|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* This is an internal development file. It should NOT be included in |
|
5
|
|
|
* any SMF distribution packages. |
|
6
|
|
|
* |
|
7
|
|
|
* This file exists to make it easy for developers to update the |
|
8
|
|
|
* Unicode data in Subs-Charset.php whenever a new version of the |
|
9
|
|
|
* Unicode Character Database is released. Just run this file from the |
|
10
|
|
|
* command line in order to perform the update. |
|
11
|
|
|
* |
|
12
|
|
|
* Simple Machines Forum (SMF) |
|
13
|
|
|
* |
|
14
|
|
|
* @package SMF |
|
15
|
|
|
* @author Simple Machines https://www.simplemachines.org |
|
16
|
|
|
* @copyright 2022 Simple Machines and individual contributors |
|
17
|
|
|
* @license https://www.simplemachines.org/about/smf/license.php BSD |
|
18
|
|
|
* |
|
19
|
|
|
* @version 2.1.2 |
|
20
|
|
|
*/ |
|
21
|
|
|
|
|
22
|
|
|
$unicode_data_url = 'https://unicode.org/Public/UCD/latest/ucd'; |
|
23
|
|
|
|
|
24
|
|
|
$sourcedir = realpath(dirname(__DIR__) . '/Sources'); |
|
25
|
|
|
$unicodedir = $sourcedir . '/Unicode'; |
|
26
|
|
|
|
|
27
|
|
|
$full_decomposition_maps = array(); |
|
28
|
|
|
$funcs = array( |
|
29
|
|
|
'utf8_normalize_d_maps' => array( |
|
30
|
|
|
'file' => 'DecompositionCanonical.php', |
|
31
|
|
|
'key_type' => 'hexchar', |
|
32
|
|
|
'val_type' => 'hexchar', |
|
33
|
|
|
'data' => array(), |
|
34
|
|
|
), |
|
35
|
|
|
'utf8_normalize_kd_maps' => array( |
|
36
|
|
|
'file' => 'DecompositionCompatibility.php', |
|
37
|
|
|
'key_type' => 'hexchar', |
|
38
|
|
|
'val_type' => 'hexchar', |
|
39
|
|
|
'data' => array(), |
|
40
|
|
|
), |
|
41
|
|
|
'utf8_compose_maps' => array( |
|
42
|
|
|
'file' => 'Composition.php', |
|
43
|
|
|
'key_type' => 'hexchar', |
|
44
|
|
|
'val_type' => 'hexchar', |
|
45
|
|
|
'data' => array(), |
|
46
|
|
|
), |
|
47
|
|
|
'utf8_combining_classes' => array( |
|
48
|
|
|
'file' => 'CombiningClasses.php', |
|
49
|
|
|
'key_type' => 'hexchar', |
|
50
|
|
|
'val_type' => 'int', |
|
51
|
|
|
'data' => array(), |
|
52
|
|
|
), |
|
53
|
|
|
'utf8_strtolower_maps' => array( |
|
54
|
|
|
'file' => 'CaseLower.php', |
|
55
|
|
|
'key_type' => 'hexchar', |
|
56
|
|
|
'val_type' => 'hexchar', |
|
57
|
|
|
'data' => array(), |
|
58
|
|
|
), |
|
59
|
|
|
'utf8_strtoupper_maps' => array( |
|
60
|
|
|
'file' => 'CaseUpper.php', |
|
61
|
|
|
'key_type' => 'hexchar', |
|
62
|
|
|
'val_type' => 'hexchar', |
|
63
|
|
|
'data' => array(), |
|
64
|
|
|
), |
|
65
|
|
|
'utf8_casefold_maps' => array( |
|
66
|
|
|
'file' => 'CaseFold.php', |
|
67
|
|
|
'key_type' => 'hexchar', |
|
68
|
|
|
'val_type' => 'hexchar', |
|
69
|
|
|
'data' => array(), |
|
70
|
|
|
), |
|
71
|
|
|
'utf8_default_ignorables' => array( |
|
72
|
|
|
'file' => 'DefaultIgnorables.php', |
|
73
|
|
|
'key_type' => 'int', |
|
74
|
|
|
'val_type' => 'hexchar', |
|
75
|
|
|
'data' => array(), |
|
76
|
|
|
), |
|
77
|
|
|
'utf8_regex_properties' => array( |
|
78
|
|
|
'file' => 'RegularExpressions.php', |
|
79
|
|
|
'key_type' => 'string', |
|
80
|
|
|
'val_type' => 'string', |
|
81
|
|
|
'propfiles' => array( |
|
82
|
|
|
'DerivedCoreProperties.txt', |
|
83
|
|
|
'PropList.txt', |
|
84
|
|
|
'emoji/emoji-data.txt', |
|
85
|
|
|
'extracted/DerivedGeneralCategory.txt', |
|
86
|
|
|
), |
|
87
|
|
|
'props' => array( |
|
88
|
|
|
'Bidi_Control', |
|
89
|
|
|
'Cn', |
|
90
|
|
|
'Default_Ignorable_Code_Point', |
|
91
|
|
|
'Emoji', |
|
92
|
|
|
'Emoji_Modifier', |
|
93
|
|
|
'Ideographic', |
|
94
|
|
|
'Join_Control', |
|
95
|
|
|
'Regional_Indicator', |
|
96
|
|
|
'Variation_Selector', |
|
97
|
|
|
), |
|
98
|
|
|
'data' => array(), |
|
99
|
|
|
), |
|
100
|
|
|
'utf8_regex_variation_selectors' => array( |
|
101
|
|
|
'file' => 'RegularExpressions.php', |
|
102
|
|
|
'key_type' => 'string', |
|
103
|
|
|
'val_type' => 'string', |
|
104
|
|
|
'data' => array(), |
|
105
|
|
|
), |
|
106
|
|
|
'utf8_regex_joining_type' => array( |
|
107
|
|
|
'file' => 'RegularExpressions.php', |
|
108
|
|
|
'key_type' => 'string', |
|
109
|
|
|
'val_type' => 'string', |
|
110
|
|
|
'data' => array(), |
|
111
|
|
|
), |
|
112
|
|
|
'utf8_regex_indic' => array( |
|
113
|
|
|
'file' => 'RegularExpressions.php', |
|
114
|
|
|
'key_type' => 'string', |
|
115
|
|
|
'val_type' => 'string', |
|
116
|
|
|
'data' => array(), |
|
117
|
|
|
), |
|
118
|
|
|
); |
|
119
|
|
|
|
|
120
|
|
|
foreach ($funcs as $func_name => $func_info) |
|
121
|
|
|
{ |
|
122
|
|
|
if (!is_file($unicodedir . '/' . $func_info['file']) || !is_writable($unicodedir . '/' . $func_info['file'])) |
|
123
|
|
|
{ |
|
124
|
|
|
die($unicodedir . '/' . $func_info['file'] . ' not found or not writable.'); |
|
125
|
|
|
} |
|
126
|
|
|
} |
|
127
|
|
|
|
|
128
|
|
|
@ini_set('memory_limit', '256M'); |
|
129
|
|
|
|
|
130
|
|
|
/********************************************* |
