aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/libs/icu/common/unistr_case.cpp
blob: 2138d60c01c2d17be8dae6c51f7e1a17403e2c43 (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
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
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
*
*   Copyright (C) 1999-2014, International Business Machines
*   Corporation and others.  All Rights Reserved.
*
*******************************************************************************
*   file name:  unistr_case.cpp
*   encoding:   UTF-8
*   tab size:   8 (not used)
*   indentation:2
*
*   created on: 2004aug19
*   created by: Markus W. Scherer
*
*   Case-mapping functions moved here from unistr.cpp
*/

#include "unicode/utypes.h"
#include "unicode/brkiter.h"
#include "unicode/casemap.h"
#include "unicode/edits.h"
#include "unicode/putil.h"
#include "cstring.h"
#include "cmemory.h"
#include "unicode/ustring.h"
#include "unicode/unistr.h"
#include "unicode/uchar.h"
#include "uassert.h"
#include "ucasemap_imp.h"
#include "uelement.h"

U_NAMESPACE_BEGIN

//========================================
// Read-only implementation
//========================================

int8_t
UnicodeString::doCaseCompare(int32_t start,
                             int32_t length,
                             const UChar *srcChars,
                             int32_t srcStart,
                             int32_t srcLength,
                             uint32_t options) const
{
  // compare illegal string values
  // treat const UChar *srcChars==NULL as an empty string
  if(isBogus()) {
    return -1;
  }

  // pin indices to legal values
  pinIndices(start, length);

  if(srcChars == NULL) {
    srcStart = srcLength = 0;
  }

  // get the correct pointer
  const UChar *chars = getArrayStart();

  chars += start;
  if(srcStart!=0) {
    srcChars += srcStart;
  }

  if(chars != srcChars) {
    UErrorCode errorCode=U_ZERO_ERROR;
    int32_t result=u_strcmpFold(chars, length, srcChars, srcLength,
                                options|U_COMPARE_IGNORE_CASE, &errorCode);
    if(result!=0) {
      return (int8_t)(result >> 24 | 1);
    }
  } else {
    // get the srcLength if necessary
    if(srcLength < 0) {
      srcLength = u_strlen(srcChars + srcStart);
    }
    if(length != srcLength) {
      return (int8_t)((length - srcLength) >> 24 | 1);
    }
  }
  return 0;
}

//========================================
// Write implementation
//========================================

UnicodeString &
UnicodeString::caseMap(int32_t caseLocale, uint32_t options, UCASEMAP_BREAK_ITERATOR_PARAM
                       UStringCaseMapper *stringCaseMapper) {
  if(isEmpty() || !isWritable()) {
    // nothing to do
    return *this;
  }

  UChar oldBuffer[2 * US_STACKBUF_SIZE];
  UChar *oldArray;
  int32_t oldLength = length();
  int32_t newLength;
  UBool writable = isBufferWritable();
  UErrorCode errorCode = U_ZERO_ERROR;

#if !UCONFIG_NO_BREAK_ITERATION
  // Read-only alias to the original string contents for the titlecasing BreakIterator.
  // We cannot set the iterator simply to *this because *this is being modified.
  UnicodeString oldString;
#endif

