|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* This file is part of the ZBateson\MbWrapper project. |
|
4
|
|
|
* |
|
5
|
|
|
* @license http://opensource.org/licenses/bsd-license.php BSD |
|
6
|
|
|
*/ |
|
7
|
|
|
namespace ZBateson\MbWrapper; |
|
8
|
|
|
|
|
9
|
|
|
/** |
|
10
|
|
|
* Helper class for converting strings between charsets, finding a multibyte |
|
11
|
|
|
* strings length, and creating a substring. |
|
12
|
|
|
* |
|
13
|
|
|
* MbWrapper prefers PHP's mb_* extension first, and reverts to iconv_* if the |
|
14
|
|
|
* charsets aren't listed as supported by mb_list_encodings(). |
|
15
|
|
|
* |
|
16
|
|
|
* A list of aliased charsets are maintained to support the greatest number of |
|
17
|
|
|
* charsets. In addition, when searching for a charset, separator characters |
|
18
|
|
|
* such as dashes are removed, and searches are always performed |
|
19
|
|
|
* case-insensitively. This is to support strange reported encodings in emails, |
|
20
|
|
|
* etc... |
|
21
|
|
|
* |
|
22
|
|
|
* @author Zaahid Bateson |
|
23
|
|
|
*/ |
|
24
|
|
|
class MbWrapper |
|
25
|
|
|
{ |
|
26
|
|
|
/** |
|
27
|
|
|
* @var array<string, string> aliased charsets supported by mb_convert_encoding. |
|
28
|
|
|
* The alias is stripped of any non-alphanumeric characters (so CP367 |
|
29
|
|
|
* is equal to CP-367) when comparing. |
|
30
|
|
|
* Some of these translations are already supported by |
|
31
|
|
|
* mb_convert_encoding on "my" PHP 5.5.9, but may not be supported in |
|
32
|
|
|
* other implementations or versions since they're not part of |
|
33
|
|
|
* documented support. |
|
34
|
|
|
*/ |
|
35
|
|
|
public static $mbAliases = [ |
|
36
|
|
|
// supported but not included in mb_list_encodings for some reason... |
|
37
|
|
|
'CP850' => 'CP850', |
|
38
|
|
|
'GB2312' => 'GB18030', |
|
39
|
|
|
'SJIS2004' => 'SJIS-2004', |
|
40
|
|
|
// aliases |
|
41
|
|
|
'ANSIX341968' => 'ASCII', |
|
42
|
|
|
'ANSIX341986' => 'ASCII', |
|
43
|
|
|
'ARABIC' => 'ISO-8859-6', |
|
44
|
|
|
'ASMO708' => 'ISO-8859-6', |
|
45
|
|
|
'BIG5' => 'BIG-5', |
|
46
|
|
|
'BIG5TW' => 'BIG-5', |
|
47
|
|
|
'CESU8' => 'UTF-8', |
|
48
|
|
|
'CHINESE' => 'GB18030', |
|
49
|
|
|
'CP367' => 'ASCII', |
|
50
|
|
|
'CP819' => 'ISO-8859-1', |
|
51
|
|
|
'CP1251' => 'WINDOWS-1251', |
|
52
|
|
|
'CP1252' => 'WINDOWS-1252', |
|
53
|
|
|
'CP1254' => 'WINDOWS-1254', |
|
54
|
|
|
'CP1255' => 'ISO-8859-8', |
|
55
|
|
|
'CSASCII' => 'ASCII', |
|
56
|
|
|
'CSBIG5' => 'BIG-5', |
|
57
|
|
|
'CSIBM866' => 'CP866', |
|
58
|
|
|
'CSISO2022JP' => 'ISO-2022-JP', |