|
131
|
|
|
* Part 1: Normalization, case folding, etc. * |
|
132
|
|
|
*********************************************/ |
|
133
|
|
|
|
|
134
|
|
|
// We need some of these for further analysis below. |
|
135
|
|
|
$derived_normalization_props = array(); |
|
136
|
|
|
foreach (file($unicode_data_url . '/DerivedNormalizationProps.txt') as $line) |
|
137
|
|
|
{ |
|
138
|
|
|
$line = substr($line, 0, strcspn($line, '#')); |
|
139
|
|
|
|
|
140
|
|
|
if (strpos($line, ';') === false) |
|
141
|
|
|
{ |
|
142
|
|
|
continue; |
|
143
|
|
|
} |
|
144
|
|
|
|
|
145
|
|
|
$fields = explode(';', $line); |
|
146
|
|
|
|
|
147
|
|
|
foreach ($fields as $key => $value) |
|
148
|
|
|
{ |
|
149
|
|
|
$fields[$key] = trim($value); |
|
150
|
|
|
} |
|
151
|
|
|
|
|
152
|
|
|
if (!isset($derived_normalization_props[$fields[1]])) |
|
153
|
|
|
{ |
|
154
|
|
|
$derived_normalization_props[$fields[1]] = array(); |
|
155
|
|
|
} |
|
156
|
|
|
|
|
157
|
|
|
if (strpos($fields[0], '..') === false) |
|
158
|
|
|
{ |
|
159
|
|
|
$entities = array('&#x' . $fields[0] . ';'); |
|
160
|
|
|
} |
|
161
|
|
|
else |
|
162
|
|
|
{ |
|
163
|
|
|
$entities = array(); |
|
164
|
|
|
|
|
165
|
|
|
list($start, $end) = explode('..', $fields[0]); |
|
166
|
|
|
|
|
167
|
|
|
$ord_s = hexdec($start); |
|
168
|
|
|
$ord_e = hexdec($end); |
|
169
|
|
|
|
|
170
|
|
|
$ord = $ord_s; |
|
171
|
|
|
while ($ord <= $ord_e) |
|
172
|
|
|
{ |
|
173
|
|
|
$entities[] = '&#x' . strtoupper(sprintf('%04s', dechex($ord++))) . ';'; |
|
174
|
|
|
} |
|
175
|
|
|
} |
|
176
|
|
|
|
|
177
|
|
|
$value = ''; |
|
178
|
|
|
if (!isset($fields[2])) |
|
179
|
|
|
{ |
|
180
|
|
|
$value = 'SAME'; |
|
181
|
|
|
} |
|
182
|
|
|
elseif (in_array($fields[1], array('FC_NFKC', 'NFKC_CF'))) |
|
183
|
|
|
{ |
|
184
|
|
|
$value = trim($fields[2]) !== '' ? '&#x' . str_replace(' ', '; &#x', trim($fields[2])) . ';' : ''; |
|
185
|
|
|
} |
|
186
|
|
|
else |
|
187
|
|
|
{ |
|
188
|
|
|
$value = $fields[2]; |
|
189
|
|
|
} |
|
190
|
|
|
|
|
191
|
|
|
foreach ($entities as $entity) |
|
192
|
|
|
{ |
|
193
|
|
|
$derived_normalization_props[$fields[1]][$entity] = $value === 'SAME' ? $entity : $value; |
|
194
|
|
|
} |
|
195
|
|
|
} |
|
196
|
|
|
|
|
197
|
|
|
// Go through all the characters in the Unicode database. |
|
198
|
|
|
$char_data = array(); |
|
199
|
|
|
foreach (file($unicode_data_url . '/UnicodeData.txt') as $line) |
|
200
|
|
|
{ |
|
201
|
|
|
$fields = explode(';', $line); |
|
202
|
|
|
|
|
203
|
|
|
if (!empty($fields[3])) |
|
204
|
|
|
{ |
|
205
|
|
|
$funcs['utf8_combining_classes']['data']['&#x' . $fields[0] . ';'] = trim($fields[3]); |
|
206
|
|
|
} |
|
207
|
|
|
|
|
208
|
|
|
// Uppercase maps. |
|
209
|
|
|
if ($fields[12] !== '') |
|
210
|
|
|
{ |
|
211
|
|
|
$funcs['utf8_strtoupper_maps']['data']['&#x' . $fields[0] . ';'] = '&#x' . $fields[12] . ';'; |
|
212
|
|
|
} |
|
213
|
|
|
|
|
214
|
|
|
// Lowercase maps. |
|
215
|
|
|
if ($fields[13] !== '') |
|
216
|
|
|
{ |
|
217
|
|
|
$funcs['utf8_strtolower_maps']['data']['&#x' . $fields[0] . ';'] = '&#x' . $fields[13] . ';'; |
|
218
|
|
|
} |
|
219
|
|
|
|
|
220
|
|
|
// Remember this character's general category for later. |
|
221
|
|
|
$char_data['&#x' . $fields[0] . ';']['General_Category'] = trim($fields[2]); |
|
222
|
|
|
|
|
223
|
|
|
if ($fields[5] === '') |
|
224
|
|
|
{ |
|
225
|
|
|
continue; |
|
226
|
|
|
} |
|
227
|
|
|
|
|
228
|
|
|
// All canonical decompositions AND all compatibility decompositions. |
|
229
|
|
|
$full_decomposition_maps['&#x' . $fields[0] . ';'] = '&#x' . str_replace(' ', '; &#x', trim(strip_tags($fields[5]))) . ';'; |
|
230
|
|
|
|
|
231
|
|
|
// Just the canonical decompositions. |
|
232
|
|
|
if (strpos($fields[5], '<') === false) |
|
233
|
|
|
{ |
|
234
|
|
|
$funcs['utf8_normalize_d_maps']['data']['&#x' . $fields[0] . ';'] = '&#x' . str_replace(' ', '; &#x', trim($fields[5])) . ';'; |
|
235
|
|
|
} |
|
236
|
|
|
} |
|
237
|
|
|
|
|
238
|
|
|
foreach (file($unicode_data_url . '/CaseFolding.txt') as $line) |
|
239
|
|
|
{ |
|
240
|
|
|
$line = substr($line, 0, strcspn($line, '#')); |
|
241
|
|
|
|
|
242
|
|
|
if (strpos($line, ';') === false) |
|
243
|
|
|
{ |
|
244
|
|
|
continue; |
|
245
|
|
|
} |
|
246
|
|
|
|
|
247
|
|
|
$fields = explode(';', $line); |
|
248
|
|
|
|
|
249
|
|
|
foreach ($fields as $key => $value) |
|
250
|
|
|
{ |
|
251
|
|
|
$fields[$key] = trim($value); |
|
252
|
|
|
} |
|
253
|
|
|
|
|
254
|
|
|
// Full casefolding. |
|
255
|
|
|
if (in_array($fields[1], array('C', 'F'))) |
|
256
|
|
|
{ |
|
257
|
|
|
$funcs['utf8_casefold_maps']['data']['&#x' . $fields[0] . ';'] = '&#x' . str_replace(' ', '; &#x', trim($fields[2])) . ';'; |
|
258
|
|
|
} |
|
259
|
|
|
|
|
260
|
|
|
// Simple casefolding. Currently unused. |
|
261
|
|
|
// if (in_array($fields[1], array('C', 'S'))) |
