|
1
|
|
|
<?php |
|
2
|
|
|
namespace TYPO3\CMS\Core\Charset; |
|
3
|
|
|
|
|
4
|
|
|
/* |
|
5
|
|
|
* This file is part of the TYPO3 CMS project. |
|
6
|
|
|
* |
|
7
|
|
|
* It is free software; you can redistribute it and/or modify it under |
|
8
|
|
|
* the terms of the GNU General Public License, either version 2 |
|
9
|
|
|
* of the License, or any later version. |
|
10
|
|
|
* |
|
11
|
|
|
* For the full copyright and license information, please read the |
|
12
|
|
|
* LICENSE.txt file that was distributed with this source code. |
|
13
|
|
|
* |
|
14
|
|
|
* The TYPO3 project - inspiring people to share! |
|
15
|
|
|
*/ |
|
16
|
|
|
|
|
17
|
|
|
use TYPO3\CMS\Core\SingletonInterface; |
|
18
|
|
|
use TYPO3\CMS\Core\Utility\ExtensionManagementUtility; |
|
19
|
|
|
use TYPO3\CMS\Core\Utility\GeneralUtility; |
|
20
|
|
|
|
|
21
|
|
|
/** |
|
22
|
|
|
* Notes on UTF-8 |
|
23
|
|
|
* |
|
24
|
|
|
* Functions working on UTF-8 strings: |
|
25
|
|
|
* |
|
26
|
|
|
* - strchr/strstr |
|
27
|
|
|
* - strrchr |
|
28
|
|
|
* - substr_count |
|
29
|
|
|
* - implode/explode/join |
|
30
|
|
|
* |
|
31
|
|
|
* Functions nearly working on UTF-8 strings: |
|
32
|
|
|
* |
|
33
|
|
|
* - trim/ltrim/rtrim: the second parameter 'charlist' won't work for characters not contained in 7-bit ASCII |
|
34
|
|
|
* - htmlentities: charset support for UTF-8 only since PHP 4.3.0 |
|
35
|
|
|
* - preg_*: Support compiled into PHP by default nowadays, but could be unavailable, need to use modifier |
|
36
|
|
|
* |
|
37
|
|
|
* Functions NOT working on UTF-8 strings: |
|
38
|
|
|
* |
|
39
|
|
|
* - str*cmp |
|
40
|
|
|
* - stristr |
|
41
|
|
|
* - stripos |
|
42
|
|
|
* - substr |
|
43
|
|
|
* - strrev |
|
44
|
|
|
* - split/spliti |
|
45
|
|
|
* - ... |
|
46
|
|
|
*/ |
|
47
|
|
|
|
|
48
|
|
|
/** |
|
49
|
|
|
* Class for conversion between charsets |
|
50
|
|
|
*/ |
|
51
|
|
|
class CharsetConverter implements SingletonInterface |
|
52
|
|
|
{ |
|
53
|
|
|
/** |
|
54
|
|
|
* ASCII Value for chars with no equivalent. |
|
55
|
|
|
* |
|
56
|
|
|
* @var int |
|
57
|
|
|
*/ |
|
58
|
|
|
public $noCharByteVal = 63; |
|
59
|
|
|
|
|
60
|
|
|
/** |
|
61
|
|
|
* This is the array where parsed conversion tables are stored (cached) |
|
62
|
|
|
* |
|
63
|
|
|
* @var array |
|
64
|
|
|
*/ |
|
65
|
|
|
public $parsedCharsets = []; |
|
66
|
|
|
|
|
67
|
|
|
/** |
|
68
|
|
|
* An array where case folding data will be stored (cached) |
|
69
|
|
|
* |
|
70
|
|
|
* @var array |
|
71
|
|
|
*/ |
|
72
|
|
|
public $caseFolding = []; |
|
73
|
|
|
|
|
74
|
|
|
/** |
|
75
|
|
|
* An array where charset-to-ASCII mappings are stored (cached) |
|
76
|
|
|
* |
|
77
|
|
|
* @var array |
|
78
|
|
|
*/ |
|
79
|
|
|
public $toASCII = []; |
|
80
|
|
|
|
|
81
|
|
|
/** |
|
82
|
|
|
* This tells the converter which charsets has two bytes per char: |
|
83
|
|
|
* |
|
84
|
|
|
* @var array |
|
85
|
|
|
*/ |
|
86
|
|
|
public $twoByteSets = [ |
|
87
|
|
|
'ucs-2' => 1 |
|
88
|
|
|
]; |
|
89
|
|
|
|
|
90
|
|
|
/** |
|
91
|
|
|
* This tells the converter which charsets use a scheme like the Extended Unix Code: |
|
92
|
|
|
* |
|
93
|
|
|
* @var array |
|
94
|
|
|
*/ |
|
95
|
|
|
public $eucBasedSets = [ |
|
96
|
|
|
'gb2312' => 1, // Chinese, simplified. |
|
97
|
|
|
'big5' => 1, // Chinese, traditional. |
|
98
|
|
|
'euc-kr' => 1, // Korean |
|
99
|
|
|
'shift_jis' => 1 |
|
100
|
|
|
]; |
|
101
|
|
|
|
|
102
|
|
|
/** |
|
103
|
|
|
* @link http://developer.apple.com/documentation/macos8/TextIntlSvcs/TextEncodingConversionManager/TEC1.5/TEC.b0.html |
|
104
|
|
|
* @link http://czyborra.com/charsets/iso8859.html |
|
105
|
|
|
* |
|
106
|
|
|
* @var array |
|
107
|
|
|
*/ |
|
108
|
|
|
public $synonyms = [ |
|
109
|
|
|
'us' => 'ascii', |
|
110
|
|
|
'us-ascii' => 'ascii', |
|
111
|
|
|
'cp819' => 'iso-8859-1', |
|
112
|
|
|
'ibm819' => 'iso-8859-1', |
|
113
|
|
|
'iso-ir-100' => 'iso-8859-1', |
|
114
|
|
|
'iso-ir-101' => 'iso-8859-2', |
|
115
|
|
|
'iso-ir-109' => 'iso-8859-3', |
|
116
|
|
|
'iso-ir-110' => 'iso-8859-4', |
|
117
|
|
|
'iso-ir-144' => 'iso-8859-5', |
|
118
|
|
|
'iso-ir-127' => 'iso-8859-6', |
|
119
|
|
|
'iso-ir-126' => 'iso-8859-7', |
|
120
|
|
|
'iso-ir-138' => 'iso-8859-8', |
|
121
|
|
|
'iso-ir-148' => 'iso-8859-9', |
|
122
|
|
|
'iso-ir-157' => 'iso-8859-10', |
|
123
|
|
|
'iso-ir-179' => 'iso-8859-13', |
|
124
|
|
|
'iso-ir-199' => 'iso-8859-14', |
|
125
|
|
|
'iso-ir-203' => 'iso-8859-15', |
|
126
|
|
|
'csisolatin1' => 'iso-8859-1', |
|
127
|
|
|
'csisolatin2' => 'iso-8859-2', |
|
128
|
|
|
'csisolatin3' => 'iso-8859-3', |
|
129
|
|
|
'csisolatin5' => 'iso-8859-9', |
|
130
|
|
|
'csisolatin8' => 'iso-8859-14', |
|
131
|
|
|
'csisolatin9' => 'iso-8859-15', |
|
132
|
|
|
'csisolatingreek' => 'iso-8859-7', |
|
133
|
|
|
'iso-celtic' => 'iso-8859-14', |
|
134
|
|
|
'latin1' => 'iso-8859-1', |
|
135
|
|
|
'latin2' => 'iso-8859-2', |
|
136
|
|
|
'latin3' => 'iso-8859-3', |
|
137
|
|
|
'latin5' => 'iso-8859-9', |
|
138
|
|
|
'latin6' => 'iso-8859-10', |
|
139
|
|
|
'latin8' => 'iso-8859-14', |
|
140
|
|
|
'latin9' => 'iso-8859-15', |
|
141
|
|
|
'l1' => 'iso-8859-1', |
|
142
|
|
|
'l2' => 'iso-8859-2', |
|
143
|
|
|
'l3' => 'iso-8859-3', |
|
144
|
|
|
'l5' => 'iso-8859-9', |
|
145
|
|
|
'l6' => 'iso-8859-10', |
|
146
|
|
|
'l8' => 'iso-8859-14', |
|
147
|
|
|
'l9' => 'iso-8859-15', |
|
148
|
|
|
'cyrillic' => 'iso-8859-5', |
|
149
|
|
|
'arabic' => 'iso-8859-6', |
|
150
|
|
|
'tis-620' => 'iso-8859-11', |
|
151
|
|
|
'win874' => 'windows-874', |
|
152
|
|
|
'win1250' => 'windows-1250', |
|
153
|
|
|
'win1251' => 'windows-1251', |
|
154
|
|
|
'win1252' => 'windows-1252', |
|
155
|
|
|
'win1253' => 'windows-1253', |
|
156
|
|
|
'win1254' => 'windows-1254', |
|
157
|
|
|
'win1255' => 'windows-1255', |
|
158
|
|
|
'win1256' => 'windows-1256', |
|
159
|
|
|
'win1257' => 'windows-1257', |
|
160
|
|
|
'win1258' => 'windows-1258', |
|
161
|
|
|
'cp1250' => 'windows-1250', |
|
162
|
|
|
'cp1251' => 'windows-1251', |
|
163
|
|
|
'cp1252' => 'windows-1252', |
|
164
|
|
|
'ms-ee' => 'windows-1250', |
|
165
|
|
|
'ms-ansi' => 'windows-1252', |
|
166
|
|
|
'ms-greek' => 'windows-1253', |
|
167
|
|
|
'ms-turk' => 'windows-1254', |
|
168
|
|
|
'winbaltrim' => 'windows-1257', |
|
169
|
|
|
'koi-8ru' => 'koi-8r', |
|
170
|
|
|
'koi8r' => 'koi-8r', |
|
171
|
|
|
'cp878' => 'koi-8r', |
|
172
|
|
|
'mac' => 'macroman', |
|
173
|
|
|
'macintosh' => 'macroman', |
|
174
|
|
|
'euc-cn' => 'gb2312', |
|
175
|
|
|
'x-euc-cn' => 'gb2312', |
|
176
|
|
|
'euccn' => 'gb2312', |
|
177
|
|
|
'cp936' => 'gb2312', |
|
178
|
|
|
'big-5' => 'big5', |
|
179
|
|
|
'cp950' => 'big5', |
|
180
|
|
|
'eucjp' => 'euc-jp', |
|
181
|
|
|
'sjis' => 'shift_jis', |
|
182
|
|
|
'shift-jis' => 'shift_jis', |
|
183
|
|
|
'cp932' => 'shift_jis', |
|