|
59
|
|
|
'CSISO2022KR' => 'ISO-2022-KR', |
|
60
|
|
|
'CSISO58GB231280' => 'GB18030', |
|
61
|
|
|
'CSISOLATIN1' => 'ISO-8859-1', |
|
62
|
|
|
'CSISOLATIN2' => 'ISO-8859-2', |
|
63
|
|
|
'CSISOLATIN3' => 'ISO-8859-3', |
|
64
|
|
|
'CSISOLATIN4' => 'ISO-8859-4', |
|
65
|
|
|
'CSISOLATIN5' => 'ISO-8859-9', |
|
66
|
|
|
'CSISOLATIN6' => 'ISO-8859-10', |
|
67
|
|
|
'CSISOLATINARABIC' => 'ISO-8859-6', |
|
68
|
|
|
'CSISOLATINCYRILLIC' => 'ISO-8859-5', |
|
69
|
|
|
'CSISOLATINGREEK' => 'ISO-8859-7', |
|
70
|
|
|
'CSISOLATINHEBREW' => 'ISO-8859-8', |
|
71
|
|
|
'CSKOI8R' => 'KOI8-R', |
|
72
|
|
|
'CSPC850MULTILINGUAL' => 'CP850', |
|
73
|
|
|
'CSSHIFTJIS' => 'SJIS', |
|
74
|
|
|
'CYRILLIC' => 'ISO-8859-5', |
|
75
|
|
|
'ECMA114' => 'ISO-8859-6', |
|
76
|
|
|
'ECMA118' => 'ISO-8859-7', |
|
77
|
|
|
'ELOT928' => 'ISO-8859-7', |
|
78
|
|
|
'EUCCN' => 'GB18030', |
|
79
|
|
|
'EUCGB2312CN' => 'GB18030', |
|
80
|
|
|
'GB180302000' => 'GB18030', |
|
81
|
|
|
'GB23121980' => 'GB18030', |
|
82
|
|
|
'GB231280' => 'GB18030', |
|
83
|
|
|
'GBK' => 'CP936', |
|
84
|
|
|
'GREEK8' => 'ISO-8859-7', |
|
85
|
|
|
'GREEK' => 'ISO-8859-7', |
|
86
|
|
|
'HEBREW' => 'ISO-8859-8', |
|
87
|
|
|
'HZGB2312' => 'HZ', |
|
88
|
|
|
'HZGB' => 'HZ', |
|
89
|
|
|
'IBM367' => 'ASCII', |
|
90
|
|
|
'IBM819' => 'ISO-8859-1', |
|
91
|
|
|
'IBM850' => 'CP850', |
|
92
|
|
|
'IBM866' => 'CP866', |
|
93
|
|
|
'ISO2022JP2004' => 'ISO-2022-JP-2004', |
|
94
|
|
|
'ISO646IRV1991' => 'ASCII', |
|
95
|
|
|
'ISO646US' => 'ASCII', |
|
96
|
|
|
'ISO8859' => 'ISO-8859-1', |
|
97
|
|
|
'ISO8859101992' => 'ISO-8859-10', |
|
98
|
|
|
'ISO885911987' => 'ISO-8859-1', |
|
99
|
|
|
'ISO8859141998' => 'ISO-8859-14', |
|
100
|
|
|
'ISO8859162001' => 'ISO-8859-16', |
|
101
|
|
|
'ISO885921987' => 'ISO-8859-2', |
|
102
|
|
|
'ISO885931988' => 'ISO-8859-3', |
|
103
|
|
|
'ISO885941988' => 'ISO-8859-4', |
|
104
|
|
|
'ISO885951988' => 'ISO-8859-5', |
|
105
|
|
|
'ISO885961987' => 'ISO-8859-6', |
|
106
|
|
|
'ISO885971987' => 'ISO-8859-7', |
|
107
|
|
|
'ISO885981988' => 'ISO-8859-8', |
|
108
|
|
|
'ISO88598I' => 'ISO-8859-8', |
|
109
|
|
|
'ISO885991989' => 'ISO-8859-9', |
|
110
|
|
|
'ISOCELTIC' => 'ISO-8859-14', |
|
111
|
|
|
'ISOIR100' => 'ISO-8859-1', |
|
112
|
|
|
'ISOIR101' => 'ISO-8859-2', |
|
113
|
|
|
'ISOIR109' => 'ISO-8859-3', |
|
114
|
|
|
'ISOIR110' => 'ISO-8859-4', |
|
115
|
|
|
'ISOIR126' => 'ISO-8859-7', |
|
116
|
|
|
'ISOIR127' => 'ISO-8859-6', |
|
117
|
|
|
'ISOIR138' => 'ISO-8859-8', |
|
118
|
|
|
'ISOIR144' => 'ISO-8859-5', |
|
119
|
|