|
262
|
|
|
// $funcs['utf8_casefold_simple_maps']['data']['&#x' . $fields[0] . ';'] = '&#x' . str_replace(' ', '; &#x', trim($fields[2])) . ';'; |
|
263
|
|
|
} |
|
264
|
|
|
|
|
265
|
|
|
// Recursively iterate until we reach the final decomposition forms. |
|
266
|
|
|
// This is necessary because some characters decompose to other characters that |
|
267
|
|
|
// themselves decompose further. |
|
268
|
|
|
$changed = true; |
|
269
|
|
|
while ($changed) |
|
270
|
|
|
{ |
|
271
|
|
|
$temp = array(); |
|
272
|
|
|
foreach ($full_decomposition_maps as $composed => $decomposed) |
|
273
|
|
|
{ |
|
274
|
|
|
$parts = strpos($decomposed, ' ') !== false ? explode(' ', $decomposed) : (array) $decomposed; |
|
275
|
|
|
|
|
276
|
|
|
foreach ($parts as $partnum => $hex) |
|
277
|
|
|
{ |
|
278
|
|
|
if (isset($full_decomposition_maps[$hex])) |
|
279
|
|
|
{ |
|
280
|
|
|
$parts[$partnum] = $full_decomposition_maps[$hex]; |
|
281
|
|
|
} |
|
282
|
|
|
} |
|
283
|
|
|
|
|
284
|
|
|
$decomposed = implode(' ', $parts); |
|
285
|
|
|
unset($parts); |
|
286
|
|
|
|
|
287
|
|
|
$temp[$composed] = $decomposed; |
|
288
|
|
|
} |
|
289
|
|
|
|
|
290
|
|
|
$changed = $full_decomposition_maps !== $temp; |
|
291
|
|
|
|
|
292
|
|
|
$full_decomposition_maps = $temp; |
|
293
|
|
|
} |
|
294
|
|
|
|
|
295
|
|
|
// Same as above, but using only canonical decompositions. |
|
296
|
|
|
$changed = true; |
|
297
|
|
|
$iteration = 0; |
|
298
|
|
|
while ($changed) |
|
299
|
|
|
{ |
|
300
|
|
|
$temp = array(); |
|
301
|
|
|
foreach ($funcs['utf8_normalize_d_maps']['data'] as $composed => $decomposed) |
|
302
|
|
|
{ |
|
303
|
|
|
if ($iteration === 0 && !in_array($composed, $derived_normalization_props['Full_Composition_Exclusion'])) |
|
304
|
|
|
{ |
|
305
|
|
|
$funcs['utf8_compose_maps']['data'][$decomposed] = $composed; |
|
306
|
|
|
} |
|
307
|
|
|
|
|
308
|
|
|
$parts = strpos($decomposed, ' ') !== false ? explode(' ', $decomposed) : (array) $decomposed; |
|
309
|
|
|
|
|
310
|
|
|
foreach ($parts as $partnum => $hex) |
|
311
|
|
|
{ |
|
312
|
|
|
if (isset($funcs['utf8_normalize_d_maps']['data'][$hex])) |
|
313
|
|
|
{ |
|
314
|
|
|
$parts[$partnum] = $funcs['utf8_normalize_d_maps']['data'][$hex]; |
|
315
|
|
|
} |
|
316
|
|
|
} |
|
317
|
|
|
|
|
318
|
|
|
$decomposed = implode(' ', $parts); |
|
319
|
|
|
unset($parts); |
|
320
|
|
|
|
|
321
|
|
|
$temp[$composed] = $decomposed; |
|
322
|
|
|
} |
|
323
|
|
|
|
|
324
|
|
|
$changed = $funcs['utf8_normalize_d_maps']['data'] !== $temp; |
|
325
|
|
|
|
|
326
|
|
|
$funcs['utf8_normalize_d_maps']['data'] = $temp; |
|
327
|
|
|
$iteration++; |
|
328
|
|
|
} |
|
329
|
|
|
|
|
330
|
|
|
$funcs['utf8_normalize_kd_maps']['data'] = array_diff_assoc($full_decomposition_maps, $funcs['utf8_normalize_d_maps']['data']); |
|
331
|
|
|
unset($full_decomposition_maps, $derived_normalization_props); |
|
332
|
|
|
|
|
333
|
|
|
// Now update the files with the data we've got so far. |
|
334
|
|
|
foreach ($funcs as $func_name => $func_info) |
|
335
|
|
|
{ |
|
336
|
|
|
if (empty($func_info['data'])) |
|
337
|
|
|
{ |
|
338
|
|
|
continue; |
|
339
|
|
|
} |
|
340
|
|
|
|
|
341
|
|
|
export_func_to_file($func_name, $func_info); |
|
342
|
|
|
|
|
343
|
|
|
// Free up some memory. |
|
344
|
|
|
if ($func_name != 'utf8_combining_classes') |
|
345
|
|
|
{ |
|
346
|
|
|
unset($funcs[$func_name]); |
|
347
|
|
|
} |
|
348
|
|
|
} |
|
349
|
|
|
|
|
350
|
|
|
/*********************************** |
|
351
|
|
|
* Part 2: Regular expression data * |
|
352
|
|
|
***********************************/ |
|
353
|
|
|
|
|
354
|
|
|
// Build regular expression classes for extended Unicode properties. |
|
355
|
|
|
foreach ($funcs['utf8_regex_properties']['propfiles'] as $filename) |
|
356
|
|
|
{ |
|
357
|
|
|
foreach (file($unicode_data_url . '/' . $filename) as $line) |
|
358
|
|
|
{ |
|
359
|
|
|
$line = substr($line, 0, strcspn($line, '#')); |
|
360
|
|
|
|
|
361
|
|
|
if (strpos($line, ';') === false) |
|
362
|
|
|
{ |
|
363
|
|
|
continue; |
|
364
|
|
|
} |
|
365
|
|
|
|
|
366
|
|
|
$fields = explode(';', $line); |
|
367
|
|
|
|
|
368
|
|
|
foreach ($fields as $key => $value) |
|
369
|
|
|
{ |
|
370
|
|
|
$fields[$key] = trim($value); |
|
371
|
|
|
} |
|
372
|
|
|
|
|
373
|
|
|
if (in_array($fields[1], $funcs['utf8_regex_properties']['props'])) |
|
374
|
|
|
{ |
|
375
|
|
|
if (!isset($funcs['utf8_regex_properties']['data'][$fields[1]])) |
|
376
|
|
|
{ |
|
377
|
|
|
$funcs['utf8_regex_properties']['data'][$fields[1]] = ''; |
|
378
|
|
|
} |
|
379
|
|
|
|
|
380
|
|
|
$funcs['utf8_regex_properties']['data'][$fields[1]] .= '\\x{' . str_replace('..', '}-\\x{', $fields[0]) . '}'; |
|
381
|
|
|
} |
|
382
|
|
|
|
|
383
|
|
|
// We also track 'Default_Ignorable_Code_Point' property in a separate array. |
|
384
|
|
|