184
|
|
|
'cp949' => 'euc-kr', |
|
185
|
|
|
'utf7' => 'utf-7', |
|
186
|
|
|
'utf8' => 'utf-8', |
|
187
|
|
|
'utf16' => 'utf-16', |
|
188
|
|
|
'utf32' => 'utf-32', |
|
189
|
|
|
'ucs2' => 'ucs-2', |
|
190
|
|
|
'ucs4' => 'ucs-4' |
|
191
|
|
|
]; |
|
192
|
|
|
|
|
193
|
|
|
/** |
|
194
|
|
|
* Normalize - changes input character set to lowercase letters. |
|
195
|
|
|
* |
|
196
|
|
|
* @param string $charset Input charset |
|
197
|
|
|
* @return string Normalized charset |
|
198
|
|
|
*/ |
|
199
|
|
|
public function parse_charset($charset) |
|
200
|
|
|
{ |
|
201
|
|
|
$charset = trim(strtolower($charset)); |
|
202
|
|
|
if (isset($this->synonyms[$charset])) { |
|
203
|
|
|
$charset = $this->synonyms[$charset]; |
|
204
|
|
|
} |
|
205
|
|
|
return $charset; |
|
206
|
|
|
} |
|
207
|
|
|
|
|
208
|
|
|
/******************************************** |
|
209
|
|
|
* |
|
210
|
|
|
* Charset Conversion functions |
|
211
|
|
|
* |
|
212
|
|
|
********************************************/ |
|
213
|
|
|
/** |
|
214
|
|
|
* Convert from one charset to another charset. |
|
215
|
|
|
* |
|
216
|
|
|
* @param string $inputString Input string |
|
217
|
|
|
* @param string $fromCharset From charset (the current charset of the string) |
|
218
|
|
|
* @param string $toCharset To charset (the output charset wanted) |
|
219
|
|
|
* @param bool $useEntityForNoChar If set, then characters that are not available in the destination character set will be encoded as numeric entities |
|
220
|
|
|
* @return string Converted string |
|
221
|
|
|
* @see convArray() |
|
222
|
|
|
*/ |
|
223
|
|
|
public function conv($inputString, $fromCharset, $toCharset, $useEntityForNoChar = false) |
|
224
|
|
|
{ |
|
225
|
|
|
if ($fromCharset === $toCharset) { |
|
226
|
|
|
return $inputString; |
|
227
|
|
|
} |
|
228
|
|
|
// PHP-libs don't support fallback to SGML entities, but UTF-8 handles everything |
|
229
|
|
|
if ($toCharset === 'utf-8' || !$useEntityForNoChar) { |
|
230
|
|
|
// Returns FALSE for unsupported charsets |
|
231
|
|
|
$convertedString = mb_convert_encoding($inputString, $toCharset, $fromCharset); |
|
232
|
|
|
if (false !== $convertedString) { |
|
233
|
|
|
return $convertedString; |
|
234
|
|
|
} |
|
235
|
|
|
} |
|
236
|
|
|
if ($fromCharset !== 'utf-8') { |
|
237
|
|
|
$inputString = $this->utf8_encode($inputString, $fromCharset); |
|
238
|
|
|
} |
|
239
|
|
|
if ($toCharset !== 'utf-8') { |
|
240
|
|
|
$inputString = $this->utf8_decode($inputString, $toCharset, $useEntityForNoChar); |
|
241
|
|
|
} |
|
242
|
|
|
return $inputString; |
|
243
|
|
|
} |
|
244
|
|
|
|
|
245
|
|
|
/** |
|
246
|
|
|
* Convert all elements in ARRAY with type string from one charset to another charset. |
|
247
|
|
|
* NOTICE: Array is passed by reference! |
|
248
|
|
|
* |
|
249
|
|
|
* @param array $array Input array, possibly multidimensional |
|
250
|
|
|
* @param string $fromCharset From charset (the current charset of the string) |
|
251
|
|
|
* @param string $toCharset To charset (the output charset wanted) |
|
252
|
|
|
* @param bool $useEntityForNoChar If set, then characters that are not available in the destination character set will be encoded as numeric entities |
|
253
|
|
|
* @see conv() |
|
254
|
|
|
*/ |
|
255
|
|
|
public function convArray(&$array, $fromCharset, $toCharset, $useEntityForNoChar = false) |
|
256
|
|
|
{ |
|
257
|
|
|
foreach ($array as $key => $value) { |
|
258
|
|
|
if (is_array($array[$key])) { |
|
259
|
|
|
$this->convArray($array[$key], $fromCharset, $toCharset, $useEntityForNoChar); |
|
260
|
|
|
} elseif (is_string($array[$key])) { |
|
261
|
|
|
$array[$key] = $this->conv($array[$key], $fromCharset, $toCharset, $useEntityForNoChar); |
|
262
|
|
|
} |
|
263
|
|
|
} |
|
264
|
|
|
} |
|
265
|
|
|
|
|
266
|
|
|
/** |
|
267
|
|
|
* Converts $str from $charset to UTF-8 |
|
268
|
|
|
* |
|
269
|
|
|
* @param string $str String in local charset to convert to UTF-8 |
|
270
|
|
|
* @param string $charset Charset, lowercase. Must be found in csconvtbl/ folder. |
|
271
|
|
|
* @return string Output string, converted to UTF-8 |
|
272
|
|
|
*/ |
|
273
|
|
|
public function utf8_encode($str, $charset) |
|
274
|
|
|
{ |
|
275
|
|
|
if ($charset === 'utf-8') { |
|
276
|
|
|
return $str; |
|
277
|
|
|
} |
|
278
|
|
|
// Charset is case-insensitive |
|
279
|
|
|
// Parse conv. table if not already |
|
280
|
|
|
if ($this->initCharset($charset)) { |
|
281
|
|
|
$strLen = strlen($str); |
|
282
|
|
|
$outStr = ''; |
|
283
|
|
|
// Traverse each char in string |
|
284
|
|
|
for ($a = 0; $a < $strLen; $a++) { |
|
285
|
|
|
$chr = substr($str, $a, 1); |
|
286
|
|
|
$ord = ord($chr); |
|
287
|
|
|
// If the charset has two bytes per char |
|
288
|
|
|
if (isset($this->twoByteSets[$charset])) { |
|
289
|
|
|
$ord2 = ord($str[$a + 1]); |
|
290
|
|
|
// Assume big endian |
|
291
|
|
|
$ord = $ord << 8 | $ord2; |
|
292
|
|
|
// If the local char-number was found in parsed conv. table then we use that, otherwise 127 (no char?) |
|
293
|
|
|
if (isset($this->parsedCharsets[$charset]['local'][$ord])) { |
|
294
|
|
|
$outStr .= $this->parsedCharsets[$charset]['local'][$ord]; |
|
295
|
|
|
} else { |
|
296
|
|
|
$outStr .= chr($this->noCharByteVal); |
|
297
|
|
|
} |
|
298
|
|
|
// No char exists |
|
299
|
|
|
$a++; |
|
300
|
|
|
} elseif ($ord > 127) { |
|
301
|
|
|
// If char has value over 127 it's a multibyte char in UTF-8 |
|
302
|
|
|
// EUC uses two-bytes above 127; we get both and advance pointer and make $ord a 16bit int. |
|
303
|
|
|
if (isset($this->eucBasedSets[$charset])) { |
|
304
|
|
|
// Shift-JIS: chars between 160 and 223 are single byte |
|
305
|
|
|
if ($charset !== 'shift_jis' || ($ord < 160 || $ord > 223)) { |
|
306
|
|
|
$a++; |
|
307
|
|
|
$ord2 = ord(substr($str, $a, 1)); |
|
308
|
|
|
$ord = $ord * 256 + $ord2; |
|
309
|
|
|
} |
|
310
|
|
|
} |
|
311
|
|
|
if (isset($this->parsedCharsets[$charset]['local'][$ord])) { |
|
312
|
|
|
// If the local char-number was found in parsed conv. table then we use that, otherwise 127 (no char?) |
|
313
|
|
|
$outStr .= $this->parsedCharsets[$charset]['local'][$ord]; |
|
314
|
|
|
} else { |
|
315
|
|
|
$outStr .= chr($this->noCharByteVal); |
|
316
|
|
|
} |
|
317
|
|
|
} else { |
|
318
|
|
|
$outStr .= $chr; |
|
319
|
|
|
} |
|
320
|
|
|
} |
|
321
|
|
|
return $outStr; |
|
322
|
|
|
} |
|
323
|
|
|
return ''; |
|
324
|
|
|
} |
|
325
|
|
|
|
|
326
|
|
|
/** |
|
327
|
|
|
* Converts $str from UTF-8 to $charset |
|
328
|
|
|
* |
|
329
|
|
|
* @param string $str String in UTF-8 to convert to local charset |
|
330
|
|
|
* @param string $charset Charset, lowercase. Must be found in csconvtbl/ folder. |
|
331
|
|
|
* @param bool $useEntityForNoChar If set, then characters that are not available in the destination character set will be encoded as numeric entities |
