /* EBRACE */ + "unbalanced []", /* EBRACK */ + "collation error - not relevant", /* ECOLLATE */ + "bad class", /* ECTYPE */ + "bad escape sequence", /* EESCAPE */ + "empty expression", /* EMPTY */ + "unbalanced ()", /* EPAREN */ + "bad range inside []", /* ERANGE */ + "expression too big", /* ESIZE */ + "failed to get memory", /* ESPACE */ + "bad back reference", /* ESUBREG */ + "bad argument", /* INVARG */ + "match failed" /* NOMATCH */ +}; + + + + +/************************************************* +* Translate error code to string * +*************************************************/ + PCREPOSIX_EXP_DEFN size_t PCRE_CALL_CONVENTION -regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size) -{ -const char *message, *addmessage; -size_t length, addlength; - -message = (errcode >= (int)(sizeof(pstring)/sizeof(char *)))? - "unknown error code" : pstring[errcode]; -length = strlen(message) + 1; - -addmessage = " at offset "; +regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size) +{ +const char *message, *addmessage; +size_t length, addlength; + +message = (errcode >= (int)(sizeof(pstring)/sizeof(char *)))? + "unknown error code" : pstring[errcode]; +length = strlen(message) + 1; + +addmessage = " at offset "; addlength = (preg != NULL && (int)preg->re_erroffset != -1)? - strlen(addmessage) + 6 : 0; - -if (errbuf_size > 0) - { - if (addlength > 0 && errbuf_size >= length + addlength) - sprintf(errbuf, "%s%s%-6d", message, addmessage, (int)preg->re_erroffset); - else - { - strncpy(errbuf, message, errbuf_size - 1); - errbuf[errbuf_size-1] = 0; - } - } - -return length + addlength; -} - - - - -/************************************************* -* Free store held by a regex * -*************************************************/ - + strlen(addmessage) + 6 : 0; + +if (errbuf_size > 0) + { + if (addlength > 0 && errbuf_size >= length + addlength) + sprintf(errbuf, "%s%s%-6d", message, addmessage, (int)preg->re_erroffset); + else + { + strncpy(errbuf, message, errbuf_size - 1); + errbuf[errbuf_size-1] = 0; + } + } + +return length + addlength; +} + + + + +/************************************************* +* Free store held by a regex * +*************************************************/ + PCREPOSIX_EXP_DEFN void PCRE_CALL_CONVENTION -regfree(regex_t *preg) -{ +regfree(regex_t *preg) +{ (PUBL(free))(preg->re_pcre); -} - - - - -/************************************************* -* Compile a regular expression * -*************************************************/ - -/* -Arguments: - preg points to a structure for recording the compiled expression - pattern the pattern to compile - cflags compilation flags - -Returns: 0 on success - various non-zero codes on failure -*/ - +} + + + + +/************************************************* +* Compile a regular expression * +*************************************************/ + +/* +Arguments: + preg points to a structure for recording the compiled expression + pattern the pattern to compile + cflags compilation flags + +Returns: 0 on success + various non-zero codes on failure +*/ + PCREPOSIX_EXP_DEFN int PCRE_CALL_CONVENTION -regcomp(regex_t *preg, const char *pattern, int cflags) -{ -const char *errorptr; -int erroffset; -int errorcode; -int options = 0; +regcomp(regex_t *preg, const char *pattern, int cflags) +{ +const char *errorptr; +int