aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/libs/icu/i18n/inputext.cpp
blob: 0351518b704e08e08d46da0715736a0ea634fcb0 (plain) (blame)
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
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
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
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
 **********************************************************************
 *   Copyright (C) 2005-2016, International Business Machines
 *   Corporation and others.  All Rights Reserved.
 **********************************************************************
 */

#include "unicode/utypes.h"

#if !UCONFIG_NO_CONVERSION

#include "inputext.h"

#include "cmemory.h"
#include "cstring.h"

#include <string.h>

U_NAMESPACE_BEGIN

#define BUFFER_SIZE 8192

#define NEW_ARRAY(type,count) (type *) uprv_malloc((count) * sizeof(type))
#define DELETE_ARRAY(array) uprv_free((void *) (array))

InputText::InputText(UErrorCode &status)
    : fInputBytes(NEW_ARRAY(uint8_t, BUFFER_SIZE)), // The text to be checked.  Markup will have been
                                                 //   removed if appropriate.
      fByteStats(NEW_ARRAY(int16_t, 256)),       // byte frequency statistics for the input text.
                                                 //   Value is percent, not absolute.
      fDeclaredEncoding(nullptr),
      fRawInput(nullptr),
      fRawLength(0)
{
    if (fInputBytes == nullptr || fByteStats == nullptr) {
        status = U_MEMORY_ALLOCATION_ERROR;
    }
}

InputText::~InputText()
{
    DELETE_ARRAY(fDeclaredEncoding);
    DELETE_ARRAY(fByteStats);
    DELETE_ARRAY(fInputBytes);
}

void InputText::setText(const char *in, int32_t len)
{
    fInputLen  = 0;
    fC1Bytes   = false;
    fRawInput  = (const uint8_t *) in;
    fRawLength = len == -1? (int32_t)uprv_strlen(in) : len;
}

void InputText::setDeclaredEncoding(const char* encoding, int32_t len)
{
    if(encoding) {
        if (len == -1) {
            len = (int32_t)uprv_strlen(encoding);
        }

        len += 1;     // to make place for the \0 at the end.
        uprv_free(fDeclaredEncoding);
        fDeclaredEncoding = NEW_ARRAY(char, len);
        uprv_strncpy(fDeclaredEncoding, encoding, len);
    }
}

UBool InputText::isSet() const 
{
    return fRawInput != nullptr;
}

/**
*  MungeInput - after getting a set of raw input data to be analyzed, preprocess
*               it by removing what appears to be html markup.
* 
* @internal
*/
void InputText::MungeInput(UBool fStripTags) {
    int     srci = 0;
    int     dsti = 0;
    uint8_t b;
    bool    inMarkup = false;
    int32_t openTags = 0;
    int32_t badTags  = 0;

    //
    //  html / xml markup stripping.
    //     quick and dirty, not 100% accurate, but hopefully good enough, statistically.
    //     discard everything within < brackets >
    //     Count how many total '<' and illegal (nested) '<' occur, so we can make some
    //     guess as to whether the input was actually marked up at all.
    // TODO: Think about how this interacts with EBCDIC charsets that are detected.
    if (fStripTags) {
        for (srci = 0; srci < fRawLength && dsti < BUFFER_SIZE; srci += 1) {
            b = fRawInput[srci];

            if (b == (uint8_t)0x3C) { /* Check for the ASCII '<' */
                if (inMarkup) {
                    badTags += 1;
                }

                inMarkup = true;
                openTags += 1;
            }

            if (! inMarkup) {
                fInputBytes[dsti++] = b;
            }

            if (b == (uint8_t)0x3E) { /* Check for the ASCII '>' */
                inMarkup = false;
            }
        }

        fInputLen = dsti;
    }

    //
    //  If it looks like this input wasn't marked up, or if it looks like it's
    //    essentially nothing but markup abandon the markup stripping.
    //    Detection will have to work on the unstripped input.
    //
    if (openTags<5 || openTags/5 < badTags || 
        (fInputLen < 100 && fRawLength>600))
    {
        int32_t limit = fRawLength;

        if (limit > BUFFER_SIZE) {
            limit = BUFFER_SIZE;
        }

        for (srci=0; srci<limit; srci++) {
            fInputBytes[srci] = fRawInput[srci];
        }

        fInputLen = srci;
    }

    //
    // Tally up the byte occurrence statistics.
    // These are available for use by the various detectors.
    //

    uprv_memset(fByteStats, 0, (sizeof fByteStats[0]) * 256);

    for (srci = 0; srci < fInputLen; srci += 1) {
        fByteStats[fInputBytes[srci]] += 1;
    }

    for (int32_t i = 0x80; i <= 0x9F; i += 1) {
        if (fByteStats[i] != 0) {
            fC1Bytes = true;
            break;
        }
    }
}

U_NAMESPACE_END
#endif