if ($fields[1] !== 'Default_Ignorable_Code_Point') |
|
385
|
|
|
{ |
|
386
|
|
|
continue; |
|
387
|
|
|
} |
|
388
|
|
|
|
|
389
|
|
|
if (strpos($fields[0], '..') === false) |
|
390
|
|
|
{ |
|
391
|
|
|
$funcs['utf8_default_ignorables']['data'][] = '&#x' . $fields[0] . ';'; |
|
392
|
|
|
} |
|
393
|
|
|
else |
|
394
|
|
|
{ |
|
395
|
|
|
list($start, $end) = explode('..', $fields[0]); |
|
396
|
|
|
|
|
397
|
|
|
$ord_s = hexdec($start); |
|
398
|
|
|
$ord_e = hexdec($end); |
|
399
|
|
|
|
|
400
|
|
|
$ord = $ord_s; |
|
401
|
|
|
while ($ord <= $ord_e) |
|
402
|
|
|
{ |
|
403
|
|
|
$funcs['utf8_default_ignorables']['data'][] = '&#x' . strtoupper(sprintf('%04s', dechex($ord++))) . ';'; |
|
404
|
|
|
} |
|
405
|
|
|
} |
|
406
|
|
|
} |
|
407
|
|
|
} |
|
408
|
|
|
ksort($funcs['utf8_regex_properties']['data']); |
|
409
|
|
|
|
|
410
|
|
|
// Build regular expression classes for filtering variation selectors. |
|
411
|
|
|
$files = array('StandardizedVariants.txt', 'emoji/emoji-variation-sequences.txt'); |
|
412
|
|
|
foreach ($files as $filename) |
|
413
|
|
|
{ |
|
414
|
|
|
foreach (file($unicode_data_url . '/' . $filename) as $line) |
|
415
|
|
|
{ |
|
416
|
|
|
$line = substr($line, 0, strcspn($line, '#')); |
|
417
|
|
|
|
|
418
|
|
|
if (strpos($line, ';') === false) |
|
419
|
|
|
{ |
|
420
|
|
|
continue; |
|
421
|
|
|
} |
|
422
|
|
|
|
|
423
|
|
|
$fields = explode(';', $line); |
|
424
|
|
|
|
|
425
|
|
|
foreach ($fields as $key => $value) |
|
426
|
|
|
{ |
|
427
|
|
|
$fields[$key] = trim($value); |
|
428
|
|
|
} |
|
429
|
|
|
|
|
430
|
|
|
list($base_char, $variation_selector) = explode(' ', $fields[0]); |
|
431
|
|
|
|
|
432
|
|
|
$funcs['utf8_regex_variation_selectors']['data']['\\x{' . $variation_selector . '}'][] = hexdec($base_char); |
|
433
|
|
|
} |
|
434
|
|
|
} |
|
435
|
|
|
foreach ($funcs['utf8_regex_variation_selectors']['data'] as $variation_selector => $ords) |
|
436
|
|
|
{ |
|
437
|
|
|
$class_string = ''; |
|
438
|
|
|
|
|
439
|
|
|
$current_range = array('start' => null, 'end' => null); |
|
440
|
|
|
foreach ($ords as $ord) |
|
441
|
|
|
{ |
|
442
|
|
|
if (!isset($current_range['start'])) |
|
443
|
|
|
{ |
|
444
|
|
|
$current_range['start'] = $ord; |
|
445
|
|
|
} |
|
446
|
|
|
|
|
447
|
|
|
if (!isset($current_range['end']) || $ord == $current_range['end'] + 1) |
|
448
|
|
|
{ |
|
449
|
|
|
$current_range['end'] = $ord; |
|
450
|
|
|
continue; |
|
451
|
|
|
} |
|
452
|
|
|
else |
|
453
|
|
|
{ |
|
454
|
|
|
$class_string .= '\\x{' . strtoupper(sprintf('%04s', dechex($current_range['start']))) . '}'; |
|
|
|
|
|
|
455
|
|
|
|
|
456
|
|
|
if ($current_range['start'] != $current_range['end']) |
|
457
|
|
|
{ |
|
458
|
|
|
$class_string .= '-\\x{' . strtoupper(sprintf('%04s', dechex($current_range['end']))) . '}'; |
|
|
|
|
|
|
459
|
|
|
} |
|
460
|
|
|
|
|
461
|
|
|
$current_range = array('start' => $ord, 'end' => $ord); |
|
462
|
|
|
} |
|
463
|
|
|
} |
|
464
|
|
|
|
|
465
|
|
|
if (isset($current_range['start'])) |
|
466
|
|
|
{ |
|
467
|
|
|
$class_string .= '\\x{' . strtoupper(sprintf('%04s', dechex($current_range['start']))) . '}'; |
|
468
|
|
|
|
|
469
|
|
|
if ($current_range['start'] != $current_range['end']) |
|
470
|
|
|
{ |
|
471
|
|
|
$class_string .= '-\\x{' . strtoupper(sprintf('%04s', dechex($current_range['end']))) . '}'; |
|
472
|
|
|
} |
|
473
|
|
|
} |
|
474
|
|
|
|
|
475
|
|
|
// As of Unicode 14.0, \x{FE0E} and \x{FE0F} work with identical ranges of base characters. |
|
476
|
|
|
if (($identical = array_search($class_string, $funcs['utf8_regex_variation_selectors']['data'])) !== false) |
|
477
|
|
|
{ |
|
478
|
|
|
unset( |
|
479
|
|
|
$funcs['utf8_regex_variation_selectors']['data'][$identical], |
|
480
|
|
|
$funcs['utf8_regex_variation_selectors']['data'][$variation_selector] |
|
481
|
|
|
); |
|
482
|
|
|
|
|
483
|
|
|
$compound_selector = array($identical, $variation_selector); |
|
484
|
|
|
sort($compound_selector); |
|
485
|
|
|
|
|
486
|
|
|
$variation_selector = implode('', $compound_selector); |
|
487
|
|
|
} |
|
488
|
|
|
|
|
489
|
|
|
$funcs['utf8_regex_variation_selectors']['data'][$variation_selector] = $class_string; |
|
490
|
|
|
} |
|
491
|
|
|
krsort($funcs['utf8_regex_variation_selectors']['data']); |
|
492
|
|
|
|
|
493
|
|
|
// The regex classes for join control tests require info about language scripts. |
|
494
|
|
|
$script_stats = array(); |
|
495
|
|
|
$script_aliases = array(); |
|
496
|
|
|
foreach (file($unicode_data_url . '/PropertyValueAliases.txt') as $line) |
|
497
|
|
|
{ |
|
498
|
|
|
$line = substr($line, 0, strcspn($line, '#')); |
|
499
|
|
|
|
|
500
|
|
|
if (strpos($line, ';') === false) |
|
501
|
|
|
{ |
|
502
|
|
|
continue; |
|
503
|
|
|
} |
|
504
|
|
|
|
|
505
|
|
|
$fields = explode(';', $line); |
|
506
|
|
|
|
|
507
|
|
|