|
'ISOIR148' => 'ISO-8859-9', |
|
120
|
|
|
'ISOIR157' => 'ISO-8859-10', |
|
121
|
|
|
'ISOIR199' => 'ISO-8859-14', |
|
122
|
|
|
'ISOIR226' => 'ISO-8859-16', |
|
123
|
|
|
'ISOIR58' => 'GB18030', |
|
124
|
|
|
'ISOIR6' => 'ASCII', |
|
125
|
|
|
'KOI8R' => 'KOI8-R', |
|
126
|
|
|
'KOREAN' => 'EUC-KR', |
|
127
|
|
|
'KSC56011987' => 'EUC-KR', |
|
128
|
|
|
'KSC5601' => 'EUC-KR', |
|
129
|
|
|
'KSX1001' => 'EUC-KR', |
|
130
|
|
|
'L1' => 'ISO-8859-1', |
|
131
|
|
|
'L2' => 'ISO-8859-2', |
|
132
|
|
|
'L3' => 'ISO-8859-3', |
|
133
|
|
|
'L4' => 'ISO-8859-4', |
|
134
|
|
|
'L5' => 'ISO-8859-9', |
|
135
|
|
|
'L6' => 'ISO-8859-10', |
|
136
|
|
|
'L8' => 'ISO-8859-14', |
|
137
|
|
|
'L10' => 'ISO-8859-16', |
|
138
|
|
|
'LATIN' => 'ISO-8859-1', |
|
139
|
|
|
'LATIN1' => 'ISO-8859-1', |
|
140
|
|
|
'LATIN2' => 'ISO-8859-2', |
|
141
|
|
|
'LATIN3' => 'ISO-8859-3', |
|
142
|
|
|
'LATIN4' => 'ISO-8859-4', |
|
143
|
|
|
'LATIN5' => 'ISO-8859-9', |
|
144
|
|
|
'LATIN6' => 'ISO-8859-10', |
|
145
|
|
|
'LATIN8' => 'ISO-8859-14', |
|
146
|
|
|
'LATIN10' => 'ISO-8859-16', |
|
147
|
|
|
'MS932' => 'CP932', |
|
148
|
|
|
'ms936' => 'CP936', |
|
149
|
|
|
'MS950' => 'CP950', |
|
150
|
|
|
'MSKANJI' => 'CP932', |
|
151
|
|
|
'SHIFTJIS2004' => 'SJIS', |
|
152
|
|
|
'SHIFTJIS' => 'SJIS', |
|
153
|
|
|
'UJIS' => 'EUC-JP', |
|
154
|
|
|
'UNICODE11UTF7' => 'UTF-7', |
|
155
|
|
|
'US' => 'ASCII', |
|
156
|
|
|
'USASCII' => 'ASCII', |
|
157
|
|
|
'WE8MSWIN1252' => 'WINDOWS-1252', |
|
158
|
|
|
'WINDOWS1251' => 'WINDOWS-1251', |
|
159
|
|
|
'WINDOWS1252' => 'WINDOWS-1252', |
|
160
|
|
|
'WINDOWS1254' => 'WINDOWS-1254', |
|
161
|
|
|
'WINDOWS1255' => 'ISO-8859-8', |
|
162
|
|
|
'0' => 'WINDOWS-1252', |
|
163
|
|
|
'128' => 'SJIS', |
|
164
|
|
|
'129' => 'EUC-KR', |
|
165
|
|
|
'134' => 'GB18030', |
|
166
|
|
|
'136' => 'BIG-5', |
|
167
|
|
|
'161' => 'WINDOWS-1253', |
|
168
|
|
|
'162' => 'WINDOWS-1254', |
|
169
|
|
|
'177' => 'WINDOWS-1255', |
|
170
|
|
|
'178' => 'WINDOWS-1256', |
|
171
|
|
|
'186' => 'WINDOWS-1257', |
|
172
|
|
|
'204' => 'WINDOWS-1251', |
|
173
|
|
|
'222' => 'WINDOWS-874', |
|
174
|
|
|
'238' => 'WINDOWS-1250', |
|
175
|
|
|
'646' => 'ASCII', |
|
176
|
|
|
'850' => 'CP850', |
|
177
|
|
|
'866' => 'CP866', |
|
178
|
|
|
'932' => 'CP932', |
|
179
|
|
|
'936' => 'CP936', |
|
180
|
|
|
'950' => 'CP950', |
|
181
|
|
|
'1251' => 'WINDOWS-1251', |
|
182
|
|
|
'1252' => 'WINDOWS-1252', |
|
183
|
|
|
'1254' => 'WINDOWS-1254', |
|
184
|
|
|
'1255' => 'ISO-8859-8', |
|
185
|
|
|
'8859' => 'ISO-8859-1', |
|
186
|
|
|
]; |