|
332
|
|
|
* @return string Output string, converted to local charset |
|
333
|
|
|
*/ |
|
334
|
|
|
public function utf8_decode($str, $charset, $useEntityForNoChar = false) |
|
335
|
|
|
{ |
|
336
|
|
|
if ($charset === 'utf-8') { |
|
337
|
|
|
return $str; |
|
338
|
|
|
} |
|
339
|
|
|
// Charset is case-insensitive. |
|
340
|
|
|
// Parse conv. table if not already |
|
341
|
|
|
if ($this->initCharset($charset)) { |
|
342
|
|
|
$strLen = strlen($str); |
|
343
|
|
|
$outStr = ''; |
|
344
|
|
|
// Traverse each char in UTF-8 string |
|
345
|
|
|
for ($a = 0, $i = 0; $a < $strLen; $a++, $i++) { |
|
346
|
|
|
$chr = substr($str, $a, 1); |
|
347
|
|
|
$ord = ord($chr); |
|
348
|
|
|
// This means multibyte! (first byte!) |
|
349
|
|
|
if ($ord > 127) { |
|
350
|
|
|
// Since the first byte must have the 7th bit set we check that. Otherwise we might be in the middle of a byte sequence. |
|
351
|
|
|
if ($ord & 64) { |
|
352
|
|
|
// Add first byte |
|
353
|
|
|
$buf = $chr; |
|
354
|
|
|
// For each byte in multibyte string |
|
355
|
|
|
for ($b = 0; $b < 8; $b++) { |
|
356
|
|
|
// Shift it left and |
|
357
|
|
|
$ord = $ord << 1; |
|
358
|
|
|
// ... and with 8th bit - if that is set, then there are still bytes in sequence. |
|
359
|
|
|
if ($ord & 128) { |
|
360
|
|
|
$a++; |
|
361
|
|
|
// ... and add the next char. |
|
362
|
|
|
$buf .= substr($str, $a, 1); |
|
363
|
|
|
} else { |
|
364
|
|
|
break; |
|
365
|
|
|
} |
|
366
|
|
|
} |
|
367
|
|
|
// If the UTF-8 char-sequence is found then... |
|
368
|
|
|
if (isset($this->parsedCharsets[$charset]['utf8'][$buf])) { |
|
369
|
|
|
// The local number |
|
370
|
|
|
$mByte = $this->parsedCharsets[$charset]['utf8'][$buf]; |
|
371
|
|
|
// If the local number is greater than 255 we will need to split the byte (16bit word assumed) in two chars. |
|
372
|
|
|
if ($mByte > 255) { |
|
373
|
|
|
$outStr .= chr(($mByte >> 8 & 255)) . chr(($mByte & 255)); |
|
374
|
|
|
} else { |
|
375
|
|
|
$outStr .= chr($mByte); |
|
376
|
|
|
} |
|
377
|
|
|
} elseif ($useEntityForNoChar) { |
|
378
|
|
|
// Create num entity: |
|
379
|
|
|
$outStr .= '&#' . $this->utf8CharToUnumber($buf, true) . ';'; |
|
380
|
|
|
} else { |
|
381
|
|
|
$outStr .= chr($this->noCharByteVal); |
|
382
|
|
|
} |
|
383
|
|
|
} else { |
|
384
|
|
|
$outStr .= chr($this->noCharByteVal); |
|
385
|
|
|
} |
|
386
|
|
|
} else { |
|
387
|
|
|
$outStr .= $chr; |
|
388
|
|
|
} |
|
389
|
|
|
} |
|
390
|
|
|
return $outStr; |
|
391
|
|
|
} |
|
392
|
|
|
return ''; |
|
393
|
|
|
} |
|
394
|
|
|
|
|
395
|
|
|
/** |
|
396
|
|
|
* Converts all chars > 127 to numeric entities. |
|
397
|
|
|
* |
|
398
|
|
|
* @param string $str Input string |
|
399
|
|
|
* @return string Output string |
|
400
|
|
|
*/ |
|
401
|
|
|
public function utf8_to_entities($str) |
|
402
|
|
|
{ |
|
403
|
|
|
$strLen = strlen($str); |
|
404
|
|
|
$outStr = ''; |
|
405
|
|
|
// Traverse each char in UTF-8 string. |
|
406
|
|
|
for ($a = 0; $a < $strLen; $a++) { |
|
407
|
|
|
$chr = substr($str, $a, 1); |
|
408
|
|
|
$ord = ord($chr); |
|
409
|
|
|
// This means multibyte! (first byte!) |
|
410
|
|
|
if ($ord > 127) { |
|
411
|
|
|
// Since the first byte must have the 7th bit set we check that. Otherwise we might be in the middle of a byte sequence. |
|
412
|
|
|
if ($ord & 64) { |
|
413
|
|
|
// Add first byte |
|
414
|
|
|
$buf = $chr; |
|
415
|
|
|
// For each byte in multibyte string... |
|
416
|
|
|
for ($b = 0; $b < 8; $b++) { |
|
417
|
|
|
// Shift it left and ... |
|
418
|
|
|
$ord = $ord << 1; |
|
419
|
|
|
// ... and with 8th bit - if that is set, then there are still bytes in sequence. |
|
420
|
|
|
if ($ord & 128) { |
|
421
|
|
|
$a++; |
|
422
|
|
|
// ... and add the next char. |
|
423
|
|
|
$buf .= substr($str, $a, 1); |
|
424
|
|
|
} else { |
|
425
|
|
|
break; |
|
426
|
|
|
} |
|
427
|
|
|
} |
|
428
|
|
|
$outStr .= '&#' . $this->utf8CharToUnumber($buf, true) . ';'; |
|
429
|
|
|
} else { |
|
430
|
|
|
$outStr .= chr($this->noCharByteVal); |
|
431
|
|
|
} |
|
432
|
|
|
} else { |
|
433
|
|
|
$outStr .= $chr; |
|
434
|
|
|
} |
|
435
|
|
|
} |
|
436
|
|
|
return $outStr; |
|
437
|
|
|
} |
|
438
|
|
|
|
|
439
|
|
|
/** |
|
440
|
|
|
* Converts numeric entities (UNICODE, eg. decimal (Ӓ) or hexadecimal ()) to UTF-8 multibyte chars. |
|
441
|
|
|
* All string-HTML entities (like & or £) will be converted as well |
|
442
|
|
|
* @param string $str Input string, UTF-8 |
|
443
|
|
|
* @return string Output string |
|
444
|
|
|
*/ |
|
445
|
|
|
public function entities_to_utf8($str) |
|
446
|
|
|
{ |
|
447
|
|
|
$trans_tbl = array_flip(get_html_translation_table(HTML_ENTITIES, ENT_COMPAT)); |
|
448
|
|
|
$token = md5(microtime()); |
|
449
|
|
|
$parts = explode($token, preg_replace('/(&([#[:alnum:]]*);)/', $token . '${2}' . $token, $str)); |
|
450
|
|
|
foreach ($parts as $k => $v) { |
|
451
|
|
|
// Only take every second element |
|
452
|
|
|
if ($k % 2 === 0) { |
|
453
|
|
|
continue; |
|
454
|
|
|
} |
|
455
|
|
|
$position = 0; |
|
456
|
|
|
// Dec or hex entities |
|
457
|
|
|
if (substr($v, $position, 1) === '#') { |
|
458
|
|
|
$position++; |
|
459
|
|
|
if (substr($v, $position, 1) === 'x') { |
|
460
|
|
|
$v = hexdec(substr($v, ++$position)); |
|
461
|
|
|
} else { |
|
462
|
|
|
$v = substr($v, $position); |
|
463
|
|
|
} |
|
464
|
|
|
$parts[$k] = $this->UnumberToChar($v); |
|
|
|
|
|
|
465
|
|
|
} elseif (isset($trans_tbl['&' . $v . ';'])) { |
|
466
|
|
|
// Other entities: |
|
467
|
|
|
$v = $trans_tbl['&' . $v . ';']; |
|
468
|
|
|
$parts[$k] = $v; |
|
469
|
|
|
} else { |
|
470
|
|
|
// No conversion: |
|
471
|
|
|
$parts[$k] = '&' . $v . ';'; |
|
472
|
|
|
} |
|
473
|
|
|
} |
|
474
|
|
|
return implode('', $parts); |
|
475
|
|
|
} |
|
476
|
|
|
|
|
477
|
|
|
/** |
|
478
|
|
|
* Converts all chars in the input UTF-8 string into integer numbers returned in an array. |
|
479
|
|
|
* All HTML entities (like & or £ or { or 㽝) will be detected as characters. |
|
480
|
|
|
* Also, instead of integer numbers the real UTF-8 char is returned. |
|
481
|
|
|
* |
|
482
|
|
|
* @param string $str Input string, UTF-8 |
|
483
|
|
|
* @return array Output array with the char numbers |
|
484
|
|
|
*/ |
|
485
|
|
|
public function utf8_to_numberarray($str) |
|
486
|
|
|
{ |
|
487
|
|
|
// Entities must be registered as well |
|
488
|
|
|
$str = $this->entities_to_utf8($str); |
|
489
|
|
|
|
|
490
|
|
|
// Do conversion: |
|
491
|
|
|
$strLen = strlen($str); |
|
492
|
|
|
$outArr = []; |
|
493
|
|
|
// Traverse each char in UTF-8 string. |
|
494
|
|
|
for ($a = 0; $a < $strLen; $a++) { |
|
495
|
|
|
$chr = substr($str, $a, 1); |
|
496
|
|
|
$ord = ord($chr); |
|
497
|
|
|
// This means multibyte! (first byte!) |
|
498
|
|
|
if ($ord > 127) { |
|
499
|
|
|