foreach ($fields as $key => $value) |
|
508
|
|
|
{ |
|
509
|
|
|
$fields[$key] = trim($value); |
|
510
|
|
|
} |
|
511
|
|
|
|
|
512
|
|
|
if ($fields[0] !== 'sc') |
|
513
|
|
|
{ |
|
514
|
|
|
continue; |
|
515
|
|
|
} |
|
516
|
|
|
|
|
517
|
|
|
$script_aliases[$fields[1]] = $fields[2]; |
|
518
|
|
|
} |
|
519
|
|
|
foreach (file($unicode_data_url . '/Scripts.txt') as $line) |
|
520
|
|
|
{ |
|
521
|
|
|
$line = substr($line, 0, strcspn($line, '#')); |
|
522
|
|
|
|
|
523
|
|
|
if (strpos($line, ';') === false) |
|
524
|
|
|
{ |
|
525
|
|
|
continue; |
|
526
|
|
|
} |
|
527
|
|
|
|
|
528
|
|
|
$fields = explode(';', $line); |
|
529
|
|
|
|
|
530
|
|
|
foreach ($fields as $key => $value) |
|
531
|
|
|
{ |
|
532
|
|
|
$fields[$key] = trim($value); |
|
533
|
|
|
} |
|
534
|
|
|
|
|
535
|
|
|
if (in_array($fields[1], array('Common', 'Inherited'))) |
|
536
|
|
|
{ |
|
537
|
|
|
continue; |
|
538
|
|
|
} |
|
539
|
|
|
|
|
540
|
|
|
if (strpos($fields[0], '..') === false) |
|
541
|
|
|
{ |
|
542
|
|
|
$char_data['&#x' . $fields[0] . ';']['scripts'][] = $fields[1]; |
|
543
|
|
|
} |
|
544
|
|
|
else |
|
545
|
|
|
{ |
|
546
|
|
|
list($start, $end) = explode('..', $fields[0]); |
|
547
|
|
|
|
|
548
|
|
|
$ord_s = hexdec($start); |
|
549
|
|
|
$ord_e = hexdec($end); |
|
550
|
|
|
|
|
551
|
|
|
$ord = $ord_s; |
|
552
|
|
|
while ($ord <= $ord_e) |
|
553
|
|
|
{ |
|
554
|
|
|
$char_data['&#x' . strtoupper(sprintf('%04s', dechex($ord++))) . ';']['scripts'][] = $fields[1]; |
|
555
|
|
|
} |
|
556
|
|
|
} |
|
557
|
|
|
} |
|
558
|
|
|
foreach (file($unicode_data_url . '/ScriptExtensions.txt') as $line) |
|
559
|
|
|
{ |
|
560
|
|
|
$line = substr($line, 0, strcspn($line, '#')); |
|
561
|
|
|
|
|
562
|
|
|
if (strpos($line, ';') === false) |
|
563
|
|
|
{ |
|
564
|
|
|
continue; |
|
565
|
|
|
} |
|
566
|
|
|
|
|
567
|
|
|
$fields = explode(';', $line); |
|
568
|
|
|
|
|
569
|
|
|
foreach ($fields as $key => $value) |
|
570
|
|
|
{ |
|
571
|
|
|
$fields[$key] = trim($value); |
|
572
|
|
|
} |
|
573
|
|
|
|
|
574
|
|
|
$char_scripts = array(); |
|
575
|
|
|
foreach (explode(' ', $fields[1]) as $alias) |
|
576
|
|
|
{ |
|
577
|
|
|
if (!in_array($script_aliases[$alias], array('Common', 'Inherited'))) |
|
578
|
|
|
{ |
|
579
|
|
|
$char_scripts[] = $script_aliases[$alias]; |
|
580
|
|
|
} |
|
581
|
|
|
} |
|
582
|
|
|
|
|
583
|
|
|
if (strpos($fields[0], '..') === false) |
|
584
|
|
|
{ |
|
585
|
|
|
foreach ($char_scripts as $char_script) |
|
586
|
|
|
{ |
|
587
|
|
|
$char_data['&#x' . $fields[0] . ';']['scripts'][] = $char_script; |
|
588
|
|
|
} |
|
589
|
|
|
} |
|
590
|
|
|
else |
|
591
|
|
|
{ |
|
592
|
|
|
list($start, $end) = explode('..', $fields[0]); |
|
593
|
|
|
|
|
594
|
|
|
$ord_s = hexdec($start); |
|
595
|
|
|
$ord_e = hexdec($end); |
|
596
|
|
|
|
|
597
|
|
|
$ord = $ord_s; |
|
598
|
|
|
while ($ord <= $ord_e) |
|
599
|
|
|
{ |
|
600
|
|
|
foreach ($char_scripts as $char_script) |
|
601
|
|
|
{ |
|
602
|
|
|
$char_data['&#x' . strtoupper(sprintf('%04s', dechex($ord++))) . ';']['scripts'][] = $char_script; |
|
603
|
|
|
} |
|
604
|
|
|
} |
|
605
|
|
|
} |
|
606
|
|
|
} |
|
607
|
|
|
foreach (file($unicode_data_url . '/DerivedAge.txt') as $line) |
|
608
|
|
|
{ |
|
609
|
|
|
$line = substr($line, 0, strcspn($line, '#')); |
|
610
|
|
|
|
|
611
|
|
|
if (strpos($line, ';') === false) |
|
612
|
|
|
{ |
|
613
|
|
|
continue; |
|
614
|
|
|
} |
|
615
|
|
|
|
|
616
|
|
|
$fields = explode(';', $line); |
|
617
|
|
|
|
|
618
|
|
|
foreach ($fields as $key => $value) |
|
619
|
|
|
{ |
|
620
|
|
|
$fields[$key] = trim($value); |
|
621
|
|
|
} |
|
622
|
|
|
|
|
623
|
|
|
$fields[1] = (float) $fields[1]; |
|
624
|
|
|
|
|
625
|
|
|
if (strpos($fields[0], '..') === false) |
|
626
|
|
|
{ |
|
627
|
|
|
$entity = '&#x' . $fields[0] . ';'; |
|
628
|
|
|
|
|
629
|
|
|
if (empty($char_data[$entity]['scripts'])) |
|
630
|
|
|
{ |
|
631
|
|
|
continue; |
|
632
|
|
|
} |
|
633
|
|
|
|
|
634
|
|
|
foreach ($char_data[$entity]['scripts'] as $char_script) |
|
635
|
|
|
{ |
|
636
|
|
|
if (!isset($script_stats[$char_script])) |
|
637
|
|
|
{ |
|
638
|
|
|
$script_stats[$char_script]['age'] = (float) $fields[1]; |
|
639
|
|
|
$script_stats[$char_script]['count'] = 1; |
|
640
|
|
|
} |
|
641
|
|
|
else |
|
642
|
|
|
{ |
|
643
|
|
|
$script_stats[$char_script]['age'] = min((float) $fields[1], $script_stats[$char_script]['age']); |
|
644
|
|
|
$script_stats[$char_script]['count']++; |
|
645
|
|
|
} |
|
646
|
|
|
} |
|
647
|
|
|
} |
|
648
|
|
|
else |
|
649
|
|
|
{ |
|
650
|
|
|
list($start, $end) = explode('..', $fields[0]); |
|
651
|
|
|
|
|
652
|
|
|
$ord_s = hexdec($start); |
|
653
|
|
|
$ord_e = hexdec($end); |
|
654
|
|
|
|
|
655
|
|
|
$ord = $ord_s; |
|
656
|
|
|
while ($ord <= $ord_e) |
|
657
|
|
|
{ |
|
658
|
|
|
$entity = '&#x' . strtoupper(sprintf('%04s', dechex($ord++))) . ';'; |