|
187
|
|
|
|
|
188
|
|
|
/** |
|
189
|
|
|
* @var array<string, string> aliased charsets supported by iconv. |
|
190
|
|
|
*/ |
|
191
|
|
|
public static $iconvAliases = [ |
|
192
|
|
|
// iconv aliases -- a lot of these may already be supported |
|
193
|
|
|
'CESU8' => 'UTF8', |
|
194
|
|
|
'CP154' => 'PT154', |
|
195
|
|
|
'CPGR' => 'CP869', |
|
196
|
|
|
'CPIS' => 'CP861', |
|
197
|
|
|
'CSHPROMAN8' => 'ROMAN8', |
|
198
|
|
|
'CSIBM037' => 'CP037', |
|
199
|
|
|
'CSIBM1026' => 'CP1026', |
|
200
|
|
|
'CSIBM424' => 'CP424', |
|
201
|
|
|
'CSIBM500' => 'CP500', |
|
202
|
|
|
'CSIBM860' => 'CP860', |
|
203
|
|
|
'CSIBM861' => 'CP861', |
|
204
|
|
|
'CSIBM863' => 'CP863', |
|
205
|
|
|
'CSIBM864' => 'CP864', |
|
206
|
|
|
'CSIBM865' => 'CP865', |
|
207
|
|
|
'CSIBM869' => 'CP869', |
|
208
|
|
|
'CSPC775BALTIC' => 'CP775', |
|
209
|
|
|
'CSPC862LATINHEBREW' => 'CP862', |
|
210
|
|
|
'CSPC8CODEPAGE437' => 'CP437', |
|
211
|
|
|
'CSPTCP154' => 'PT154', |
|
212
|
|
|
'CYRILLICASIAN' => 'PT154', |
|
213
|
|
|
'EBCDICCPBE' => 'CP500', |
|
214
|
|
|
'EBCDICCPCA' => 'CP037', |
|
215
|
|
|
'EBCDICCPCH' => 'CP500', |
|
216
|
|
|
'EBCDICCPHE' => 'CP424', |
|
217
|
|
|
'EBCDICCPNL' => 'CP037', |
|
218
|
|
|
'EBCDICCPUS' => 'CP037', |
|
219
|
|
|
'EBCDICCPWT' => 'CP037', |
|
220
|
|
|
'HKSCS' => 'BIG5HKSCS', |
|
221
|
|
|
'HPROMAN8' => 'ROMAN8', |
|
222
|
|
|
'IBM037' => 'CP037', |
|
223
|
|
|
'IBM039' => 'CP037', |
|
224
|
|
|
'IBM424' => 'CP424', |
|
225
|
|
|
'IBM437' => 'CP437', |
|
226
|
|
|
'IBM500' => 'CP500', |
|
227
|
|
|
'IBM775' => 'CP775', |
|
228
|
|
|
'IBM860' => 'CP860', |
|
229
|
|
|
'IBM861' => 'CP861', |
|
230
|
|
|
'IBM862' => 'CP862', |
|
231
|
|
|
'IBM863' => 'CP863', |
|
232
|
|
|
'IBM864' => 'CP864', |
|
233
|
|
|
'IBM865' => 'CP865', |
|
234
|
|
|
'IBM869' => 'CP869', |
|
235
|
|
|
'IBM1026' => 'CP1026', |
|
236
|
|
|
'IBM1140' => 'CP1140', |
|
237
|
|
|
'ISO2022JP2' => 'ISO2022JP2', |
|
238
|
|
|
'ISO8859112001' => 'ISO885911', |
|
239
|
|
|
'ISO885911' => 'ISO885911', |
|
240
|
|
|
'ISOIR166' => 'TIS620', |
|
241
|
|
|
'JOHAB' => 'CP1361', |
|
242
|
|
|
'MACCYRILLIC' => 'MACCYRILLIC', |
|
243
|
|
|
'MS1361' => 'CP1361', |
|
244
|
|
|
'MS949' => 'CP949', |
|
245
|
|
|
'PTCP154' => 'PT154', |
|
246
|
|
|
'R8' => 'ROMAN8', |
|
247
|
|
|
'ROMAN8' => 'ROMAN8', |
|
248
|
|
|
'THAI' => 'ISO885911', |
|
249
|
|
|
'TIS6200' => 'TIS620', |
|
250
|
|
|
'TIS62025290' => 'TIS620', |
|
251
|
|
|
'TIS62025291' => 'TIS620', |
|
252
|
|
|
'TIS620' => 'TIS620', |
|
253
|
|
|
'UHC' => 'CP949', |
|
254
|
|
|