// Since the first byte must have the 7th bit set we check that. Otherwise we might be in the middle of a byte sequence. |
|
500
|
|
|
if ($ord & 64) { |
|
501
|
|
|
// Add first byte |
|
502
|
|
|
$buf = $chr; |
|
503
|
|
|
// For each byte in multibyte string... |
|
504
|
|
|
for ($b = 0; $b < 8; $b++) { |
|
505
|
|
|
// Shift it left and ... |
|
506
|
|
|
$ord = $ord << 1; |
|
507
|
|
|
// ... and with 8th bit - if that is set, then there are still bytes in sequence. |
|
508
|
|
|
if ($ord & 128) { |
|
509
|
|
|
$a++; |
|
510
|
|
|
// ... and add the next char. |
|
511
|
|
|
$buf .= substr($str, $a, 1); |
|
512
|
|
|
} else { |
|
513
|
|
|
break; |
|
514
|
|
|
} |
|
515
|
|
|
} |
|
516
|
|
|
$outArr[] = $buf; |
|
517
|
|
|
} else { |
|
518
|
|
|
$outArr[] = chr($this->noCharByteVal); |
|
519
|
|
|
} |
|
520
|
|
|
} else { |
|
521
|
|
|
$outArr[] = chr($ord); |
|
522
|
|
|
} |
|
523
|
|
|
} |
|
524
|
|
|
return $outArr; |
|
525
|
|
|
} |
|
526
|
|
|
|
|
527
|
|
|
/** |
|
528
|
|
|
* Converts a UNICODE number to a UTF-8 multibyte character |
|
529
|
|
|
* Algorithm based on script found at From: http://czyborra.com/utf/ |
|
530
|
|
|
* Unit-tested by Kasper |
|
531
|
|
|
* |
|
532
|
|
|
* The binary representation of the character's integer value is thus simply spread across the bytes |
|
533
|
|
|
* and the number of high bits set in the lead byte announces the number of bytes in the multibyte sequence: |
|
534
|
|
|
* |
|
535
|
|
|
* bytes | bits | representation |
|
536
|
|
|
* 1 | 7 | 0vvvvvvv |
|
537
|
|
|
* 2 | 11 | 110vvvvv 10vvvvvv |
|
538
|
|
|
* 3 | 16 | 1110vvvv 10vvvvvv 10vvvvvv |
|
539
|
|
|
* 4 | 21 | 11110vvv 10vvvvvv 10vvvvvv 10vvvvvv |
|
540
|
|
|
* 5 | 26 | 111110vv 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv |
|
541
|
|
|
* 6 | 31 | 1111110v 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv |
|
542
|
|
|
* |
|
543
|
|
|
* @param int $unicodeInteger UNICODE integer |
|
544
|
|
|
* @return string UTF-8 multibyte character string |
|
545
|
|
|
* @see utf8CharToUnumber() |
|
546
|
|
|
*/ |
|
547
|
|
|
public function UnumberToChar($unicodeInteger) |
|
548
|
|
|
{ |
|
549
|
|
|
$str = ''; |
|
550
|
|
|
if ($unicodeInteger < 128) { |
|
551
|
|
|
$str .= chr($unicodeInteger); |
|
552
|
|
|
} elseif ($unicodeInteger < 2048) { |
|
553
|
|
|
$str .= chr(192 | $unicodeInteger >> 6); |
|
554
|
|
|
$str .= chr(128 | $unicodeInteger & 63); |
|
555
|
|
|
} elseif ($unicodeInteger < 65536) { |
|
556
|
|
|
$str .= chr(224 | $unicodeInteger >> 12); |
|
557
|
|
|
$str .= chr(128 | $unicodeInteger >> 6 & 63); |
|
558
|
|
|
$str .= chr(128 | $unicodeInteger & 63); |
|
559
|
|
|
} elseif ($unicodeInteger < 2097152) { |
|
560
|
|
|
$str .= chr(240 | $unicodeInteger >> 18); |
|
561
|
|
|
$str .= chr(128 | $unicodeInteger >> 12 & 63); |
|
562
|
|
|
$str .= chr(128 | $unicodeInteger >> 6 & 63); |
|
563
|
|
|
$str .= chr(128 | $unicodeInteger & 63); |
|
564
|
|
|
} elseif ($unicodeInteger < 67108864) { |
|
565
|
|
|
$str .= chr(248 | $unicodeInteger >> 24); |
|
566
|
|
|
$str .= chr(128 | $unicodeInteger >> 18 & 63); |
|
567
|
|
|
$str .= chr(128 | $unicodeInteger >> 12 & 63); |
|
568
|
|
|
$str .= chr(128 | $unicodeInteger >> 6 & 63); |
|
569
|
|
|
$str .= chr(128 | $unicodeInteger & 63); |
|
570
|
|
|
} elseif ($unicodeInteger < 2147483648) { |
|
571
|
|
|
$str .= chr(252 | $unicodeInteger >> 30); |
|
572
|
|
|
$str .= chr(128 | $unicodeInteger >> 24 & 63); |
|
573
|
|
|
$str .= chr(128 | $unicodeInteger >> 18 & 63); |
|
574
|
|
|
$str .= chr(128 | $unicodeInteger >> 12 & 63); |
|
575
|
|
|
$str .= chr(128 | $unicodeInteger >> 6 & 63); |
|
576
|
|
|
$str .= chr(128 | $unicodeInteger & 63); |
|
577
|
|
|
} else { |
|
578
|
|
|
// Cannot express a 32-bit character in UTF-8 |
|
579
|
|
|
$str .= chr($this->noCharByteVal); |
|
580
|
|
|
} |
|
581
|
|
|
return $str; |
|
582
|
|
|
} |
|
583
|
|
|
|
|
584
|
|
|
/** |
|
585
|
|
|
* Converts a UTF-8 Multibyte character to a UNICODE number |
|
586
|
|
|
* Unit-tested by Kasper |
|
587
|
|
|
* |
|
588
|
|
|
* @param string $str UTF-8 multibyte character string |
|
589
|
|
|
* @param bool $hex If set, then a hex. number is returned. |
|
590
|
|
|
* @return int UNICODE integer |
|
591
|
|
|
* @see UnumberToChar() |
|
592
|
|
|
*/ |
|
593
|
|
|
public function utf8CharToUnumber($str, $hex = false) |
|
594
|
|
|
{ |
|
595
|
|
|
// First char |
|
596
|
|
|
$ord = ord($str[0]); |
|
597
|
|
|
// This verifies that it IS a multi byte string |
|
598
|
|
|
if (($ord & 192) === 192) { |
|
599
|
|
|
$binBuf = ''; |
|
600
|
|
|
$b = 0; |
|
601
|
|
|
// For each byte in multibyte string... |
|
602
|
|
|
for (; $b < 8; $b++) { |
|
603
|
|
|
// Shift it left and ... |
|
604
|
|
|
$ord = $ord << 1; |
|
605
|
|
|
// ... and with 8th bit - if that is set, then there are still bytes in sequence. |
|
606
|
|
|
if ($ord & 128) { |
|
607
|
|
|
$binBuf .= substr('00000000' . decbin(ord(substr($str, ($b + 1), 1))), -6); |
|
608
|
|
|
} else { |
|
609
|
|
|
break; |
|
610
|
|
|
} |
|
611
|
|
|
} |
|
612
|
|
|
$binBuf = substr(('00000000' . decbin(ord($str[0]))), -(6 - $b)) . $binBuf; |
|
613
|
|
|
$int = bindec($binBuf); |
|
614
|
|
|
} else { |
|
615
|
|
|
$int = $ord; |
|
616
|
|
|
} |
|
617
|
|
|
return $hex ? 'x' . dechex($int) : $int; |
|
|
|
|
|
|
618
|
|
|
} |
|
619
|
|
|
|
|
620
|
|
|
/******************************************** |
|
621
|
|
|
* |
|
622
|
|
|
* Init functions |
|
623
|
|
|
* |
|
624
|
|
|
********************************************/ |
|
625
|
|
|
/** |
|
626
|
|
|
* This will initialize a charset for use if it's defined in the 'typo3/sysext/core/Resources/Private/Charsets/csconvtbl/' folder |
|
627
|
|
|
* This function is automatically called by the conversion functions |
|
628
|
|
|
* |
|
629
|
|
|
* PLEASE SEE: http://www.unicode.org/Public/MAPPINGS/ |
|
630
|
|
|
* |
|
631
|
|
|
* @param string $charset The charset to be initialized. Use lowercase charset always (the charset must match exactly with a filename in csconvtbl/ folder ([charset].tbl) |
|
632
|
|
|
* @return int Returns '1' if already loaded, '2' if the charset conversion table was found and parsed. |
|
633
|
|
|
* @throws UnknownCharsetException if no charset table was found |
|
634
|
|
|
* @access private |
|
635
|
|
|
*/ |
|
636
|
|
|
public function initCharset($charset) |
|
637
|
|
|
{ |
|
638
|
|
|
// Only process if the charset is not yet loaded: |
|
639
|
|
|
if (!is_array($this->parsedCharsets[$charset])) { |
|
640
|
|
|
// Conversion table filename: |
|
641
|
|
|
$charsetConvTableFile = ExtensionManagementUtility::extPath('core') . 'Resources/Private/Charsets/csconvtbl/' . $charset . '.tbl'; |
|
642
|
|
|
// If the conversion table is found: |
|
643
|
|
|
if ($charset && GeneralUtility::validPathStr($charsetConvTableFile) && @is_file($charsetConvTableFile)) { |