|
659
|
|
|
|
|
660
|
|
|
if (empty($char_data[$entity]['scripts'])) |
|
661
|
|
|
{ |
|
662
|
|
|
continue; |
|
663
|
|
|
} |
|
664
|
|
|
|
|
665
|
|
|
foreach ($char_data[$entity]['scripts'] as $char_script) |
|
666
|
|
|
{ |
|
667
|
|
|
if (!isset($script_stats[$char_script])) |
|
668
|
|
|
{ |
|
669
|
|
|
$script_stats[$char_script]['age'] = $fields[1]; |
|
670
|
|
|
$script_stats[$char_script]['count'] = 1; |
|
671
|
|
|
} |
|
672
|
|
|
else |
|
673
|
|
|
{ |
|
674
|
|
|
$script_stats[$char_script]['age'] = min($fields[1], $script_stats[$char_script]['age']); |
|
675
|
|
|
$script_stats[$char_script]['count']++; |
|
676
|
|
|
} |
|
677
|
|
|
} |
|
678
|
|
|
} |
|
679
|
|
|
} |
|
680
|
|
|
} |
|
681
|
|
|
|
|
682
|
|
|
// Build regex classes for join control tests in utf8_sanitize_invisibles: |
|
683
|
|
|
// 1. Cursive scripts like Arabic. |
|
684
|
|
|
foreach (file($unicode_data_url . '/extracted/DerivedJoiningType.txt') as $line) |
|
685
|
|
|
{ |
|
686
|
|
|
$line = substr($line, 0, strcspn($line, '#')); |
|
687
|
|
|
|
|
688
|
|
|
if (strpos($line, ';') === false) |
|
689
|
|
|
{ |
|
690
|
|
|
continue; |
|
691
|
|
|
} |
|
692
|
|
|
|
|
693
|
|
|
$fields = explode(';', $line); |
|
694
|
|
|
|
|
695
|
|
|
foreach ($fields as $key => $value) |
|
696
|
|
|
{ |
|
697
|
|
|
$fields[$key] = trim($value); |
|
698
|
|
|
} |
|
699
|
|
|
|
|
700
|
|
|
switch ($fields[1]) |
|
701
|
|
|
{ |
|
702
|
|
|
case 'C': |
|
703
|
|
|
$joining_type = 'Join_Causing'; |
|
704
|
|
|
break; |
|
705
|
|
|
|
|
706
|
|
|
case 'D': |
|
707
|
|
|
$joining_type = 'Dual_Joining'; |
|
708
|
|
|
break; |
|
709
|
|
|
|
|
710
|
|
|
case 'R': |
|
711
|
|
|
$joining_type = 'Right_Joining'; |
|
712
|
|
|
break; |
|
713
|
|
|
|
|
714
|
|
|
case 'L': |
|
715
|
|
|
$joining_type = 'Left_Joining'; |
|
716
|
|
|
break; |
|
717
|
|
|
|
|
718
|
|
|
case 'T': |
|
719
|
|
|
$joining_type = 'Transparent'; |
|
720
|
|
|
break; |
|
721
|
|
|
|
|
722
|
|
|
default: |
|
723
|
|
|
$joining_type = null; |
|
724
|
|
|
break; |
|
725
|
|
|
} |
|
726
|
|
|
|
|
727
|
|
|
if (!isset($joining_type)) |
|
728
|
|
|
{ |
|
729
|
|
|
continue; |
|
730
|
|
|
} |
|
731
|
|
|
|
|
732
|
|
|
$entity = '&#x' . substr($fields[0], 0, strcspn($fields[0], '.')) . ';'; |
|
733
|
|
|
|
|
734
|
|
|
if (empty($char_data[$entity]['scripts'])) |
|
735
|
|
|
{ |
|
736
|
|
|
continue; |
|
737
|
|
|
} |
|
738
|
|
|
|
|
739
|
|
|
foreach ($char_data[$entity]['scripts'] as $char_script) |
|
740
|
|
|
{ |
|
741
|
|
|
if (!isset($funcs['utf8_regex_joining_type']['data'][$char_script]['stats'])) |
|
742
|
|
|
{ |
|
743
|
|
|
$funcs['utf8_regex_joining_type']['data'][$char_script]['stats'] = $script_stats[$char_script]; |
|
744
|
|
|
} |
|
745
|
|
|
|
|
746
|
|
|
if (!isset($funcs['utf8_regex_joining_type']['data'][$char_script][$joining_type])) |
|
747
|
|
|
{ |
|
748
|
|
|
$funcs['utf8_regex_joining_type']['data'][$char_script][$joining_type] = array(); |
|
749
|
|
|
} |
|
750
|
|
|
|
|
751
|
|
|
$funcs['utf8_regex_joining_type']['data'][$char_script][$joining_type][] = '\\x{' . str_replace('..', '}-\\x{', $fields[0]) . '}'; |
|
752
|
|
|
} |
|
753
|
|
|
} |
|
754
|
|
|
// This sort works decently well to ensure widely used scripts are ranked before rare scripts. |
|
755
|
|
|
uasort($funcs['utf8_regex_joining_type']['data'], function ($a, $b) |
|
756
|
|
|
{ |
|
757
|
|
|
if ($a['stats']['age'] == $b['stats']['age']) |
|
758
|
|
|
{ |
|
759
|
|
|
return $b['stats']['count'] - $a['stats']['count']; |
|
760
|
|
|
} |
|
761
|
|
|
else |
|
762
|
|
|
{ |
|
763
|
|
|
return $a['stats']['age'] - $b['stats']['age']; |
|
764
|
|
|
} |
|
765
|
|
|
}); |
|
766
|
|
|
foreach ($funcs['utf8_regex_joining_type']['data'] as $char_script => $joining_types) |
|
767
|
|
|
{ |
|
768
|
|
|
unset($funcs['utf8_regex_joining_type']['data'][$char_script]['stats'], $joining_types['stats']); |
|
769
|
|
|
|
|
770
|
|
|
// If the only joining type in this script is transparent, we don't care about it. |
|
771
|
|
|
if (array_keys($joining_types) === array('Transparent')) |
|
772
|
|
|
{ |
|
773
|
|
|
unset($funcs['utf8_regex_joining_type']['data'][$char_script]); |
|
774
|
|
|
continue; |
|
775
|
|
|
} |
|
776
|
|
|
|
|
777
|
|
|
foreach ($joining_types as $joining_type => $value) |
|
778
|
|
|
{ |
|
779
|
|
|
sort($value); |
|
780
|
|
|
$funcs['utf8_regex_joining_type']['data'][$char_script][$joining_type] = implode('', $value); |
|
781
|
|
|
} |
|
782
|
|
|
} |
|
783
|
|
|
|
|
784
|
|
|
// 2. Indic scripts like Devanagari. |
|
785
|
|
|
foreach (file($unicode_data_url . '/IndicSyllabicCategory.txt') as $line) |
|
786
|
|
|
{ |
|
787
|
|
|
$line = substr($line, 0, strcspn($line, '#')); |
|
788
|
|
|
|
|
789
|
|
|
if (strpos($line, ';') === false) |
|
790
|
|
|
{ |
|
791
|
|
|
continue; |
|
792
|
|
|
} |