'WINDOWS1250' => 'CP1250', |
|
255
|
|
|
'WINDOWS1253' => 'CP1253', |
|
256
|
|
|
'WINDOWS1256' => 'CP1256', |
|
257
|
|
|
'WINDOWS1257' => 'CP1257', |
|
258
|
|
|
'WINDOWS1258' => 'CP1258', |
|
259
|
|
|
'037' => 'CP037', |
|
260
|
|
|
'424' => 'CP424', |
|
261
|
|
|
'437' => 'CP437', |
|
262
|
|
|
'500' => 'CP500', |
|
263
|
|
|
'775' => 'CP775', |
|
264
|
|
|
'860' => 'CP860', |
|
265
|
|
|
'861' => 'CP861', |
|
266
|
|
|
'862' => 'CP862', |
|
267
|
|
|
'863' => 'CP863', |
|
268
|
|
|
'864' => 'CP864', |
|
269
|
|
|
'865' => 'CP865', |
|
270
|
|
|
'869' => 'CP869', |
|
271
|
|
|
'949' => 'CP949', |
|
272
|
|
|
'1026' => 'CP1026', |
|
273
|
|
|
'1140' => 'CP1140', |
|
274
|
|
|
'1250' => 'CP1250', |
|
275
|
|
|
'1253' => 'CP1253', |
|
276
|
|
|
'1256' => 'CP1256', |
|
277
|
|
|
'1257' => 'CP1257', |
|
278
|
|
|
'1258' => 'CP1258', |
|
279
|
|
|
]; |
|
280
|
|
|
|
|
281
|
|
|
/** |
|
282
|
|
|
* @var string[] cached lookups for quicker retrieval |
|
283
|
|
|
*/ |
|
284
|
|
|
protected $mappedMbCharsets = [ |
|
285
|
|
|
'UTF8' => 'UTF-8', |
|
286
|
|
|
'USASCII' => 'US-ASCII', |
|
287
|
|
|
'ISO88591' => 'ISO-8859-1', |
|
288
|
|
|
]; |
|
289
|
|
|
|
|
290
|
|
|
/** |
|
291
|
|
|
* @var string[] An array of encodings supported by the mb_* extension, as |
|
292
|
|
|
* returned by mb_list_encodings(), with the key set to the charset's |
|
293
|
|
|
* name afte |
|
294
|
|
|
*/ |
|
295
|
|
|
private static $mbListedEncodings; |
|
296
|
|
|
|
|
297
|
|
|
/** |
|
298
|
|
|
* Initializes the static mb_* encoding array. |
|
299
|
|
|
*/ |
|
300
|
14 |
|
public function __construct() |
|
301
|
|
|
{ |
|
302
|
14 |
|
if (self::$mbListedEncodings === null) { |
|
|
|
|
|
|
303
|
1 |
|
$cs = \mb_list_encodings(); |
|
304
|
1 |
|
$keys = $this->getNormalizedCharset($cs); |
|
305
|
1 |
|
self::$mbListedEncodings = \array_combine($keys, $cs); |
|
306
|
|
|
} |
|
307
|
|
|
} |
|
308
|
|
|
|
|
309
|
|
|
/** |
|
310
|
|
|
* The passed charset is uppercased, and stripped of non-alphanumeric |
|
311
|
|
|
* characters before being returned. |
|
312
|
|
|
* |
|
313
|
|
|
* @param string|string[] $charset |
|
314
|
|
|
* @return string|string[] |
|
315
|
|
|
*/ |
|
316
|
14 |
|
private function getNormalizedCharset($charset) |
|
317
|
|
|
{ |
|
318
|
14 |
|
$upper = null; |
|
319
|
14 |
|
if (\is_array($charset)) { |
|
320
|
1 |
|
$upper = \array_map('strtoupper', $charset); |
|
321
|
|
|
} else { |
|
322
|
14 |
|
$upper = \strtoupper($charset); |
|
323
|
|
|
} |
|
324
|
14 |
|
return \preg_replace('/[^A-Z0-9]+/', '', $upper); |
