|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* webtrees: online genealogy |
|
5
|
|
|
* Copyright (C) 2022 webtrees development team |
|
6
|
|
|
* This program is free software: you can redistribute it and/or modify |
|
7
|
|
|
* it under the terms of the GNU General Public License as published by |
|
8
|
|
|
* the Free Software Foundation, either version 3 of the License, or |
|
9
|
|
|
* (at your option) any later version. |
|
10
|
|
|
* This program is distributed in the hope that it will be useful, |
|
11
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
12
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
13
|
|
|
* GNU General Public License for more details. |
|
14
|
|
|
* You should have received a copy of the GNU General Public License |
|
15
|
|
|
* along with this program. If not, see <https://www.gnu.org/licenses/>. |
|
16
|
|
|
*/ |
|
17
|
|
|
|
|
18
|
|
|
declare(strict_types=1); |
|
19
|
|
|
|
|
20
|
|
|
namespace Fisharebest\Webtrees\Functions; |
|
21
|
|
|
|
|
22
|
|
|
use Fisharebest\Webtrees\I18N; |
|
23
|
|
|
|
|
24
|
|
|
use function str_contains; |
|
25
|
|
|
|
|
26
|
|
|
/** |
|
27
|
|
|
* RTL Functions for use in the PDF/HTML reports |
|
28
|
|
|
*/ |
|
29
|
|
|
class FunctionsRtl |
|
30
|
|
|
{ |
|
31
|
|
|
private const UTF8_LRM = "\xE2\x80\x8E"; // U+200E (Left to Right mark: zero-width character with LTR directionality) |
|
32
|
|
|
private const UTF8_RLM = "\xE2\x80\x8F"; // U+200F (Right to Left mark: zero-width character with RTL directionality) |
|
33
|
|
|
private const UTF8_LRO = "\xE2\x80\xAD"; // U+202D (Left to Right override: force everything following to LTR mode) |
|
34
|
|
|
private const UTF8_RLO = "\xE2\x80\xAE"; // U+202E (Right to Left override: force everything following to RTL mode) |
|
35
|
|
|
private const UTF8_LRE = "\xE2\x80\xAA"; // U+202A (Left to Right embedding: treat everything following as LTR text) |
|
36
|
|
|
private const UTF8_RLE = "\xE2\x80\xAB"; // U+202B (Right to Left embedding: treat everything following as RTL text) |
|
37
|
|
|
private const UTF8_PDF = "\xE2\x80\xAC"; // U+202C (Pop directional formatting: restore state prior to last LRO, RLO, LRE, RLE) |
|
38
|
|
|
|
|
39
|
|
|
private const OPEN_PARENTHESES = '([{'; |
|
40
|
|
|
|
|
41
|
|
|
private const CLOSE_PARENTHESES = ')]}'; |
|
42
|
|
|
|
|
43
|
|
|
private const NUMBERS = '0123456789'; |
|
44
|
|
|
|
|
45
|
|
|
private const NUMBER_PREFIX = '+-'; // Treat these like numbers when at beginning or end of numeric strings |
|
46
|
|
|
|
|
47
|
|
|
private const NUMBER_PUNCTUATION = '- ,.:/'; // Treat these like numbers when inside numeric strings |
|
48
|
|
|
|
|
49
|
|
|
private const PUNCTUATION = ',.:;?!'; |
|
50
|
|
|
|
|
51
|
|
|
// Markup |
|
52
|
|
|
private const START_LTR = '<LTR>'; |
|
53
|
|
|
private const END_LTR = '</LTR>'; |
|
54
|
|
|
private const START_RTL = '<RTL>'; |
|
55
|
|
|
private const END_RTL = '</RTL>'; |
|
56
|
|
|
private const LENGTH_START = 5; |
|
57
|
|
|
private const LENGTH_END = 6; |
|
58
|
|
|
|
|
59
|
|
|
/** @var string Were we previously processing LTR or RTL. */ |
|
60
|
|
|
private static $previousState; |
|
61
|
|
|
|
|
62
|
|
|
/** @var string Are we currently processing LTR or RTL. */ |
|
63
|
|
|
private static $currentState; |
|
64
|
|
|
|
|
65
|
|
|
/** @var string Text waiting to be processed. */ |
|
66
|
|
|
private static $waitingText; |
|
67
|
|
|
|
|
68
|
|
|
/** @var int Offset into the text. */ |
|
69
|
|
|
private static $posSpanStart; |
|
70
|
|
|
|
|
71
|
|
|
/** |
|
72
|
|
|
* This function strips ‎ and ‏ from the input string. It should be used for all |
|
73
|
|
|
* text that has been passed through the PrintReady() function before that text is stored |
|
74
|
|
|
* in the database. The database should NEVER contain these characters. |