erroffset; +int errorcode; +int options = 0; int re_nsub = 0; - + if ((cflags & REG_ICASE) != 0) options |= PCRE_CASELESS; if ((cflags & REG_NEWLINE) != 0) options |= PCRE_MULTILINE; if ((cflags & REG_DOTALL) != 0) options |= PCRE_DOTALL; @@ -281,14 +281,14 @@ if ((cflags & REG_NOSUB) != 0) options |= PCRE_NO_AUTO_CAPTURE; if ((cflags & REG_UTF8) != 0) options |= PCRE_UTF8; if ((cflags & REG_UCP) != 0) options |= PCRE_UCP; if ((cflags & REG_UNGREEDY) != 0) options |= PCRE_UNGREEDY; - -preg->re_pcre = pcre_compile2(pattern, options, &errorcode, &errorptr, - &erroffset, NULL); -preg->re_erroffset = erroffset; - + +preg->re_pcre = pcre_compile2(pattern, options, &errorcode, &errorptr, + &erroffset, NULL); +preg->re_erroffset = erroffset; + /* Safety: if the error code is too big for the translation vector (which should not happen, but we all make mistakes), return REG_BADPAT. */ - + if (preg->re_pcre == NULL) { return (errorcode < (int)(sizeof(eint)/sizeof(const int)))? @@ -299,69 +299,69 @@ if (preg->re_pcre == NULL) &re_nsub); preg->re_nsub = (size_t)re_nsub; preg->re_erroffset = (size_t)(-1); /* No meaning after successful compile */ -return 0; -} - - - - -/************************************************* -* Match a regular expression * -*************************************************/ - -/* Unfortunately, PCRE requires 3 ints of working space for each captured -substring, so we have to get and release working store instead of just using -the POSIX structures as was done in earlier releases when PCRE needed only 2 -ints. However, if the number of possible capturing brackets is small, use a -block of store on the stack, to reduce the use of malloc/free. The threshold is -in a macro that can be changed at configure time. - -If REG_NOSUB was specified at compile time, the PCRE_NO_AUTO_CAPTURE flag will -be set. When this is the case, the nmatch and pmatch arguments are ignored, and -the only result is yes/no/error. */ - +return 0; +} + + + + +/************************************************* +* Match a regular expression * +*************************************************/ + +/* Unfortunately, PCRE requires 3 ints of working space for each captured +substring, so we have to get and release working store instead of just using +the POSIX structures as was done in earlier releases when PCRE needed only 2 +ints. However, if the number of possible capturing brackets is small, use a +block of store on the stack, to reduce the use of malloc/free. The threshold is +in a macro that can be changed at configure time. + +If REG_NOSUB was specified at compile time, the PCRE_NO_AUTO_CAPTURE flag will +be set. When this is the case, the nmatch and pmatch arguments are ignored, and +the only result is yes/no/error. */ + PCREPOSIX_EXP_DEFN int PCRE_CALL_CONVENTION -regexec(const regex_t *preg, const char *string, size_t nmatch, - regmatch_t pmatch[], int eflags) -{ +regexec(const regex_t *preg, const char *string, size_t nmatch, + regmatch_t pmatch[], int eflags) +{ int rc, so, eo; -int options = 0; -int *ovector = NULL; -int small_ovector[POSIX_MALLOC_THRESHOLD * 3]; -BOOL allocated_ovector = FALSE; -BOOL nosub = +int options = 0; +int *ovector = NULL; +int small_ovector[POSIX_MALLOC_THRESHOLD * 3]; +BOOL allocated_ovector = FALSE; +BOOL nosub = (REAL_PCRE_OPTIONS((const pcre *)preg->re_pcre) & PCRE_NO_AUTO_CAPTURE) != 