|
644
|
|
|
// Cache file for charsets: |
|
645
|
|
|
// Caching brought parsing time for gb2312 down from 2400 ms to 150 ms. For other charsets we are talking 11 ms down to zero. |
|
646
|
|
|
$cacheFile = GeneralUtility::getFileAbsFileName('typo3temp/var/charset/charset_' . $charset . '.tbl'); |
|
647
|
|
|
if ($cacheFile && @is_file($cacheFile)) { |
|
648
|
|
|
$this->parsedCharsets[$charset] = unserialize(file_get_contents($cacheFile)); |
|
649
|
|
|
} else { |
|
650
|
|
|
// Parse conversion table into lines: |
|
651
|
|
|
$lines = GeneralUtility::trimExplode(LF, file_get_contents($charsetConvTableFile), true); |
|
652
|
|
|
// Initialize the internal variable holding the conv. table: |
|
653
|
|
|
$this->parsedCharsets[$charset] = ['local' => [], 'utf8' => []]; |
|
654
|
|
|
// traverse the lines: |
|
655
|
|
|
$detectedType = ''; |
|
656
|
|
|
foreach ($lines as $value) { |
|
657
|
|
|
// Comment line or blanks are ignored. |
|
658
|
|
|
if (trim($value) && $value[0] !== '#') { |
|
659
|
|
|
// Detect type if not done yet: (Done on first real line) |
|
660
|
|
|
// The "whitespaced" type is on the syntax "0x0A 0x000A #LINE FEED" while "ms-token" is like "B9 = U+00B9 : SUPERSCRIPT ONE" |
|
661
|
|
|
if (!$detectedType) { |
|
662
|
|
|
$detectedType = preg_match('/[[:space:]]*0x([[:alnum:]]*)[[:space:]]+0x([[:alnum:]]*)[[:space:]]+/', $value) ? 'whitespaced' : 'ms-token'; |
|
663
|
|
|
} |
|
664
|
|
|
$hexbyte = ''; |
|
665
|
|
|
$utf8 = ''; |
|
666
|
|
|
if ($detectedType === 'ms-token') { |
|
667
|
|
|
list($hexbyte, $utf8) = preg_split('/[=:]/', $value, 3); |
|
668
|
|
|
} elseif ($detectedType === 'whitespaced') { |
|
669
|
|
|
$regA = []; |
|
670
|
|
|
preg_match('/[[:space:]]*0x([[:alnum:]]*)[[:space:]]+0x([[:alnum:]]*)[[:space:]]+/', $value, $regA); |
|
671
|
|
|
$hexbyte = $regA[1]; |
|
672
|
|
|
$utf8 = 'U+' . $regA[2]; |
|
673
|
|
|
} |
|
674
|
|
|
$decval = hexdec(trim($hexbyte)); |
|
675
|
|
|
if ($decval > 127) { |
|
676
|
|
|
$utf8decval = hexdec(substr(trim($utf8), 2)); |
|
677
|
|
|
$this->parsedCharsets[$charset]['local'][$decval] = $this->UnumberToChar($utf8decval); |
|
|
|
|
|
|
678
|
|
|
$this->parsedCharsets[$charset]['utf8'][$this->parsedCharsets[$charset]['local'][$decval]] = $decval; |
|
679
|
|
|
} |
|
680
|
|
|
} |
|
681
|
|
|
} |
|
682
|
|
|
if ($cacheFile) { |
|
683
|
|
|
GeneralUtility::writeFileToTypo3tempDir($cacheFile, serialize($this->parsedCharsets[$charset])); |
|
684
|
|
|
} |
|
685
|
|
|
} |
|
686
|
|
|
return 2; |
|
687
|
|
|
} |
|
688
|
|
|
throw new UnknownCharsetException(sprintf('Unknown charset "%s"', $charset), 1508916031); |
|
689
|
|
|
} |
|
690
|
|
|
return 1; |
|
691
|
|
|
} |
|
692
|
|
|
|
|
693
|
|
|
/** |
|
694
|
|
|
* This function initializes all UTF-8 character data tables. |
|
695
|
|
|
* |
|
696
|
|
|
* PLEASE SEE: http://www.unicode.org/Public/UNIDATA/ |
|
697
|
|
|
* |
|
698
|
|
|
* @param string $mode Mode ("case", "ascii", ...) |
|
699
|
|
|
* @return int Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached). |
|
700
|
|
|
* @access private |
|
701
|
|
|
*/ |
|
702
|
|
|
public function initUnicodeData($mode = null) |
|
703
|
|
|
{ |
|
704
|
|
|
// Cache files |
|
705
|
|
|
$cacheFileCase = GeneralUtility::getFileAbsFileName('typo3temp/var/charset/cscase_utf-8.tbl'); |
|
706
|
|
|
$cacheFileASCII = GeneralUtility::getFileAbsFileName('typo3temp/var/charset/csascii_utf-8.tbl'); |
|
707
|
|
|
// Only process if the tables are not yet loaded |
|
708
|
|
|
switch ($mode) { |
|
709
|
|
|
case 'case': |
|
710
|
|
|
if (is_array($this->caseFolding['utf-8'])) { |
|
711
|
|
|
return 1; |
|
712
|
|
|
} |
|
713
|
|
|
// Use cached version if possible |
|
714
|
|
|
if ($cacheFileCase && @is_file($cacheFileCase)) { |
|
715
|
|
|
$this->caseFolding['utf-8'] = unserialize(file_get_contents($cacheFileCase)); |
|
716
|
|
|
return 2; |
|
717
|
|
|
} |
|
718
|
|
|
break; |
|
719
|
|
|
case 'ascii': |
|
720
|
|
|
if (is_array($this->toASCII['utf-8'])) { |
|
721
|
|
|
return 1; |
|
722
|
|
|
} |
|
723
|
|
|
// Use cached version if possible |
|
724
|
|
|
if ($cacheFileASCII && @is_file($cacheFileASCII)) { |
|
725
|
|
|
$this->toASCII['utf-8'] = unserialize(file_get_contents($cacheFileASCII)); |
|
726
|
|
|
return 2; |
|
727
|
|
|
} |
|
728
|
|
|
break; |
|
729
|
|
|
} |
|
730
|
|
|
// Process main Unicode data file |
|
731
|
|
|
$unicodeDataFile = ExtensionManagementUtility::extPath('core') . 'Resources/Private/Charsets/unidata/UnicodeData.txt'; |
|
732
|
|
|
if (!(GeneralUtility::validPathStr($unicodeDataFile) && @is_file($unicodeDataFile))) { |
|
733
|
|
|
return false; |
|
|
|
|
|
|
734
|
|
|
} |
|
735
|
|
|
$fh = fopen($unicodeDataFile, 'rb'); |
|
736
|
|
|
if (!$fh) { |
|
737
|
|
|
return false; |
|
|
|
|
|
|
738
|
|
|
} |
|
739
|
|
|
// key = utf8 char (single codepoint), value = utf8 string (codepoint sequence) |
|
740
|
|
|
// Note: we use the UTF-8 characters here and not the Unicode numbers to avoid conversion roundtrip in utf8_strtolower/-upper) |
|
741
|
|
|
$this->caseFolding['utf-8'] = []; |
|
742
|
|
|
$utf8CaseFolding = &$this->caseFolding['utf-8']; |
|
743
|
|
|
// a shorthand |
|
744
|
|
|
$utf8CaseFolding['toUpper'] = []; |
|
745
|
|
|
$utf8CaseFolding['toLower'] = []; |
|
746
|
|
|
$utf8CaseFolding['toTitle'] = []; |
|
747
|
|
|
// Array of temp. decompositions |
|
748
|
|
|
$decomposition = []; |
|
749
|
|
|
// Array of chars that are marks (eg. composing accents) |
|
750
|
|
|
$mark = []; |
|
751
|
|
|
// Array of chars that are numbers (eg. digits) |
|
752
|
|
|
$number = []; |
|
753
|
|
|
// Array of chars to be omitted (eg. Russian hard sign) |
|
754
|
|
|
$omit = []; |
|
755
|
|
|
while (!feof($fh)) { |
|
756
|
|
|
$line = fgets($fh, 4096); |
|
757
|
|
|
// Has a lot of info |
|
758
|
|
|
list($char, $name, $cat, , , $decomp, , , $num, , , , $upper, $lower, $title, ) = explode(';', rtrim($line)); |
|
759
|
|
|
$ord = hexdec($char); |
|
760
|
|
|
if ($ord > 65535) { |
|
761
|
|
|
// Only process the BMP |
|
762
|
|
|
break; |
|
763
|
|
|
} |
|
764
|
|
|
$utf8_char = $this->UnumberToChar($ord); |
|
765
|
|
|
if ($upper) { |
|
766
|
|
|
$utf8CaseFolding['toUpper'][$utf8_char] = $this->UnumberToChar(hexdec($upper)); |
|
767
|
|
|
} |
|
768
|
|
|
if ($lower) { |
|
769
|
|
|
$utf8CaseFolding['toLower'][$utf8_char] = $this->UnumberToChar(hexdec($lower)); |
|
770
|
|
|
} |
|
771
|
|
|
// Store "title" only when different from "upper" (only a few) |
|
772
|
|
|
if ($title && $title !== $upper) { |
|
773
|
|
|
$utf8CaseFolding['toTitle'][$utf8_char] = $this->UnumberToChar(hexdec($title)); |
|
774
|
|
|
} |
|
775
|
|
|
switch ($cat[0]) { |
|
776
|
|
|
case 'M': |
|
777
|
|
|
// mark (accent, umlaut, ...) |
|
778
|
|
|
$mark['U+' . $char] = 1; |
|
779
|
|