|
793
|
|
|
|
|
794
|
|
|
$fields = explode(';', $line); |
|
795
|
|
|
|
|
796
|
|
|
foreach ($fields as $key => $value) |
|
797
|
|
|
{ |
|
798
|
|
|
$fields[$key] = trim($value); |
|
799
|
|
|
} |
|
800
|
|
|
|
|
801
|
|
|
$insc = $fields[1]; |
|
802
|
|
|
|
|
803
|
|
|
if (!in_array($insc, array('Virama', 'Vowel_Dependent'))) |
|
804
|
|
|
{ |
|
805
|
|
|
continue; |
|
806
|
|
|
} |
|
807
|
|
|
|
|
808
|
|
|
$char_scripts = $char_data['&#x' . substr($fields[0], 0, strcspn($fields[0], '.')) . ';']['scripts']; |
|
809
|
|
|
|
|
810
|
|
|
if (empty($char_scripts)) |
|
811
|
|
|
{ |
|
812
|
|
|
continue; |
|
813
|
|
|
} |
|
814
|
|
|
|
|
815
|
|
|
foreach ($char_scripts as $char_script) |
|
816
|
|
|
{ |
|
817
|
|
|
if (!isset($funcs['utf8_regex_indic']['data'][$char_script]['stats'])) |
|
818
|
|
|
{ |
|
819
|
|
|
$funcs['utf8_regex_indic']['data'][$char_script]['stats'] = $script_stats[$char_script]; |
|
820
|
|
|
} |
|
821
|
|
|
|
|
822
|
|
|
if (!isset($funcs['utf8_regex_indic']['data'][$char_script][$insc])) |
|
823
|
|
|
{ |
|
824
|
|
|
$funcs['utf8_regex_indic']['data'][$char_script][$insc] = array(); |
|
825
|
|
|
} |
|
826
|
|
|
|
|
827
|
|
|
$funcs['utf8_regex_indic']['data'][$char_script][$insc][] = '\\x{' . str_replace('..', '}-\\x{', $fields[0]) . '}'; |
|
828
|
|
|
} |
|
829
|
|
|
} |
|
830
|
|
|
// Again, sort commonly used scripts before rare scripts. |
|
831
|
|
|
uasort($funcs['utf8_regex_indic']['data'], function ($a, $b) |
|
832
|
|
|
{ |
|
833
|
|
|
if ($a['stats']['age'] == $b['stats']['age']) |
|
834
|
|
|
{ |
|
835
|
|
|
return $b['stats']['count'] - $a['stats']['count']; |
|
836
|
|
|
} |
|
837
|
|
|
else |
|
838
|
|
|
{ |
|
839
|
|
|
return $a['stats']['age'] - $b['stats']['age']; |
|
840
|
|
|
} |
|
841
|
|
|
}); |
|
842
|
|
|
// We only want scripts with viramas. |
|
843
|
|
|
foreach ($funcs['utf8_regex_indic']['data'] as $char_script => $inscs) |
|
844
|
|
|
{ |
|
845
|
|
|
unset($funcs['utf8_regex_indic']['data'][$char_script]['stats'], $inscs['stats']); |
|
846
|
|
|
|
|
847
|
|
|
if (!isset($inscs['Virama'])) |
|
848
|
|
|
{ |
|
849
|
|
|
unset($funcs['utf8_regex_indic']['data'][$char_script]); |
|
850
|
|
|
continue; |
|
851
|
|
|
} |
|
852
|
|
|
} |
|
853
|
|
|
// Now add some more classes that we need for each script. |
|
854
|
|
|
foreach ($char_data as $entity => $info) |
|
855
|
|
|
{ |
|
856
|
|
|
if (empty($info['scripts'])) |
|
857
|
|
|
{ |
|
858
|
|
|
continue; |
|
859
|
|
|
} |
|
860
|
|
|
|
|
861
|
|
|
$ord = hexdec(trim($entity, '&#x;')); |
|
862
|
|
|
|
|
863
|
|
|
foreach ($info['scripts'] as $char_script) |
|
864
|
|
|
{ |
|
865
|
|
|
if (!isset($funcs['utf8_regex_indic']['data'][$char_script])) |
|
866
|
|
|
{ |
|
867
|
|
|
continue; |
|
868
|
|
|
} |
|
869
|
|
|
|
|
870
|
|
|
$funcs['utf8_regex_indic']['data'][$char_script]['All'][] = $ord; |
|
871
|
|
|
|
|
872
|
|
|
if (empty($info['General_Category'])) |
|
873
|
|
|
{ |
|
874
|
|
|
continue; |
|
875
|
|
|
} |
|
876
|
|
|
elseif ($info['General_Category'] == 'Mn') |
|
877
|
|
|
{ |
|
878
|
|
|
$funcs['utf8_regex_indic']['data'][$char_script]['Nonspacing_Mark'][] = $ord; |
|
879
|
|
|
|
|
880
|
|
|
if (!empty($funcs['utf8_combining_classes']['data'][$entity])) |
|
881
|
|
|
{ |
|
882
|
|
|
$funcs['utf8_regex_indic']['data'][$char_script]['Nonspacing_Combining_Mark'][] = $ord; |
|
883
|
|
|
} |
|
884
|
|
|
} |
|
885
|
|
|
elseif (substr($info['General_Category'], 0, 1) == 'L') |
|
886
|
|
|
{ |
|
887
|
|
|
$funcs['utf8_regex_indic']['data'][$char_script]['Letter'][] = $ord; |
|
888
|
|
|
} |
|
889
|
|
|
} |
|
890
|
|
|
} |
|
891
|
|
|
foreach ($funcs['utf8_regex_indic']['data'] as $char_script => $inscs) |
|
892
|
|
|
{ |
|
893
|
|
|
foreach ($inscs as $insc => $value) |
|
894
|
|
|
{ |
|
895
|
|
|
sort($value); |
|
896
|
|
|
|
|
897
|
|
|
if (!in_array($insc, array('All', 'Letter', 'Nonspacing_Mark', 'Nonspacing_Combining_Mark'))) |
|
898
|
|
|
{ |
|
899
|
|
|
$funcs['utf8_regex_indic']['data'][$char_script][$insc] = implode('', $value); |
|
900
|
|
|
continue; |
|
901
|
|
|
} |
|
902
|
|
|
|
|
903
|
|
|
$class_string = ''; |
|
904
|
|
|
|
|
905
|
|
|
$current_range = array('start' => null, 'end' => null); |
|
906
|
|
|
foreach ($value as $ord) |
|
907
|
|
|
{ |
|
908
|
|
|
if (!isset($current_range['start'])) |
|
909
|
|
|
{ |
|
910
|
|
|
$current_range['start'] = $ord; |
|
911
|
|
|
} |
|
912
|
|
|
|
|
913
|
|
|
if (!isset($current_range['end']) || $ord == $current_range['end'] + 1) |
|
914
|
|
|
{ |
|
915
|
|
|
$current_range['end'] = $ord; |
|
916
|
|
|
continue; |
|
917
|
|
|
} |
|
918
|
|
|
else |
|
919
|
|
|
{ |
|
920
|
|
|
$class_string .= '\\x{' . strtoupper(sprintf('%04s', dechex($current_range['start']))) . '}'; |
|
921
|
|
|
|
|
922
|
|
|
if ($current_range['start'] != $current_range['end']) |
|
923
|
|
|
{ |
|