  // Try to avoid heap-allocating a new character array for this string.
  if (writable ? oldLength <= UPRV_LENGTHOF(oldBuffer) : oldLength < US_STACKBUF_SIZE) {
    // Short string: Copy the contents into a temporary buffer and
    // case-map back into the current array, or into the stack buffer.
    UChar *buffer = getArrayStart();
    int32_t capacity;
    oldArray = oldBuffer;
    u_memcpy(oldBuffer, buffer, oldLength);
    if (writable) {
      capacity = getCapacity();
    } else {
      // Switch from the read-only alias or shared heap buffer to the stack buffer.
      if (!cloneArrayIfNeeded(US_STACKBUF_SIZE, US_STACKBUF_SIZE, /* doCopyArray= */ FALSE)) {
        return *this;
      }
      U_ASSERT(fUnion.fFields.fLengthAndFlags & kUsingStackBuffer);
      buffer = fUnion.fStackFields.fBuffer;
      capacity = US_STACKBUF_SIZE;
    }
#if !UCONFIG_NO_BREAK_ITERATION
    if (iter != nullptr) {
      oldString.setTo(FALSE, oldArray, oldLength);
      iter->setText(oldString);
    }
#endif
    newLength = stringCaseMapper(caseLocale, options, UCASEMAP_BREAK_ITERATOR
                                 buffer, capacity,
                                 oldArray, oldLength, NULL, errorCode);
    if (U_SUCCESS(errorCode)) {
      setLength(newLength);
      return *this;
    } else if (errorCode == U_BUFFER_OVERFLOW_ERROR) {
      // common overflow handling below
    } else {
      setToBogus();
      return *this;
    }
  } else {
    // Longer string or read-only buffer:
    // Collect only changes and then apply them to this string.
    // Case mapping often changes only small parts of a string,
    // and often does not change its length.
    oldArray = getArrayStart();
    Edits edits;
    UChar replacementChars[200];
#if !UCONFIG_NO_BREAK_ITERATION
    if (iter != nullptr) {
      oldString.setTo(FALSE, oldArray, oldLength);
      iter->setText(oldString);
    }
#endif
    stringCaseMapper(caseLocale, options | U_OMIT_UNCHANGED_TEXT, UCASEMAP_BREAK_ITERATOR
                     replacementChars, UPRV_LENGTHOF(replacementChars),
                     oldArray, oldLength, &edits, errorCode);
    if (U_SUCCESS(errorCode)) {
      // Grow the buffer at most once, not for multiple doReplace() calls.
      newLength = oldLength + edits.lengthDelta();
      if (newLength > oldLength && !cloneArrayIfNeeded(newLength, newLength)) {
        return *this;
      }
      for (Edits::Iterator ei = edits.getCoarseChangesIterator(); ei.next(errorCode);) {
        doReplace(ei.destinationIndex(), ei.oldLength(),
                  replacementChars, ei.replacementIndex(), ei.newLength());
      }
      if (U_FAILURE(errorCode)) {
        setToBogus();
      }
      return *this;
    } else if (errorCode == U_BUFFER_OVERFLOW_ERROR) {
      // common overflow handling below
      newLength = oldLength + edits.lengthDelta();
    } else {
      setToBogus();
      return *this;
    }
  }

  // Handle buffer overflow, newLength is known.
  // We need to allocate a new buffer for the internal string case mapping function.
  // This is very similar to how doReplace() keeps the old array pointer
  // and deletes the old array itself after it is done.
  // In addition, we are forcing cloneArrayIfNeeded() to always allocate a new array.
  int32_t *bufferToDelete = 0;
  if (!cloneArrayIfNeeded(newLength, newLength, FALSE, &bufferToDelete, TRUE)) {
    return *this;
  }
  errorCode = U_ZERO_ERROR;
  // No need to iter->setText() again: The case mapper restarts via iter->first().
  newLength = stringCaseMapper(caseLocale, options, UCASEMAP_BREAK_ITERATOR
                               getArrayStart(), getCapacity(),
                               oldArray, oldLength, NULL, errorCode);
  if (bufferToDelete) {
    uprv_free(bufferToDelete);
  }
  if (U_SUCCESS(errorCode)) {
    setLength(newLength);
  } else {
    setToBogus();
  }
  return *this;
}

UnicodeString &
UnicodeString::foldCase(uint32_t options) {
  return caseMap(UCASE_LOC_ROOT, options, UCASEMAP_BREAK_ITERATOR_NULL ustrcase_internalFold);
}

U_NAMESPACE_END

// Defined here to reduce dependencies on break iterator
U_CAPI int32_t U_EXPORT2
uhash_hashCaselessUnicodeString(const UElement key) {
    U_NAMESPACE_USE
    const UnicodeString *str = (const UnicodeString*) key.pointer;
    if (str == NULL) {
        return 0;
    }
    // Inefficient; a better way would be to have a hash function in
    // UnicodeString that does case folding on the fly.
    UnicodeString copy(*str);
    return copy.foldCase().hashCode();
}

// Defined here to reduce dependencies on break iterator
U_CAPI UBool U_EXPORT2
uhash_compareCaselessUnicodeString(const UElement key1, const UElement key2) {
    U_NAMESPACE_USE
    const UnicodeString *str1 = (const UnicodeString*) key1.pointer;
    const UnicodeString *str2 = (const UnicodeString*) key2.pointer;
    if (str1 == str2) {
        return TRUE;
    }
    if (str1 == NULL || str2 == NULL) {
        return FALSE;
    }
    return str1->caseCompare(*str2, U_FOLD_CASE_DEFAULT) == 0;
}