|
break; |
|
780
|
|
|
case 'N': |
|
781
|
|
|
// numeric value |
|
782
|
|
|
if ($ord > 128 && $num !== '') { |
|
783
|
|
|
$number['U+' . $char] = $num; |
|
784
|
|
|
} |
|
785
|
|
|
} |
|
786
|
|
|
// Accented Latin letters without "official" decomposition |
|
787
|
|
|
$match = []; |
|
788
|
|
|
if (preg_match('/^LATIN (SMALL|CAPITAL) LETTER ([A-Z]) WITH/', $name, $match) && !$decomp) { |
|
789
|
|
|
$c = ord($match[2]); |
|
790
|
|
|
if ($match[1] === 'SMALL') { |
|
791
|
|
|
$c += 32; |
|
792
|
|
|
} |
|
793
|
|
|
$decomposition['U+' . $char] = [dechex($c)]; |
|
794
|
|
|
continue; |
|
795
|
|
|
} |
|
796
|
|
|
$match = []; |
|
797
|
|
|
if (preg_match('/(<.*>)? *(.+)/', $decomp, $match)) { |
|
798
|
|
|
switch ($match[1]) { |
|
799
|
|
|
case '<circle>': |
|
800
|
|
|
// add parenthesis as circle replacement, eg (1) |
|
801
|
|
|
$match[2] = '0028 ' . $match[2] . ' 0029'; |
|
802
|
|
|
break; |
|
803
|
|
|
case '<square>': |
|
804
|
|
|
// add square brackets as square replacement, eg [1] |
|
805
|
|
|
$match[2] = '005B ' . $match[2] . ' 005D'; |
|
806
|
|
|
break; |
|
807
|
|
|
case '<compat>': |
|
808
|
|
|
// ignore multi char decompositions that start with a space |
|
809
|
|
|
if (preg_match('/^0020 /', $match[2])) { |
|
810
|
|
|
continue 2; |
|
811
|
|
|
} |
|
812
|
|
|
break; |
|
813
|
|
|
case '<initial>': |
|
814
|
|
|
case '<medial>': |
|
815
|
|
|
case '<final>': |
|
816
|
|
|
case '<isolated>': |
|
817
|
|
|
case '<vertical>': |
|
818
|
|
|
continue 2; |
|
819
|
|
|
} |
|
820
|
|
|
$decomposition['U+' . $char] = explode(' ', $match[2]); |
|
821
|
|
|
} |
|
822
|
|
|
} |
|
823
|
|
|
fclose($fh); |
|
824
|
|
|
// Process additional Unicode data for casing (allow folded characters to expand into a sequence) |
|
825
|
|
|
$specialCasingFile = ExtensionManagementUtility::extPath('core') . 'Resources/Private/Charsets/unidata/SpecialCasing.txt'; |
|
826
|
|
|
if (GeneralUtility::validPathStr($specialCasingFile) && @is_file($specialCasingFile)) { |
|
827
|
|
|
$fh = fopen($specialCasingFile, 'rb'); |
|
828
|
|
|
if ($fh) { |
|
829
|
|
|
while (!feof($fh)) { |
|
830
|
|
|
$line = fgets($fh, 4096); |
|
831
|
|
|
if ($line[0] !== '#' && trim($line) !== '') { |
|
832
|
|
|
list($char, $lower, $title, $upper, $cond) = GeneralUtility::trimExplode(';', $line); |
|
833
|
|
|
if ($cond === '' || $cond[0] === '#') { |
|
834
|
|
|
$utf8_char = $this->UnumberToChar(hexdec($char)); |
|
835
|
|
|
if ($char !== $lower) { |
|
836
|
|
|
$arr = explode(' ', $lower); |
|
837
|
|
|
for ($i = 0; isset($arr[$i]); $i++) { |
|
838
|
|
|
$arr[$i] = $this->UnumberToChar(hexdec($arr[$i])); |
|
839
|
|
|
} |
|
840
|
|
|
$utf8CaseFolding['toLower'][$utf8_char] = implode('', $arr); |
|
841
|
|
|
} |
|
842
|
|
|
if ($char !== $title && $title !== $upper) { |
|
843
|
|
|
$arr = explode(' ', $title); |
|
844
|
|
|
for ($i = 0; isset($arr[$i]); $i++) { |
|
845
|
|
|
$arr[$i] = $this->UnumberToChar(hexdec($arr[$i])); |
|
846
|
|
|
} |
|
847
|
|
|
$utf8CaseFolding['toTitle'][$utf8_char] = implode('', $arr); |
|
848
|
|
|
} |
|
849
|
|
|
if ($char !== $upper) { |
|
850
|
|
|
$arr = explode(' ', $upper); |
|
851
|
|
|
for ($i = 0; isset($arr[$i]); $i++) { |
|
852
|
|
|
$arr[$i] = $this->UnumberToChar(hexdec($arr[$i])); |
|
853
|
|
|
} |
|
854
|
|
|
$utf8CaseFolding['toUpper'][$utf8_char] = implode('', $arr); |
|
855
|
|
|
} |
|
856
|
|
|
} |
|
857
|
|
|
} |
|
858
|
|
|
} |
|
859
|
|
|
fclose($fh); |
|
860
|
|
|
} |
|
861
|
|
|
} |
|
862
|
|
|
// Process custom decompositions |
|
863
|
|
|
$customTranslitFile = ExtensionManagementUtility::extPath('core') . 'Resources/Private/Charsets/unidata/Translit.txt'; |
|
864
|
|
|
if (GeneralUtility::validPathStr($customTranslitFile) && @is_file($customTranslitFile)) { |
|
865
|
|
|
$fh = fopen($customTranslitFile, 'rb'); |
|
866
|
|
|
if ($fh) { |
|
867
|
|
|
while (!feof($fh)) { |
|
868
|
|
|
$line = fgets($fh, 4096); |
|
869
|
|
|
if ($line[0] !== '#' && trim($line) !== '') { |
|
870
|
|
|
list($char, $translit) = GeneralUtility::trimExplode(';', $line); |
|
871
|
|
|
if (!$translit) { |
|
872
|
|
|
$omit['U+' . $char] = 1; |
|
873
|
|
|
} |
|
874
|
|
|
$decomposition['U+' . $char] = explode(' ', $translit); |
|
875
|
|
|
} |
|
876
|
|
|
} |
|
877
|
|
|
fclose($fh); |
|
878
|
|
|
} |
|
879
|
|
|
} |
|
880
|
|
|
// Decompose and remove marks; inspired by unac (Loic Dachary <[email protected]>) |
|
881
|
|
|
foreach ($decomposition as $from => $to) { |
|
882
|
|
|
$code_decomp = []; |
|
883
|
|
|
while ($code_value = array_shift($to)) { |
|
884
|
|
|
// Do recursive decomposition |
|
885
|
|
|
if (isset($decomposition['U+' . $code_value])) { |
|
886
|
|
|
foreach (array_reverse($decomposition['U+' . $code_value]) as $cv) { |
|
887
|
|
|
array_unshift($to, $cv); |
|
888
|
|
|
} |
|
889
|
|
|
} elseif (!isset($mark['U+' . $code_value])) { |
|
890
|
|
|
// remove mark |
|
891
|
|
|
$code_decomp[] = $code_value; |
|
892
|
|
|
} |
|
893
|
|
|
} |
|
894
|
|
|
if (!empty($code_decomp) || isset($omit[$from])) { |
|
895
|
|
|
$decomposition[$from] = $code_decomp; |
|
896
|
|
|
} else { |
|
897
|
|
|
unset($decomposition[$from]); |
|
898
|
|
|
} |
|
899
|
|
|
} |
|
900
|
|
|
// Create ascii only mapping |
|
901
|
|
|
$this->toASCII['utf-8'] = []; |
|
902
|
|
|
$ascii = &$this->toASCII['utf-8']; |
|
903
|
|
|
foreach ($decomposition as $from => $to) { |
|
904
|
|
|
$code_decomp = []; |
|
905
|
|
|
while ($code_value = array_shift($to)) { |
|
906
|
|
|
$ord = hexdec($code_value); |
|
907
|
|
|
if ($ord > 127) { |
|
908
|
|
|
continue 2; |
|
909
|
|
|
} |
|
910
|
|
|
// Skip decompositions containing non-ASCII chars |
|
911
|
|
|
$code_decomp[] = chr($ord); |
|
912
|
|
|
} |
|
913
|
|
|
$ascii[$this->UnumberToChar(hexdec($from))] = implode('', $code_decomp); |
|
914
|
|
|
} |
|
915
|
|
|
// Add numeric decompositions |
|
916
|
|
|
foreach ($number as $from => $to) { |
|
917
|
|
|
$utf8_char = $this->UnumberToChar(hexdec($from)); |
|
918
|
|
|
if (!isset($ascii[$utf8_char])) { |
|
919
|
|
|
$ascii[$utf8_char] = $to; |
|
920
|
|
|
} |
|
921
|
|
|
} |
|
922
|
|
|
if ($cacheFileCase) { |
|
923
|
|
|
GeneralUtility::writeFileToTypo3tempDir($cacheFileCase, serialize($utf8CaseFolding)); |
|
924
|
|
|
} |
|
925
|
|
|
if ($cacheFileASCII) { |
|
926
|
|
|
GeneralUtility::writeFileToTypo3tempDir($cacheFileASCII, serialize($ascii)); |
|
927
|
|
|
} |
|
928
|
|
|
return 3; |
|
929
|
|
|
} |
|
930
|
|
|
|
|
931
|
|
|
/** |
|
932
|
|
|
* This function initializes the folding table for a charset other than UTF-8. |
|
933
|
|
|
* This function is automatically called by the case folding functions. |
|
934
|
|
|
* |
|
935
|
|
|
* @param string $charset Charset for which to initialize case folding. |
|
936
|
|
|
* @return int Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached). |
