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
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
|
/*
* Copyright (C) 2004, 2006, 2008 Apple Inc. All rights reserved.
* Copyright (C) 2006 Alexey Proskuryakov <ap@nypop.com>
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. 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.
*
* THIS SOFTWARE IS PROVIDED BY APPLE COMPUTER, INC. ``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 APPLE COMPUTER, INC. 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.
*/
#include "config.h"
#include "TextCodecMac.h"
#include "CString.h"
#include "CharacterNames.h"
#include "CharsetData.h"
#include "PlatformString.h"
#include "ThreadGlobalData.h"
#include <wtf/Assertions.h>
#include <wtf/PassOwnPtr.h>
#include <wtf/Threading.h>
using namespace std;
namespace WebCore {
// We need to keep this because ICU doesn't support some of the encodings that we need:
// <http://bugs.webkit.org/show_bug.cgi?id=4195>.
const size_t ConversionBufferSize = 16384;
static TECConverterWrapper& cachedConverterTEC()
{
return threadGlobalData().cachedConverterTEC();
}
void TextCodecMac::registerEncodingNames(EncodingNameRegistrar registrar)
{
TECTextEncodingID lastEncoding = invalidEncoding;
const char* lastName = 0;
for (size_t i = 0; CharsetTable[i].name; ++i) {
if (CharsetTable[i].encoding != lastEncoding) {
lastEncoding = CharsetTable[i].encoding;
lastName = CharsetTable[i].name;
}
registrar(CharsetTable[i].name, lastName);
}
}
static PassOwnPtr<TextCodec> newTextCodecMac(const TextEncoding&, const void* additionalData)
{
return new TextCodecMac(*static_cast<const TECTextEncodingID*>(additionalData));
}
void TextCodecMac::registerCodecs(TextCodecRegistrar registrar)
{
TECTextEncodingID lastEncoding = invalidEncoding;
for (size_t i = 0; CharsetTable[i].name; ++i)
if (CharsetTable[i].encoding != lastEncoding) {
registrar(CharsetTable[i].name, newTextCodecMac, &CharsetTable[i].encoding);
lastEncoding = CharsetTable[i].encoding;
}
}
TextCodecMac::TextCodecMac(TECTextEncodingID encoding)
: m_encoding(encoding)
, m_numBufferedBytes(0)
, m_converterTEC(0)
{
}
TextCodecMac::~TextCodecMac()
{
releaseTECConverter();
}
void TextCodecMac::releaseTECConverter() const
{
if (m_converterTEC) {
TECConverterWrapper& cachedConverter = cachedConverterTEC();
if (cachedConverter.converter)
TECDisposeConverter(cachedConverter.converter);
cachedConverter.converter = m_converterTEC;
cachedConverter.encoding = m_encoding;
m_converterTEC = 0;
}
}
OSStatus TextCodecMac::createTECConverter() const
{
TECConverterWrapper& cachedConverter = cachedConverterTEC();
bool cachedEncodingEqual = cachedConverter.encoding == m_encoding;
cachedConverter.encoding = invalidEncoding;
if (cachedEncodingEqual && cachedConverter.converter) {
m_converterTEC = cachedConverter.converter;
cachedConverter.converter = 0;
TECClearConverterContextInfo(m_converterTEC);
} else {
OSStatus status = TECCreateConverter(&m_converterTEC, m_encoding,
CreateTextEncoding(kTextEncodingUnicodeDefault, kTextEncodingDefaultVariant, kUnicode16BitFormat));
if (status)
return status;
TECSetBasicOptions(m_converterTEC, kUnicodeForceASCIIRangeMask);
}
return noErr;
}
OSStatus TextCodecMac::decode(const unsigned char* inputBuffer, int inputBufferLength, int& inputLength,
void *outputBuffer, int outputBufferLength, int& outputLength)
{
OSStatus status;
unsigned long bytesRead = 0;
unsigned long bytesWritten = 0;
if (m_numBufferedBytes != 0) {
// Finish converting a partial character that's in our buffer.