0; - -if ((eflags & REG_NOTBOL) != 0) options |= PCRE_NOTBOL; -if ((eflags & REG_NOTEOL) != 0) options |= PCRE_NOTEOL; + +if ((eflags & REG_NOTBOL) != 0) options |= PCRE_NOTBOL; +if ((eflags & REG_NOTEOL) != 0) options |= PCRE_NOTEOL; if ((eflags & REG_NOTEMPTY) != 0) options |= PCRE_NOTEMPTY; - + /* When no string data is being returned, or no vector has been passed in which to put it, ensure that nmatch is zero. Otherwise, ensure the vector for holding the return data is large enough. */ - + if (nosub || pmatch == NULL) nmatch = 0; - -else if (nmatch > 0) - { - if (nmatch <= POSIX_MALLOC_THRESHOLD) - { - ovector = &(small_ovector[0]); - } - else - { - if (nmatch > INT_MAX/(sizeof(int) * 3)) return REG_ESPACE; - ovector = (int *)malloc(sizeof(int) * nmatch * 3); - if (ovector == NULL) return REG_ESPACE; - allocated_ovector = TRUE; - } - } - + +else if (nmatch > 0) + { + if (nmatch <= POSIX_MALLOC_THRESHOLD) + { + ovector = &(small_ovector[0]); + } + else + { + if (nmatch > INT_MAX/(sizeof(int) * 3)) return REG_ESPACE; + ovector = (int *)malloc(sizeof(int) * nmatch * 3); + if (ovector == NULL) return REG_ESPACE; + allocated_ovector = TRUE; + } + } + /* REG_STARTEND is a BSD extension, to allow for non-NUL-terminated strings. The man page from OS X says "REG_STARTEND affects only the location of the string, not how it is matched". That is why the "so" value is used to bump the start location rather than being passed as a PCRE "starting offset". */ - + if ((eflags & REG_STARTEND) != 0) { if (pmatch == NULL) return REG_INVARG; @@ -373,7 +373,7 @@ else so = 0; eo = (int)strlen(string); } - + rc = pcre_exec((const pcre *)preg->re_pcre, NULL, string + so, (eo - so), 0, options, ovector, (int)(nmatch * 3)); @@ -381,27 +381,27 @@ if (rc == 0) rc = (int)nmatch; /* All captured slots were filled in */ /* Successful match */ -if (rc >= 0) - { - size_t i; - if (!nosub) - { - for (i = 0; i < (size_t)rc; i++) - { +if (rc >= 0) + { + size_t i; + if (!nosub) + { + for (i = 0; i < (size_t)rc; i++) + { pmatch[i].rm_so = (ovector[i*2] < 0)? -1 : ovector[i*2] + so; pmatch[i].rm_eo = (ovector[i*2+1] < 0)? -1: ovector[i*2+1] + so; - } - if (allocated_ovector) free(ovector); - for (; i < nmatch; i++) pmatch[i].rm_so = pmatch[i].rm_eo = -1; - } - return 0; - } - + } + if (allocated_ovector) free(ovector); + for (; i < nmatch; i++) pmatch[i].rm_so = pmatch[i].rm_eo = -1; + } + return 0; + } + /* Unsuccessful match */ if (allocated_ovector) free(ovector); switch(rc) - { + { /* ========================================================================== */ /* These cases are never obeyed. This is a fudge that causes a compile-time error if the vector eint, which is indexed by compile-time error number, is @@ -425,7 +425,7 @@ switch(rc) case PCRE_ERROR_BADUTF8_OFFSET: return REG_INVARG; case PCRE_ERROR_BADMODE: return REG_INVARG; default: return REG_ASSERT; - } -} - -/* End of pcreposix.c */ + } +} + +/* End of pcreposix.c */ diff --git a/contrib/libs/pcre/pcreposix.h b/contrib/libs/pcre/pcreposix.h index 62cf33ae17..4667ea388d 100644 --- a/contrib/libs/pcre/pcreposix.h +++ b/contrib/libs/pcre/pcreposix.h @@ -1,57 +1,57 @@ -/************************************************* -* Perl-Compatible Regular Expressions * -*************************************************/ - -#ifndef _PCREPOSIX_H -#define _PCREPOSIX_H - -/* This is the header for the POSIX wrapper interface to the PCRE Perl- -Compatible Regular Expression library. It defines the things POSIX says should -be there. I hope. - +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +#ifndef _PCREPOSIX_H +#define _PCREPOSIX_H + +/* This is the header for the POSIX wrapper interface to the PCRE Perl- +Compatible Regular Expression library. It defines the things POSIX says should +be there. I hope. + Copyright (c) 1997-2012 University of Cambridge - ------------------------------------------------------------------------------ -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, - this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - * Neither the name of the University of Cambridge nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE -LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -POSSIBILITY OF SUCH DAMAGE. ------------------------------------------------------------------------------ -*/ - -/* Have to include stdlib.h in order to ensure that size_t is defined. */ - -#include <stdlib.h> - -/* Allow for C++ users */ - -#ifdef __cplusplus -extern "C" { -#endif - + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + +/* Have to include stdlib.h in order to ensure that size_t is defined. */ + +#include <stdlib.h> + +/* Allow for C++ users */ + +#ifdef __cplusplus +extern "C" { +#endif + /* Options, mostly defined by POSIX, but with some extras. */ - + #define REG_ICASE 0x0001 /* Maps to PCRE_CASELESS */ #define REG_NEWLINE 0x0002 /* Maps to PCRE_MULTILINE */ #define REG_NOTBOL 0x0004 /* Maps to PCRE_NOTBOL */ @@ -63,64 +63,64 @@ extern "C" { #define REG_NOTEMPTY 0x0100 /* NOT defined by POSIX; maps to PCRE_NOTEMPTY */ #define REG_UNGREEDY 0x0200 /* NOT defined by POSIX; maps to PCRE_UNGREEDY */ #define REG_UCP 0x0400 /* NOT defined by POSIX; maps to PCRE_UCP */ - -/* This is not used by PCRE, but by defining it we make it easier -to slot PCRE into existing programs that make POSIX calls. */ - -#define REG_EXTENDED 0 - -/* Error values. Not all these are relevant or used by the wrapper. */ - -enum { - REG_ASSERT = 1, /* internal error ? */ - REG_BADBR, /* invalid repeat counts in {} */ - REG_BADPAT, /* pattern error */ - REG_BADRPT, /* ? * + invalid */ - REG_EBRACE, /* unbalanced {} */ - REG_EBRACK, /* unbalanced [] */ - REG_ECOLLATE, /* collation error - not relevant */ - REG_ECTYPE, /* bad class */ - REG_EESCAPE, /* bad escape sequence */ - REG_EMPTY, /* empty expression */ - REG_EPAREN, /* unbalanced () */ - REG_ERANGE, /* bad range inside [] */ - REG_ESIZE, /* expression too big */ - REG_ESPACE, /* failed to get memory */ - REG_ESUBREG, /* bad back reference */ - REG_INVARG, /* bad argument */ - REG_NOMATCH /* match failed */ -}; - - -/* The structure representing a compiled regular expression. */ - -typedef struct { - void *re_pcre; - size_t re_nsub; - size_t re_erroffset; -} regex_t; - -/* The structure in which a captured offset is returned. */ - -typedef int regoff_t; - -typedef struct { - regoff_t rm_so; - regoff_t rm_eo; -} regmatch_t; - -/* When an application links to a PCRE DLL in Windows, the symbols that are -imported have to be identified as such. When building PCRE, the appropriate -export settings are needed, and are set in pcreposix.c before including this -file. */ - + +/* This is not used by PCRE, but by defining