924
|
|
|
$class_string .= '-\\x{' . strtoupper(sprintf('%04s', dechex($current_range['end']))) . '}'; |
|
925
|
|
|
} |
|
926
|
|
|
|
|
927
|
|
|
$current_range = array('start' => $ord, 'end' => $ord); |
|
928
|
|
|
} |
|
929
|
|
|
} |
|
930
|
|
|
|
|
931
|
|
|
if (isset($current_range['start'])) |
|
932
|
|
|
{ |
|
933
|
|
|
$class_string .= '\\x{' . strtoupper(sprintf('%04s', dechex($current_range['start']))) . '}'; |
|
934
|
|
|
|
|
935
|
|
|
if ($current_range['start'] != $current_range['end']) |
|
936
|
|
|
{ |
|
937
|
|
|
$class_string .= '-\\x{' . strtoupper(sprintf('%04s', dechex($current_range['end']))) . '}'; |
|
938
|
|
|
} |
|
939
|
|
|
} |
|
940
|
|
|
|
|
941
|
|
|
$funcs['utf8_regex_indic']['data'][$char_script][$insc] = $class_string; |
|
942
|
|
|
} |
|
943
|
|
|
|
|
944
|
|
|
ksort($funcs['utf8_regex_indic']['data'][$char_script]); |
|
945
|
|
|
} |
|
946
|
|
|
unset($funcs['utf8_combining_classes']); |
|
947
|
|
|
|
|
948
|
|
|
foreach ($funcs as $func_name => $func_info) |
|
949
|
|
|
{ |
|
950
|
|
|
export_func_to_file($func_name, $func_info); |
|
951
|
|
|
} |
|
952
|
|
|
|
|
953
|
|
|
/** |
|
954
|
|
|
* Updates a Unicode data function in its designated file. |
|
955
|
|
|
* |
|
956
|
|
|
* @param string $func_name The name of the function. |
|
957
|
|
|
* @param array $func_info Info about the function, including its data. |
|
958
|
|
|
*/ |
|
959
|
|
|
function export_func_to_file($func_name, $func_info) |
|
960
|
|
|
{ |
|
961
|
|
|
global $unicodedir; |
|
962
|
|
|
|
|
963
|
|
|
$file_contents = file_get_contents($unicodedir . '/' . $func_info['file']); |
|
964
|
|
|
|
|
965
|
|
|
$func_text = 'function ' . $func_name . '()' . "\n" . '{'; |
|
966
|
|
|
|
|
967
|
|
|
$func_regex = '/' . preg_quote($func_text, '/') . '.+?\n}/s'; |
|
968
|
|
|
|
|
969
|
|
|
$func_text .= "\n\t" . 'return array(' . "\n"; |
|
970
|
|
|
|
|
971
|
|
|
build_func_array($func_text, $func_info['data'], $func_info['key_type'], $func_info['val_type']); |
|
972
|
|
|
|
|
973
|
|
|
$func_text .= "\t" . ');' . "\n" . '}'; |
|
974
|
|
|
|
|
975
|
|
|
$file_contents = preg_replace($func_regex, $func_text, $file_contents); |
|
976
|
|
|
|
|
977
|
|
|
file_put_contents($unicodedir . '/' . $func_info['file'], $file_contents); |
|
978
|
|
|
} |
|
979
|
|
|
|
|
980
|
|
|
/** |
|
981
|
|
|
* Helper for export_func_to_file(). Builds the function's data array. |
|
982
|
|
|
* |
|
983
|
|
|
* @param string &$func_text The raw string that contains function code. |
|
984
|
|
|
* @param array $data Data to format as an array. |
|
985
|
|
|
* @param string $key_type How to format the array keys. |
|
986
|
|
|
* @param string $val_type How to format the array values. |
|
987
|
|
|
*/ |
|
988
|
|
|
function build_func_array(&$func_text, $data, $key_type, $val_type) |
|
989
|
|
|
{ |
|
990
|
|
|
static $indent = 2; |
|
991
|
|
|
|
|
992
|
|
|
foreach ($data as $key => $value) |
|
993
|
|
|
{ |
|
994
|
|
|
$func_text .= str_repeat("\t", $indent); |
|
995
|
|
|
|
|
996
|
|
|
if ($key_type == 'hexchar') |
|
997
|
|
|
{ |
|
998
|
|
|
$func_text .= '"'; |
|
999
|
|
|
|
|
1000
|
|
|
$key = mb_decode_numericentity(str_replace(' ', '', $key), array(0, 0x10FFFF, 0, 0xFFFFFF), 'UTF-8'); |
|
1001
|
|
|
|
|
1002
|
|
|
foreach (unpack('C*', $key) as $byte_value) |
|
1003
|
|
|
{ |
|
1004
|
|
|
$func_text .= '\\x' . strtoupper(dechex($byte_value)); |
|
1005
|
|
|
} |
|
1006
|
|
|
|
|
1007
|
|
|
$func_text .= '" => '; |
|
1008
|
|
|
} |
|
1009
|
|
|
elseif ($key_type == 'string') |
|
1010
|
|
|
{ |
|
1011
|
|
|
$func_text .= var_export($key, true) . ' => '; |
|
1012
|
|
|
} |
|
1013
|
|
|
|
|
1014
|
|
|
if (is_array($value)) |
|
1015
|
|
|
{ |
|
1016
|
|
|
$func_text .= 'array(' . "\n"; |
|
1017
|
|
|
|
|
1018
|
|
|
$indent++; |
|
1019
|
|
|
build_func_array($func_text, $value, $key_type, $val_type); |
|
1020
|
|
|
$indent--; |
|
1021
|
|
|
|
|
1022
|
|
|
$func_text .= str_repeat("\t", $indent) . ')'; |
|
1023
|
|
|
} |
|
1024
|
|
|
elseif ($val_type == 'hexchar') |
|
1025
|
|
|
{ |
|
1026
|
|
|
$func_text .= '"'; |
|
1027
|
|
|
|
|
1028
|
|
|
$value = mb_decode_numericentity(str_replace(' ', '', $value), array(0, 0x10FFFF, 0, 0xFFFFFF), 'UTF-8'); |
|
1029
|
|
|
foreach (unpack('C*', $value) as $byte_value) |
|
1030
|
|
|
{ |
|
1031
|
|
|
$func_text .= '\\x' . strtoupper(dechex($byte_value)); |
|
1032
|
|
|
} |
|
1033
|
|
|
|
|
1034
|
|
|
$func_text .= '"'; |
|
1035
|
|
|
} |
|
1036
|
|
|
elseif ($val_type == 'string') |
|
1037
|
|
|
{ |
|
1038
|
|
|
$func_text .= var_export($value, true); |
|
1039
|
|
|
} |
|
1040
|
|
|
else |
|
1041
|
|
|
{ |
|
1042
|
|
|
$func_text .= $value; |
|
1043
|
|
|
} |
|
1044
|
|
|
|
|
1045
|
|
|
$func_text .= ',' . "\n"; |
|
1046
|
|
|
} |
|
1047
|
|
|
} |
|
1048
|
|
|
|
|
1049
|
|
|
?> |
SimpleMachines /
SMF2.1
Loading history...