// First, fill the partial character buffer with as many bytes as are available.
ASSERT(m_numBufferedBytes < sizeof(m_bufferedBytes));
const int spaceInBuffer = sizeof(m_bufferedBytes) - m_numBufferedBytes;
const int bytesToPutInBuffer = min(spaceInBuffer, inputBufferLength);
ASSERT(bytesToPutInBuffer != 0);
memcpy(m_bufferedBytes + m_numBufferedBytes, inputBuffer, bytesToPutInBuffer);
// Now, do a conversion on the buffer.
status = TECConvertText(m_converterTEC, m_bufferedBytes, m_numBufferedBytes + bytesToPutInBuffer, &bytesRead,
reinterpret_cast<unsigned char*>(outputBuffer), outputBufferLength, &bytesWritten);
ASSERT(bytesRead <= m_numBufferedBytes + bytesToPutInBuffer);
if (status == kTECPartialCharErr && bytesRead == 0) {
// Handle the case where the partial character was not converted.
if (bytesToPutInBuffer >= spaceInBuffer) {
LOG_ERROR("TECConvertText gave a kTECPartialCharErr but read none of the %zu bytes in the buffer", sizeof(m_bufferedBytes));
m_numBufferedBytes = 0;
status = kTECUnmappableElementErr; // should never happen, but use this error code
} else {
// Tell the caller we read all the source bytes and keep them in the buffer.
m_numBufferedBytes += bytesToPutInBuffer;
bytesRead = bytesToPutInBuffer;
status = noErr;
}
} else {
// We are done with the partial character buffer.
// Also, we have read some of the bytes from the main buffer.
if (bytesRead > m_numBufferedBytes) {
bytesRead -= m_numBufferedBytes;
} else {
LOG_ERROR("TECConvertText accepted some bytes it previously rejected with kTECPartialCharErr");
bytesRead = 0;
}
m_numBufferedBytes = 0;
if (status == kTECPartialCharErr) {
// While there may be a partial character problem in the small buffer,
// we have to try again and not get confused and think there is a partial
// character problem in the large buffer.
status = noErr;
}
}
} else {
status = TECConvertText(m_converterTEC, inputBuffer, inputBufferLength, &bytesRead,
static_cast<unsigned char*>(outputBuffer), outputBufferLength, &bytesWritten);
ASSERT(static_cast<int>(bytesRead) <= inputBufferLength);
}
// Work around bug 3351093, where sometimes we get kTECBufferBelowMinimumSizeErr instead of kTECOutputBufferFullStatus.
if (status == kTECBufferBelowMinimumSizeErr && bytesWritten != 0)
status = kTECOutputBufferFullStatus;
inputLength = bytesRead;
outputLength = bytesWritten;
return status;
}
String TextCodecMac::decode(const char* bytes, size_t length, bool flush, bool stopOnError, bool& sawError)
{
// Get a converter for the passed-in encoding.
if (!m_converterTEC && createTECConverter() != noErr)
return String();
Vector<UChar> result;
const unsigned char* sourcePointer = reinterpret_cast<const unsigned char*>(bytes);
int sourceLength = length;
bool bufferWasFull = false;
UniChar buffer[ConversionBufferSize];
while ((sourceLength || bufferWasFull) && !sawError) {
int bytesRead = 0;
int bytesWritten = 0;
OSStatus status = decode(sourcePointer, sourceLength, bytesRead, buffer, sizeof(buffer), bytesWritten);
ASSERT(bytesRead <= sourceLength);
sourcePointer += bytesRead;
sourceLength -= bytesRead;
switch (status) {
case noErr:
case kTECOutputBufferFullStatus:
break;
case kTextMalformedInputErr:
case kTextUndefinedElementErr:
// FIXME: Put FFFD character into the output string in this case?