|
937
|
|
|
* @access private |
|
938
|
|
|
*/ |
|
939
|
|
|
public function initCaseFolding($charset) |
|
940
|
|
|
{ |
|
941
|
|
|
// Only process if the case table is not yet loaded: |
|
942
|
|
|
if (is_array($this->caseFolding[$charset])) { |
|
943
|
|
|
return 1; |
|
944
|
|
|
} |
|
945
|
|
|
// Use cached version if possible |
|
946
|
|
|
$cacheFile = GeneralUtility::getFileAbsFileName('typo3temp/var/charset/cscase_' . $charset . '.tbl'); |
|
947
|
|
|
if ($cacheFile && @is_file($cacheFile)) { |
|
948
|
|
|
$this->caseFolding[$charset] = unserialize(file_get_contents($cacheFile)); |
|
949
|
|
|
return 2; |
|
950
|
|
|
} |
|
951
|
|
|
// init UTF-8 conversion for this charset |
|
952
|
|
|
if (!$this->initCharset($charset)) { |
|
953
|
|
|
return false; |
|
|
|
|
|
|
954
|
|
|
} |
|
955
|
|
|
// UTF-8 case folding is used as the base conversion table |
|
956
|
|
|
if (!$this->initUnicodeData('case')) { |
|
957
|
|
|
return false; |
|
|
|
|
|
|
958
|
|
|
} |
|
959
|
|
|
$nochar = chr($this->noCharByteVal); |
|
960
|
|
|
foreach ($this->parsedCharsets[$charset]['local'] as $ci => $utf8) { |
|
961
|
|
|
// Reconvert to charset (don't use chr() of numeric value, might be muli-byte) |
|
962
|
|
|
$c = $this->utf8_decode($utf8, $charset); |
|
963
|
|
|
$cc = $this->utf8_decode($this->caseFolding['utf-8']['toUpper'][$utf8], $charset); |
|
964
|
|
|
if ($cc !== '' && $cc !== $nochar) { |
|
965
|
|
|
$this->caseFolding[$charset]['toUpper'][$c] = $cc; |
|
966
|
|
|
} |
|
967
|
|
|
$cc = $this->utf8_decode($this->caseFolding['utf-8']['toLower'][$utf8], $charset); |
|
968
|
|
|
if ($cc !== '' && $cc !== $nochar) { |
|
969
|
|
|
$this->caseFolding[$charset]['toLower'][$c] = $cc; |
|
970
|
|
|
} |
|
971
|
|
|
$cc = $this->utf8_decode($this->caseFolding['utf-8']['toTitle'][$utf8], $charset); |
|
972
|
|
|
if ($cc !== '' && $cc !== $nochar) { |
|
973
|
|
|
$this->caseFolding[$charset]['toTitle'][$c] = $cc; |
|
974
|
|
|
} |
|
975
|
|
|
} |
|
976
|
|
|
// Add the ASCII case table |
|
977
|
|
|
$start = ord('a'); |
|
978
|
|
|
$end = ord('z'); |
|
979
|
|
|
for ($i = $start; $i <= $end; $i++) { |
|
980
|
|
|
$this->caseFolding[$charset]['toUpper'][chr($i)] = chr($i - 32); |
|
981
|
|
|
} |
|
982
|
|
|
$start = ord('A'); |
|
983
|
|
|
$end = ord('Z'); |
|
984
|
|
|
for ($i = $start; $i <= $end; $i++) { |
|
985
|
|
|
$this->caseFolding[$charset]['toLower'][chr($i)] = chr($i + 32); |
|
986
|
|
|
} |
|
987
|
|
|
if ($cacheFile) { |
|
988
|
|
|
GeneralUtility::writeFileToTypo3tempDir($cacheFile, serialize($this->caseFolding[$charset])); |
|
989
|
|
|
} |
|
990
|
|
|
return 3; |
|
991
|
|
|
} |
|
992
|
|
|
|
|
993
|
|
|
/** |
|
994
|
|
|
* This function initializes the to-ASCII conversion table for a charset other than UTF-8. |
|
995
|
|
|
* This function is automatically called by the ASCII transliteration functions. |
|
996
|
|
|
* |
|
997
|
|
|
* @param string $charset Charset for which to initialize conversion. |
|
998
|
|
|
* @return int Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached). |
|
999
|
|
|
* @access private |
|
1000
|
|
|
*/ |
|
1001
|
|
|
public function initToASCII($charset) |
|
1002
|
|
|
{ |
|
1003
|
|
|
// Only process if the case table is not yet loaded: |
|
1004
|
|
|
if (is_array($this->toASCII[$charset])) { |
|
1005
|
|
|
return 1; |
|
1006
|
|
|
} |
|
1007
|
|
|
// Use cached version if possible |
|
1008
|
|
|
$cacheFile = GeneralUtility::getFileAbsFileName('typo3temp/var/charset/csascii_' . $charset . '.tbl'); |
|
1009
|
|
|
if ($cacheFile && @is_file($cacheFile)) { |
|
1010
|
|
|
$this->toASCII[$charset] = unserialize(file_get_contents($cacheFile)); |
|
1011
|
|
|
return 2; |
|
1012
|
|
|
} |
|
1013
|
|
|
// Init UTF-8 conversion for this charset |
|
1014
|
|
|
if (!$this->initCharset($charset)) { |
|
1015
|
|
|
return false; |
|
|
|
|
|
|
1016
|
|
|
} |
|
1017
|
|
|
// UTF-8/ASCII transliteration is used as the base conversion table |
|
1018
|
|
|
if (!$this->initUnicodeData('ascii')) { |
|
1019
|
|
|
return false; |
|
|
|
|
|
|
1020
|
|
|
} |
|
1021
|
|
|
foreach ($this->parsedCharsets[$charset]['local'] as $ci => $utf8) { |
|
1022
|
|
|
// Reconvert to charset (don't use chr() of numeric value, might be muli-byte) |
|
1023
|
|
|
$c = $this->utf8_decode($utf8, $charset); |
|
1024
|
|
|
if (isset($this->toASCII['utf-8'][$utf8])) { |
|
1025
|
|
|
$this->toASCII[$charset][$c] = $this->toASCII['utf-8'][$utf8]; |
|
1026
|
|
|
} |
|
1027
|
|
|
} |
|
1028
|
|
|
if ($cacheFile) { |
|
1029
|
|
|
GeneralUtility::writeFileToTypo3tempDir($cacheFile, serialize($this->toASCII[$charset])); |
|
1030
|
|
|
} |
|
1031
|
|
|
return 3; |
|
1032
|
|
|
} |
|
1033
|
|
|
|
|
1034
|
|
|
/******************************************** |
|
1035
|
|
|
* |
|
1036
|
|
|
* String operation functions |
|
1037
|
|
|
* |
|
1038
|
|
|
********************************************/ |
|
1039
|
|
|
|
|
1040
|
|
|
/** |
|
1041
|
|
|
* Truncates a string and pre-/appends a string. |
|
1042
|
|
|
* Unit tested by Kasper |
|
1043
|
|
|
* |
|
1044
|
|
|
* @param string $charset The character set |
|
1045
|
|
|
* @param string $string Character string |
|
1046
|
|
|
* @param int $len Length (in characters) |
|
1047
|
|
|
* @param string $crop Crop signifier |
|
1048
|
|
|
* @return string The shortened string |
|
1049
|
|
|
* @see substr(), mb_strimwidth() |
|
1050
|
|
|
*/ |
|
1051
|
|
|
public function crop($charset, $string, $len, $crop = '') |
|
1052
|
|
|
{ |
|
1053
|
|
|
if ((int)$len === 0 || mb_strlen($string, $charset) <= abs($len)) { |
|
1054
|
|
|
return $string; |
|
1055
|
|
|
} |
|
1056
|
|
|
if ($len > 0) { |
|
1057
|
|
|
$string = mb_substr($string, 0, $len, $charset) . $crop; |
|
1058
|
|
|
} else { |
|
1059
|
|
|
$string = $crop . mb_substr($string, $len, mb_strlen($string, $charset), $charset); |
|
1060
|
|
|
} |
|
1061
|
|
|
return $string; |
|
1062
|
|
|
} |
|
1063
|
|
|
|
|
1064
|
|
|
/** |
|
1065
|
|
|
* Equivalent of lcfirst/ucfirst but using character set. |
|
1066
|
|
|
* |
|
1067
|
|
|
* @param string $charset |
|
1068
|
|
|
* @param string $string |
|
1069
|
|
|
* @param string $case can be 'toLower' or 'toUpper' |
|
1070
|
|
|
* @return string |
|
1071
|
|
|
*/ |
|
1072
|
|
|
public function convCaseFirst($charset, $string, $case) |
|
1073
|
|
|
{ |
|
1074
|
|
|
$firstChar = mb_substr($string, 0, 1, $charset); |
|
1075
|
|
|
$firstChar = $case === 'toLower' |
|
1076
|
|
|
? mb_strtolower($firstChar, $charset) |
|
1077
|
|
|
: mb_strtoupper($firstChar, $charset); |
|
1078
|
|
|
$remainder = mb_substr($string, 1, null, $charset); |
|
1079
|
|
|
return $firstChar . $remainder; |
|
1080
|
|
|
} |
|
1081
|
|
|
|
|
1082
|
|
|
/** |
|
1083
|
|
|
* Converts special chars (like æøåÆØÅ, umlauts etc) to ascii equivalents (usually double-bytes, like æ => ae etc.) |