|
325
|
|
|
} |
|
326
|
|
|
|
|
327
|
9 |
|
private function iconv(string $fromCharset, string $toCharset, string $str) : string |
|
328
|
|
|
{ |
|
329
|
9 |
|
$ret = @\iconv($fromCharset, $toCharset . '//TRANSLIT//IGNORE', $str); |
|
330
|
9 |
|
if ($ret === false) { |
|
331
|
1 |
|
throw new UnsupportedCharsetException("Unable to convert from charsets: $fromCharset to $toCharset"); |
|
332
|
|
|
} |
|
333
|
8 |
|
return $ret; |
|
334
|
|
|
} |
|
335
|
|
|
|
|
336
|
4 |
|
private function iconvStrlen(string $str, string $charset) : int |
|
337
|
|
|
{ |
|
338
|
4 |
|
$ret = @\iconv_strlen($str, $charset . '//TRANSLIT//IGNORE'); |
|
339
|
4 |
|
if ($ret === false) { |
|
340
|
2 |
|
throw new UnsupportedCharsetException("Charset $charset is not supported"); |
|
341
|
|
|
} |
|
342
|
2 |
|
return $ret; |
|
343
|
|
|
} |
|
344
|
|
|
|
|
345
|
4 |
|
private function iconvSubstr(string $str, string $charset, int $start, ?int $length = null) : string |
|
346
|
|
|
{ |
|
347
|
4 |
|
$ret = @\iconv_substr($str, $start, $length, $charset . '//TRANSLIT//IGNORE'); |
|
348
|
4 |
|
if ($ret === false) { |
|
349
|
1 |
|
$strLength = $this->iconvStrlen($str, $charset); |
|
350
|
|
|
if ($start > $strLength) { |
|
351
|
|
|
// returns empty to keep in line with mb_substr functionality |
|
352
|
|
|
return ''; |
|
353
|
|
|
} |
|
354
|
|
|
throw new UnsupportedCharsetException("Charset $charset is not supported"); |
|
355
|
|
|
} |
|
356
|
3 |
|
return $ret; |
|
357
|
|
|
} |
|
358
|
|
|
|
|
359
|
|
|
/** |
|
360
|
|
|
* Converts the passed string's charset from the passed $fromCharset to the |
|
361
|
|
|
* passed $toCharset |
|
362
|
|
|
* |
|
363
|
|
|
* The function attempts to use mb_convert_encoding if possible, and falls |
|
364
|
|
|
* back to iconv if not. If the source or destination character sets aren't |
|
365
|
|
|
* supported, a blank string is returned. |
|
366
|
|
|
* |
|
367
|
|
|
* @throws UnsupportedCharsetException if iconv fails |
|
368
|
|
|
*/ |
|
369
|
11 |
|
public function convert(string $str, string $fromCharset, string $toCharset) : string |
|
370
|
|
|
{ |
|
371
|
|
|
// there may be some mb-supported encodings not supported by iconv (on my libiconv for instance |
|
372
|
|
|
// HZ isn't supported), and so it may happen that failing an mb_convert_encoding, an iconv |
|
373
|
|
|
// may also fail even though both support an encoding separately. |
|
374
|
|
|
// For cases like that, a two-way encoding is done with UTF-8 as an intermediary. |
|
375
|
|
|
|
|
376
|
11 |
|