it we make it easier +to slot PCRE into existing programs that make POSIX calls. */ + +#define REG_EXTENDED 0 + +/* Error values. Not all these are relevant or used by the wrapper. */ + +enum { + REG_ASSERT = 1, /* internal error ? */ + REG_BADBR, /* invalid repeat counts in {} */ + REG_BADPAT, /* pattern error */ + REG_BADRPT, /* ? * + invalid */ + REG_EBRACE, /* unbalanced {} */ + REG_EBRACK, /* unbalanced [] */ + REG_ECOLLATE, /* collation error - not relevant */ + REG_ECTYPE, /* bad class */ + REG_EESCAPE, /* bad escape sequence */ + REG_EMPTY, /* empty expression */ + REG_EPAREN, /* unbalanced () */ + REG_ERANGE, /* bad range inside [] */ + REG_ESIZE, /* expression too big */ + REG_ESPACE, /* failed to get memory */ + REG_ESUBREG, /* bad back reference */ + REG_INVARG, /* bad argument */ + REG_NOMATCH /* match failed */ +}; + + +/* The structure representing a compiled regular expression. */ + +typedef struct { + void *re_pcre; + size_t re_nsub; + size_t re_erroffset; +} regex_t; + +/* The structure in which a captured offset is returned. */ + +typedef int regoff_t; + +typedef struct { + regoff_t rm_so; + regoff_t rm_eo; +} regmatch_t; + +/* When an application links to a PCRE DLL in Windows, the symbols that are +imported have to be identified as such. When building PCRE, the appropriate +export settings are needed, and are set in pcreposix.c before including this +file. */ + #if defined(_WIN32) && !defined(PCRE_STATIC) && !defined(PCREPOSIX_EXP_DECL) # define PCREPOSIX_EXP_DECL extern __declspec(dllimport) # define PCREPOSIX_EXP_DEFN __declspec(dllimport) #endif -/* By default, we use the standard "extern" declarations. */ - +/* By default, we use the standard "extern" declarations. */ + #ifndef PCREPOSIX_EXP_DECL # ifdef __cplusplus # define PCREPOSIX_EXP_DECL extern "C" @@ -131,8 +131,8 @@ file. */ # endif #endif -/* The functions */ - +/* The functions */ + #define regcomp pcre_regcomp #define regexec pcre_regexec #define regerror pcre_regerror @@ -140,12 +140,12 @@ file. */ PCREPOSIX_EXP_DECL int regcomp(regex_t *, const char *, int); PCREPOSIX_EXP_DECL int regexec(const regex_t *, const char *, size_t, - regmatch_t *, int); + regmatch_t *, int); PCREPOSIX_EXP_DECL size_t regerror(int, const regex_t *, char *, size_t); PCREPOSIX_EXP_DECL void regfree(regex_t *); - -#ifdef __cplusplus -} /* extern "C" */ -#endif - -#endif /* End of pcreposix.h */ + +#ifdef __cplusplus +} /* extern "C" */ +#endif + +#endif /* End of pcreposix.h */ diff --git a/contrib/libs/pcre/ucp.h b/contrib/libs/pcre/ucp.h index 2fa00296e4..a052b6d069 100644 --- a/contrib/libs/pcre/ucp.h +++ b/contrib/libs/pcre/ucp.h @@ -1,14 +1,14 @@ -/************************************************* -* Unicode Property Table handler * -*************************************************/ - -#ifndef _UCP_H -#define _UCP_H - -/* This file contains definitions of the property values that are returned by +/************************************************* +* Unicode Property Table handler * +*************************************************/ + +#ifndef _UCP_H +#define _UCP_H + +/* This file contains definitions of the property values that are returned by the UCD access macros. New values that are added for new releases of Unicode should always be at the end of each enum, for backwards compatibility. - + IMPORTANT: Note also that