|
1084
|
|
|
* |
|
1085
|
|
|
* @param string $charset Character set of string |
|
1086
|
|
|
* @param string $string Input string to convert |
|
1087
|
|
|
* @return string The converted string |
|
1088
|
|
|
*/ |
|
1089
|
|
|
public function specCharsToASCII($charset, $string) |
|
1090
|
|
|
{ |
|
1091
|
|
|
if ($charset === 'utf-8') { |
|
1092
|
|
|
$string = $this->utf8_char_mapping($string); |
|
1093
|
|
|
} elseif (isset($this->eucBasedSets[$charset])) { |
|
1094
|
|
|
$string = $this->euc_char_mapping($string, $charset); |
|
1095
|
|
|
} else { |
|
1096
|
|
|
// Treat everything else as single-byte encoding |
|
1097
|
|
|
$string = $this->sb_char_mapping($string, $charset); |
|
1098
|
|
|
} |
|
1099
|
|
|
return $string; |
|
1100
|
|
|
} |
|
1101
|
|
|
|
|
1102
|
|
|
/******************************************** |
|
1103
|
|
|
* |
|
1104
|
|
|
* Internal string operation functions |
|
1105
|
|
|
* |
|
1106
|
|
|
********************************************/ |
|
1107
|
|
|
/** |
|
1108
|
|
|
* Maps all characters of a string in a single byte charset. |
|
1109
|
|
|
* |
|
1110
|
|
|
* @param string $str The string |
|
1111
|
|
|
* @param string $charset The charset |
|
1112
|
|
|
* @return string The converted string |
|
1113
|
|
|
*/ |
|
1114
|
|
|
public function sb_char_mapping($str, $charset) |
|
1115
|
|
|
{ |
|
1116
|
|
|
if (!$this->initToASCII($charset)) { |
|
1117
|
|
|
return $str; |
|
1118
|
|
|
} |
|
1119
|
|
|
// Do nothing |
|
1120
|
|
|
$map = &$this->toASCII[$charset]; |
|
1121
|
|
|
$out = ''; |
|
1122
|
|
|
for ($i = 0; isset($str[$i]); $i++) { |
|
1123
|
|
|
$c = $str[$i]; |
|
1124
|
|
|
if (isset($map[$c])) { |
|
1125
|
|
|
$out .= $map[$c]; |
|
1126
|
|
|
} else { |
|
1127
|
|
|
$out .= $c; |
|
1128
|
|
|
} |
|
1129
|
|
|
} |
|
1130
|
|
|
return $out; |
|
1131
|
|
|
} |
|
1132
|
|
|
|
|
1133
|
|
|
/******************************************** |
|
1134
|
|
|
* |
|
1135
|
|
|
* Internal UTF-8 string operation functions |
|
1136
|
|
|
* |
|
1137
|
|
|
********************************************/ |
|
1138
|
|
|
|
|
1139
|
|
|
/** |
|
1140
|
|
|
* Translates a character position into an 'absolute' byte position. |
|
1141
|
|
|
* Unit tested by Kasper. |
|
1142
|
|
|
* |
|
1143
|
|
|
* @param string $str UTF-8 string |
|
1144
|
|
|
* @param int $pos Character position (negative values start from the end) |
|
1145
|
|
|
* @return int Byte position |
|
1146
|
|
|
*/ |
|
1147
|
|
|
public function utf8_char2byte_pos($str, $pos) |
|
1148
|
|
|
{ |
|
1149
|
|
|
// Number of characters found |
|
1150
|
|
|
$n = 0; |
|
1151
|
|
|
// Number of characters wanted |
|
1152
|
|
|
$p = abs($pos); |
|
1153
|
|
|
if ($pos >= 0) { |
|
1154
|
|
|
$i = 0; |
|
1155
|
|
|
$d = 1; |
|
1156
|
|
|
} else { |
|
1157
|
|
|
$i = strlen($str) - 1; |
|
1158
|
|
|
$d = -1; |
|
1159
|
|
|
} |
|
1160
|
|
|
for (; isset($str[$i]) && $n < $p; $i += $d) { |
|
1161
|
|
|
$c = (int)ord($str[$i]); |
|
1162
|
|
|
// single-byte (0xxxxxx) |
|
1163
|
|
|
if (!($c & 128)) { |
|
1164
|
|
|
$n++; |
|
1165
|
|
|
} elseif (($c & 192) === 192) { |
|
1166
|
|
|
// Multi-byte starting byte (11xxxxxx) |
|
1167
|
|
|
$n++; |
|
1168
|
|
|
} |
|
1169
|
|
|
} |
|
1170
|
|
|
if (!isset($str[$i])) { |
|
1171
|
|
|
// Offset beyond string length |
|
1172
|
|
|
return false; |
|
|
|
|
|
|
1173
|
|
|
} |
|
1174
|
|
|
if ($pos >= 0) { |
|
1175
|
|
|
// Skip trailing multi-byte data bytes |
|
1176
|
|
|
while (ord($str[$i]) & 128 && !(ord($str[$i]) & 64)) { |
|
1177
|
|
|
$i++; |
|
1178
|
|
|
} |
|
1179
|
|
|
} else { |
|
1180
|
|
|
// Correct offset |
|
1181
|
|
|
$i++; |
|
1182
|
|
|
} |
|
1183
|
|
|
return $i; |
|
1184
|
|
|
} |
|
1185
|
|
|
|
|
1186
|
|
|
/** |
|
1187
|
|
|
* Maps all characters of an UTF-8 string. |
|
1188
|
|
|
* |
|
1189
|
|
|
* @param string $str UTF-8 string |
|
1190
|
|
|
* @return string The converted string |
|
1191
|
|
|
*/ |
|
1192
|
|
|
public function utf8_char_mapping($str) |
|
1193
|
|
|
{ |
|
1194
|
|
|
if (!$this->initUnicodeData('ascii')) { |
|
1195
|
|
|
// Do nothing |
|
1196
|
|
|
return $str; |
|
1197
|
|
|
} |
|
1198
|
|
|
$out = ''; |
|
1199
|
|
|
$map = &$this->toASCII['utf-8']; |
|
1200
|
|
|
for ($i = 0; isset($str[$i]); $i++) { |
|
1201
|
|
|
$c = ord($str[$i]); |
|
1202
|
|
|
$mbc = ''; |
|
1203
|
|
|
// single-byte (0xxxxxx) |
|
1204
|
|
|
if (!($c & 128)) { |
|
1205
|
|
|
$mbc = $str[$i]; |
|
1206
|
|
|
} elseif (($c & 192) === 192) { |
|
1207
|
|
|
$bc = 0; |
|
1208
|
|
|
// multi-byte starting byte (11xxxxxx) |
|
1209
|
|
|
for (; $c & 128; $c = $c << 1) { |
|
1210
|
|
|
$bc++; |
|
1211
|
|
|
} |
|
1212
|
|
|
// calculate number of bytes |
|
1213
|
|
|
$mbc = substr($str, $i, $bc); |
|
1214
|
|
|
$i += $bc - 1; |
|
1215
|
|
|
} |
|
1216
|
|
|
if (isset($map[$mbc])) { |
|
1217
|
|
|
$out .= $map[$mbc]; |
|
1218
|
|
|
} else { |
|
1219
|
|
|
$out .= $mbc; |
|
1220
|
|
|
} |
|
1221
|
|
|
} |
|
1222
|
|
|
return $out; |
|
1223
|
|
|
} |
|
1224
|
|
|
|
|
1225
|
|
|
/******************************************** |
|
1226
|
|
|
* |
|
1227
|
|
|
* Internal EUC string operation functions |
|
1228
|
|
|
* |
|
1229
|
|
|
* Extended Unix Code: |
|
1230
|
|
|
* ASCII compatible 7bit single bytes chars |
|
1231
|
|
|
* 8bit two byte chars |
|
1232
|
|
|
* |
|
1233
|
|
|
* Shift-JIS is treated as a special case. |
|
1234
|
|
|
* |
|
1235
|
|
|
********************************************/ |
|
1236
|
|
|
|
|
1237
|
|
|
/** |
|
1238
|
|
|
* Maps all characters of a string in the EUC charset family. |
|
1239
|
|
|
* |
|
1240
|
|
|
* @param string $str EUC multibyte character string |
|
1241
|
|
|
* @param string $charset The charset |
|
1242
|
|
|
* @return string The converted string |
|
1243
|
|
|
*/ |
|
1244
|
|
|
public function euc_char_mapping($str, $charset) |
|
1245
|
|
|
{ |
|
1246
|
|
|
if (!$this->initToASCII($charset)) { |
|
1247
|
|
|
return $str; |
|
1248
|
|
|
} |
|
1249
|
|
|
// do nothing |
|
1250
|
|
|
$map = &$this->toASCII[$charset]; |
|
1251
|
|
|
$out = ''; |
|
1252
|
|
|
for ($i = 0; isset($str[$i]); $i++) { |
|
1253
|
|
|
$mbc = $str[$i]; |
|
1254
|
|
|
$c = ord($mbc); |
|
1255
|
|
|
if ($charset === 'shift_jis') { |
|
1256
|
|
|
// A double-byte char |
|
1257
|
|
|
if ($c >= 128 && $c < 160 || $c >= 224) { |
|
1258
|
|
|
$mbc = substr($str, $i, 2); |
|
1259
|
|
|
$i++; |
|
1260
|
|
|
} |
|
1261
|
|
|
} else { |
|
1262
|
|
|
// A double-byte char |
|
1263
|
|
|
if ($c >= 128) { |
|
1264
|
|
|
$mbc = substr($str, $i, 2); |
|
1265
|
|
|
$i++; |
|
1266
|
|
|
} |
|
1267
|
|
|
} |
|
1268
|
|
|
if (isset($map[$mbc])) { |
|
1269
|
|
|
$out .= $map[$mbc]; |
|
1270
|
|
|
} else { |
|
1271
|
|
|
$out .= $mbc; |
|
1272
|
|
|
} |
|
1273
|
|
|
} |
|
1274
|
|
|
return $out; |
|
1275
|
|
|
} |
|
1276
|
|
|
} |
|
1277
|
|
|
|