$from = $this->getMbCharset($fromCharset); |
|
377
|
11 |
|
$to = $this->getMbCharset($toCharset); |
|
378
|
|
|
|
|
379
|
11 |
|
if ($str !== '') { |
|
380
|
9 |
|
if ($from !== false && $to === false) { |
|
381
|
9 |
|
$str = \mb_convert_encoding($str, 'UTF-8', $from); |
|
382
|
9 |
|
return $this->iconv('UTF-8', $this->getIconvAlias($toCharset), $str); |
|
|
|
|
|
|
383
|
8 |
|
} elseif ($from === false && $to !== false) { |
|
384
|
7 |
|
$str = $this->iconv($this->getIconvAlias($fromCharset), 'UTF-8', $str); |
|
385
|
7 |
|
return \mb_convert_encoding($str, $to, 'UTF-8'); |
|
386
|
8 |
|
} elseif ($from !== false && $to !== false) { |
|
387
|
6 |
|
return \mb_convert_encoding($str, $to, $from); |
|
388
|
|
|
} |
|
389
|
2 |
|
return $this->iconv( |
|
390
|
2 |
|
$this->getIconvAlias($fromCharset), |
|
391
|
2 |
|
$this->getIconvAlias($toCharset), |
|
392
|
2 |
|
$str |
|
393
|
2 |
|
); |
|
394
|
|
|
} |
|
395
|
2 |
|
return $str; |
|
396
|
|
|
} |
|
397
|
|
|
|
|
398
|
|
|
/** |
|
399
|
|
|
* Returns true if the passed string is valid in the $charset encoding. |
|
400
|
|
|
* |
|
401
|
|
|
* Either uses mb_check_encoding, or iconv if it's not a supported mb |
|
402
|
|
|
* encoding. |
|
403
|
|
|
*/ |
|
404
|
|
|
public function checkEncoding(string $str, string $charset) : bool |
|
405
|
|
|
{ |
|
406
|
|
|
$mb = $this->getMbCharset($charset); |
|
407
|
|
|
if ($mb !== false) { |
|
408
|
|
|
return \mb_check_encoding($str, $mb); |
|
409
|
|
|
} |
|
410
|
|
|
$ic = $this->getIconvAlias($charset); |
|
411
|
|
|
return (@\iconv($ic, $ic . '//TRANSLIT//IGNORE', $str) !== false); |
|
412
|
|
|
} |
|
413
|
|
|
|
|
414
|
|
|
/** |
|
415
|
|
|
* Uses either mb_strlen or iconv_strlen to return the number of characters |
|
416
|
|
|
* in the passed $str for the given $charset |
|
417
|
|
|
* |
|
418
|
|
|
* @throws UnsupportedCharsetException if iconv fails |
|
419
|
|
|
*/ |
|
420
|
3 |
|
public function getLength(string $str, string $charset) : int |
|
421
|
|
|
{ |
|
422
|
3 |
|
$mb = $this->getMbCharset($charset); |
|
423
|
3 |
|
if ($mb !== false) { |
|
424
|
2 |
|
return \mb_strlen($str, $mb); |
|
425
|
|
|
} |
|
426
|
3 |
|
return $this->iconvStrlen($str, $this->getIconvAlias($charset)); |
|
427
|
|
|
} |
|
428
|
|
|
|
|
429
|
|
|
/** |
|
430
|
|
|
* Uses either mb_substr or iconv_substr to create and return a substring of |
|
431
|
|
|
* the passed $str. |
|
432
|
|
|
* |
|
433
|
|
|
* If the offset provided in $start is greater than the length of the |
|
434
|
|
|
* string, an empty string is returned. |
|
435
|
|
|