the specific numeric values of the enums have to be the same as the values that are generated by the maint/MultiStage2.py script, where the equivalent property descriptive names are listed in vectors. @@ -16,53 +16,53 @@ where the equivalent property descriptive names are listed in vectors. ALSO: The specific values of the first two enums are assumed for the table called catposstab in pcre_compile.c. */ -/* These are the general character categories. */ - -enum { - ucp_C, /* Other */ - ucp_L, /* Letter */ - ucp_M, /* Mark */ - ucp_N, /* Number */ - ucp_P, /* Punctuation */ - ucp_S, /* Symbol */ - ucp_Z /* Separator */ -}; - +/* These are the general character categories. */ + +enum { + ucp_C, /* Other */ + ucp_L, /* Letter */ + ucp_M, /* Mark */ + ucp_N, /* Number */ + ucp_P, /* Punctuation */ + ucp_S, /* Symbol */ + ucp_Z /* Separator */ +}; + /* These are the particular character categories. */ - -enum { - ucp_Cc, /* Control */ - ucp_Cf, /* Format */ - ucp_Cn, /* Unassigned */ - ucp_Co, /* Private use */ - ucp_Cs, /* Surrogate */ - ucp_Ll, /* Lower case letter */ - ucp_Lm, /* Modifier letter */ - ucp_Lo, /* Other letter */ - ucp_Lt, /* Title case letter */ - ucp_Lu, /* Upper case letter */ - ucp_Mc, /* Spacing mark */ - ucp_Me, /* Enclosing mark */ - ucp_Mn, /* Non-spacing mark */ - ucp_Nd, /* Decimal number */ - ucp_Nl, /* Letter number */ - ucp_No, /* Other number */ - ucp_Pc, /* Connector punctuation */ - ucp_Pd, /* Dash punctuation */ - ucp_Pe, /* Close punctuation */ - ucp_Pf, /* Final punctuation */ - ucp_Pi, /* Initial punctuation */ - ucp_Po, /* Other punctuation */ - ucp_Ps, /* Open punctuation */ - ucp_Sc, /* Currency symbol */ - ucp_Sk, /* Modifier symbol */ - ucp_Sm, /* Mathematical symbol */ - ucp_So, /* Other symbol */ - ucp_Zl, /* Line separator */ - ucp_Zp, /* Paragraph separator */ - ucp_Zs /* Space separator */ -}; - + +enum { + ucp_Cc, /* Control */ + ucp_Cf, /* Format */ + ucp_Cn, /* Unassigned */ + ucp_Co, /* Private use */ + ucp_Cs, /* Surrogate */ + ucp_Ll, /* Lower case letter */ + ucp_Lm, /* Modifier letter */ + ucp_Lo, /* Other letter */ + ucp_Lt, /* Title case letter */ + ucp_Lu, /* Upper case letter */ + ucp_Mc, /* Spacing mark */ + ucp_Me, /* Enclosing mark */ + ucp_Mn, /* Non-spacing mark */ + ucp_Nd, /* Decimal number */ + ucp_Nl, /* Letter number */ + ucp_No, /* Other number */ + ucp_Pc, /* Connector punctuation */ + ucp_Pd, /* Dash punctuation */ + ucp_Pe, /* Close punctuation */ + ucp_Pf, /* Final punctuation */ + ucp_Pi, /* Initial punctuation */ + ucp_Po, /* Other punctuation */ + ucp_Ps, /* Open punctuation */ + ucp_Sc, /* Currency symbol */ + ucp_Sk, /* Modifier symbol */ + ucp_Sm, /* Mathematical symbol */ + ucp_So, /* Other symbol */ + ucp_Zl, /* Line separator */ + ucp_Zp, /* Paragraph separator */ + ucp_Zs /* Space separator */ +}; + /* These are grapheme break properties. Note that the code for processing them assumes that the values are less than 16. If more values are added that take the number to 16 or more, the code will have to be rewritten. */ @@ -83,70 +83,70 @@ enum { ucp_gbOther /* 12 */ }; -/* These are the script identifications. */ - -enum { - ucp_Arabic, - ucp_Armenian, - ucp_Bengali, - ucp_Bopomofo, - ucp_Braille, - ucp_Buginese, - ucp_Buhid, - ucp_Canadian_Aboriginal, - ucp_Cherokee, - ucp_Common, - ucp_Coptic, - ucp_Cypriot, - ucp_Cyrillic, - ucp_Deseret, - ucp_Devanagari, - ucp_Ethiopic, - ucp_Georgian, - ucp_Glagolitic, - ucp_Gothic, - ucp_Greek, - ucp_Gujarati, - ucp_Gurmukhi, - ucp_Han, - ucp_Hangul, - ucp_Hanunoo, - ucp_Hebrew, - ucp_Hiragana, - ucp_Inherited, - ucp_Kannada, - ucp_Katakana, - ucp_Kharoshthi, - ucp_Khmer, - ucp_Lao, - ucp_Latin, - ucp_Limbu, - ucp_Linear_B, - ucp_Malayalam, - ucp_Mongolian, - ucp_Myanmar, - ucp_New_Tai_Lue, - ucp_Ogham, - ucp_Old_Italic, - ucp_Old_Persian, - ucp_Oriya, - ucp_Osmanya, - ucp_Runic, - ucp_Shavian, - ucp_Sinhala, - ucp_Syloti_Nagri, - ucp_Syriac, - ucp_Tagalog, - ucp_Tagbanwa, - ucp_Tai_Le, - ucp_Tamil, - ucp_Telugu, - ucp_Thaana, - ucp_Thai, - ucp_Tibetan, - ucp_Tifinagh, - ucp_Ugaritic, - ucp_Yi, +/* These are the script identifications. */ + +enum { + ucp_Arabic, + ucp_Armenian, + ucp_Bengali, + ucp_Bopomofo, + ucp_Braille, + ucp_Buginese, + ucp_Buhid, + ucp_Canadian_Aboriginal, + ucp_Cherokee, + ucp_Common, + ucp_Coptic, + ucp_Cypriot, + ucp_Cyrillic, + ucp_Deseret, + ucp_Devanagari, + ucp_Ethiopic, + ucp_Georgian, + ucp_Glagolitic, + ucp_Gothic, + ucp_Greek, + ucp_Gujarati, + ucp_Gurmukhi, + ucp_Han, + ucp_Hangul, + ucp_Hanunoo, + ucp_Hebrew, + ucp_Hiragana, + ucp_Inherited, + ucp_Kannada, + ucp_Katakana, + ucp_Kharoshthi, + ucp_Khmer, + ucp_Lao, + ucp_Latin, + ucp_Limbu, + ucp_Linear_B, + ucp_Malayalam, + ucp_Mongolian, + ucp_Myanmar, + ucp_New_Tai_Lue, + ucp_Ogham, + ucp_Old_Italic, + ucp_Old_Persian, + ucp_Oriya, + ucp_Osmanya, + ucp_Runic, + ucp_Shavian, + ucp_Sinhala, + ucp_Syloti_Nagri, + ucp_Syriac, + ucp_Tagalog, + ucp_Tagbanwa, + ucp_Tai_Le, + ucp_Tamil, + ucp_Telugu, + ucp_Thaana, + ucp_Thai, + ucp_Tibetan, + ucp_Tifinagh, + ucp_Ugaritic, + ucp_Yi, /* New for Unicode 5.0: */ ucp_Balinese, ucp_Cuneiform, @@ -217,8 +217,8 @@ enum { ucp_Siddham, ucp_Tirhuta, ucp_Warang_Citi -}; - -#endif - -/* End of ucp.h */ +}; + +#endif + +/* End of ucp.h */ diff --git a/contrib/libs/pcre/ya.make b/contrib/libs/pcre/ya.make index 85ae150697..26b8b1a571 100644 --- a/contrib/libs/pcre/ya.make +++ b/contrib/libs/pcre/ya.make @@ -1,12 +1,12 @@ # Generated by devtools/yamaker from nixpkgs a58a0b5098f0c2a389ee70eb69422a052982d990. -LIBRARY() +LIBRARY() OWNER( orivej g:cpp-contrib ) - + VERSION(8.44) ORIGINAL_SOURCE(https://ftp.pcre.org/pub/pcre/pcre-8.44.tar.bz2) @@ -26,14 +26,14 @@ ADDINCL( ) NO_COMPILER_WARNINGS() - + NO_RUNTIME() CFLAGS( GLOBAL -DPCRE_STATIC -DHAVE_CONFIG_H ) - + # JIT adds ≈108KB to binary size which may be critical for mobile and embedded devices binary distributions DEFAULT(ARCADIA_PCRE_ENABLE_JIT yes) @@ -43,32 +43,32 @@ IF (ARCADIA_PCRE_ENABLE_JIT) ) ENDIF() -SRCS( +SRCS( pcre_byte_order.c pcre_chartables.c - pcre_compile.c - pcre_config.c - pcre_dfa_exec.c - pcre_exec.c - pcre_fullinfo.c - pcre_get.c - pcre_globals.c + pcre_compile.c + pcre_config.c + pcre_dfa_exec.c + pcre_exec.c + pcre_fullinfo.c + pcre_get.c + pcre_globals.c pcre_jit_compile.c - pcre_maketables.c - pcre_newline.c - pcre_ord2utf8.c - pcre_refcount.c + pcre_maketables.c + pcre_newline.c + pcre_ord2utf8.c + pcre_refcount.c pcre_string_utils.c - pcre_study.c - pcre_tables.c + pcre_study.c + pcre_tables.c pcre_ucd.c - pcre_valid_utf8.c - pcre_version.c - pcre_xclass.c - pcreposix.c -) - -END() + pcre_valid_utf8.c + pcre_version.c + pcre_xclass.c + pcreposix.c +) + +END() RECURSE( pcre16 |