TECClearConverterContextInfo(m_converterTEC);
if (stopOnError) {
sawError = true;
break;
}
if (sourceLength) {
sourcePointer += 1;
sourceLength -= 1;
}
break;
case kTECPartialCharErr: {
// Put the partial character into the buffer.
ASSERT(m_numBufferedBytes == 0);
const int bufferSize = sizeof(m_numBufferedBytes);
if (sourceLength < bufferSize) {
memcpy(m_bufferedBytes, sourcePointer, sourceLength);
m_numBufferedBytes = sourceLength;
} else {
LOG_ERROR("TECConvertText gave a kTECPartialCharErr, but left %u bytes in the buffer", sourceLength);
}
sourceLength = 0;
break;
}
default:
sawError = true;
return String();
}
ASSERT(!(bytesWritten % sizeof(UChar)));
result.append(buffer, bytesWritten / sizeof(UChar));
bufferWasFull = status == kTECOutputBufferFullStatus;
}
if (flush) {
unsigned long bytesWritten = 0;
TECFlushText(m_converterTEC, reinterpret_cast<unsigned char*>(buffer), sizeof(buffer), &bytesWritten);
ASSERT(!(bytesWritten % sizeof(UChar)));
result.append(buffer, bytesWritten / sizeof(UChar));
}
String resultString = String::adopt(result);
// <rdar://problem/3225472>
// Simplified Chinese pages use the code A3A0 to mean "full-width space".
// But GB18030 decodes it to U+E5E5, which is correct in theory but not in practice.
// To work around, just change all occurences of U+E5E5 to U+3000 (ideographic space).
if (m_encoding == kCFStringEncodingGB_18030_2000)
resultString.replace(0xE5E5, ideographicSpace);
return resultString;
}
CString TextCodecMac::encode(const UChar* characters, size_t length, UnencodableHandling handling)
{
// FIXME: We should really use TEC here instead of CFString for consistency with the other direction.
// FIXME: Since there's no "force ASCII range" mode in CFString, we change the backslash into a yen sign.
// Encoding will change the yen sign back into a backslash.
String copy(characters, length);
copy.replace('\\', m_backslashAsCurrencySymbol);
CFStringRef cfs = copy.createCFString();
CFIndex startPos = 0;
CFIndex charactersLeft = CFStringGetLength(cfs);
Vector<char> result;
size_t size = 0;
UInt8 lossByte = handling == QuestionMarksForUnencodables ? '?' : 0;
while (charactersLeft > 0) {
CFRange range = CFRangeMake(startPos, charactersLeft);
CFIndex bufferLength;
CFStringGetBytes(cfs, range, m_encoding, lossByte, false, NULL, 0x7FFFFFFF, &bufferLength);
result.grow(size + bufferLength);
unsigned char* buffer = reinterpret_cast<unsigned char*>(result.data() + size);
CFIndex charactersConverted = CFStringGetBytes(cfs, range, m_encoding, lossByte, false, buffer, bufferLength, &bufferLength);
size += bufferLength;
if (charactersConverted != charactersLeft) {
unsigned badChar = CFStringGetCharacterAtIndex(cfs, startPos + charactersConverted);
++charactersConverted;
if ((badChar & 0xFC00) == 0xD800 && charactersConverted != charactersLeft) { // is high surrogate
UniChar low = CFStringGetCharacterAtIndex(cfs, startPos + charactersConverted);
if ((low & 0xFC00) == 0xDC00) { // is low surrogate
badChar <<= 10;
badChar += low;
badChar += 0x10000 - (0xD800 << 10) - 0xDC00;
++charactersConverted;
}
}
UnencodableReplacementArray entity;
int entityLength = getUnencodableReplacement(badChar, handling, entity);
result.grow(size + entityLength);
memcpy(result.data() + size, entity, entityLength);
size += entityLength;
}
startPos += charactersConverted;
charactersLeft -= charactersConverted;
}
CFRelease(cfs);
return CString(result.data(), size);
}
} // namespace WebCore
|