* |
|
436
|
|
|
* @throws UnsupportedCharsetException if iconv fails |
|
437
|
|
|
*/ |
|
438
|
4 |
|
public function getSubstr(string $str, string $charset, int $start, ?int $length = null) : string |
|
439
|
|
|
{ |
|
440
|
4 |
|
$mb = $this->getMbCharset($charset); |
|
441
|
4 |
|
if ($mb !== false) { |
|
442
|
2 |
|
return \mb_substr($str, $start, $length, $mb); |
|
443
|
|
|
} |
|
444
|
4 |
|
$ic = $this->getIconvAlias($charset); |
|
445
|
4 |
|
if ($ic === 'CP1258') { |
|
446
|
|
|
// iconv_substr fails with CP1258 for some reason, and returns only |
|
447
|
|
|
// a subset of characters (e.g. the first 5, instead of $length) |
|
448
|
|
|
$str = $this->convert($str, $ic, 'UTF-8'); |
|
449
|
|
|
return $this->convert($this->getSubstr($str, 'UTF-8', $start, $length), 'UTF-8', $ic); |
|
450
|
|
|
} |
|
451
|
4 |
|
return $this->iconvSubstr($str, $ic, $start, $length); |
|
452
|
|
|
} |
|
453
|
|
|
|
|
454
|
|
|
/** |
|
455
|
|
|
* Looks up a charset from mb_list_encodings and identified aliases, |
|
456
|
|
|
* checking if the lookup has been cached already first. |
|
457
|
|
|
* |
|
458
|
|
|
* If the encoding is not listed, the method will return false. |
|
459
|
|
|
* |
|
460
|
|
|
* On success, the method will return the charset name as accepted by mb_*. |
|
461
|
|
|
* |
|
462
|
|
|
* @return string|bool |
|
463
|
|
|
*/ |
|
464
|
14 |
|
private function getMbCharset(string $cs) |
|
465
|
|
|
{ |
|
466
|
14 |
|
$normalized = $this->getNormalizedCharset($cs); |
|
467
|
14 |
|
if (\array_key_exists($normalized, self::$mbListedEncodings)) { |
|
468
|
11 |
|
return self::$mbListedEncodings[$normalized]; |
|
469
|
12 |
|
} elseif (\array_key_exists($normalized, self::$mbAliases)) { |
|
470
|
5 |
|
return self::$mbAliases[$normalized]; |
|
471
|
|
|
} |
|
472
|
12 |
|
return false; |
|
473
|
|
|
} |
|
474
|
|
|
|
|
475
|
|
|
/** |
|
476
|
|
|
* Looks up the passed charset in self::$iconvAliases, returning the mapped |
|
477
|
|
|
* charset if applicable. Otherwise returns charset. |
|
478
|
|
|
* |
|
479
|
|
|
* @return string the mapped charset (if mapped) or $cs otherwise |
|
480
|
|
|
*/ |
|
481
|
12 |
|
private function getIconvAlias(string $cs) : string |
|
482
|
|
|
{ |
|
483
|
12 |
|
$normalized = $this->getNormalizedCharset($cs); |
|
484
|
12 |
|
if (\array_key_exists($normalized, self::$iconvAliases)) { |
|
485
|
8 |
|
return static::$iconvAliases[$normalized]; |
|
486
|
|
|
} |
|
487
|
11 |
|
return $cs; |
|
488
|
|
|
} |
|
489
|
|
|
} |
|
490
|
|
|
|
zbateson /
mb-wrapper
Loading history...