|
75
|
|
|
* |
|
76
|
|
|
* @param string $inputText The string from which the ‎ and ‏ characters should be stripped |
|
77
|
|
|
* |
|
78
|
|
|
* @return string The input string, with ‎ and ‏ stripped |
|
79
|
|
|
*/ |
|
80
|
|
|
public static function stripLrmRlm(string $inputText): string |
|
81
|
|
|
{ |
|
82
|
|
|
return str_replace([ |
|
83
|
|
|
self::UTF8_LRM, |
|
84
|
|
|
self::UTF8_RLM, |
|
85
|
|
|
self::UTF8_LRO, |
|
86
|
|
|
self::UTF8_RLO, |
|
87
|
|
|
self::UTF8_LRE, |
|
88
|
|
|
self::UTF8_RLE, |
|
89
|
|
|
self::UTF8_PDF, |
|
90
|
|
|
'‎', |
|
91
|
|
|
'‏', |
|
92
|
|
|
'&LRM;', |
|
93
|
|
|
'&RLM;', |
|
94
|
|
|
], '', $inputText); |
|
95
|
|
|
} |
|
96
|
|
|
|
|
97
|
|
|
/** |
|
98
|
|
|
* This function encapsulates all texts in the input with <span dir='xxx'> and </span> |
|
99
|
|
|
* according to the directionality specified. |
|
100
|
|
|
* |
|
101
|
|
|
* @param string $inputText Raw input |
|
102
|
|
|
* |
|
103
|
|
|
* @return string The string with all texts encapsulated as required |
|
104
|
|
|
*/ |
|
105
|
|
|
public static function spanLtrRtl(string $inputText): string |
|
106
|
|
|
{ |
|
107
|
|
|
if ($inputText === '') { |
|
108
|
|
|
// Nothing to do |
|
109
|
|
|
return ''; |
|
110
|
|
|
} |
|
111
|
|
|
|
|
112
|
|
|
$workingText = str_replace("\n", '<br>', $inputText); |
|
113
|
|
|
$workingText = str_replace([ |
|
114
|
|
|
'<span class="starredname"><br>', |
|
115
|
|
|
'<span<br>class="starredname">', |
|
116
|
|
|
], '<br><span class="starredname">', $workingText); // Reposition some incorrectly placed line breaks |
|
117
|
|
|
$workingText = self::stripLrmRlm($workingText); // Get rid of any existing UTF8 control codes |
|
118
|
|
|
|
|
119
|
|
|
self::$previousState = ''; |
|
120
|
|
|
self::$currentState = strtoupper(I18N::direction()); |
|
121
|
|
|
$numberState = false; // Set when we're inside a numeric string |
|
122
|
|
|
$result = ''; |
|
123
|
|
|
self::$waitingText = ''; |
|
124
|
|
|
$openParDirection = []; |
|
125
|
|
|
|
|
126
|
|
|
self::beginCurrentSpan($result); |
|
127
|
|
|
|
|
128
|
|
|
while ($workingText !== '') { |
|
129
|
|
|
$charArray = self::getChar($workingText, 0); // Get the next ASCII or UTF-8 character |
|
130
|
|
|
$currentLetter = $charArray['letter']; |
|
131
|
|
|
$currentLen = $charArray['length']; |
|
132
|
|
|
|
|
133
|
|
|
$openParIndex = strpos(self::OPEN_PARENTHESES, $currentLetter); // Which opening parenthesis is this? |
|
134
|
|
|
$closeParIndex = strpos(self::CLOSE_PARENTHESES, $currentLetter); // Which closing parenthesis is this? |
|
135
|
|
|
|
|
136
|
|
|
switch ($currentLetter) { |
|
137
|
|
|
case '<': |
|
138
|
|
|
// Assume this '<' starts an HTML element |
|
139
|
|
|
$endPos = strpos($workingText, '>'); // look for the terminating '>' |
|
140
|
|
|
if ($endPos === false) { |
|
141
|
|
|
$endPos = 0; |
|
142
|
|
|
} |
|
143
|
|
|
$currentLen += $endPos; |
|
144
|
|
|
$element = substr($workingText, 0, $currentLen); |
|
145
|
|
|
$temp = strtolower(substr($element, 0, 3)); |
|
146
|
|
|
if (strlen($element) < 7 && $temp === '<br') { |
|
147
|
|
|
if ($numberState) { |
|
148
|
|
|
$numberState = false; |
|
149
|
|
|
if (self::$currentState === 'RTL') { |
|
150
|
|
|
self::$waitingText .= self::UTF8_PDF; |
|
151
|
|
|
} |
|
152
|
|
|
} |
|
153
|
|
|
self::breakCurrentSpan($result); |
|
154
|
|
|
} elseif (self::$waitingText === '') { |
|
155
|
|
|
$result .= $element; |
|
156
|
|
|
} else { |
|
157
|
|
|
self::$waitingText .= $element; |
|
158
|
|
|
} |
|
159
|
|
|
$workingText = substr($workingText, $currentLen); |
|
160
|
|
|
break; |
|
161
|
|
|
case '&': |
|
162
|
|
|
// Assume this '&' starts an HTML entity |
|
163
|
|
|
$endPos = strpos($workingText, ';'); // look for the terminating ';' |
|
164
|
|
|
if ($endPos === false) { |
|
165
|
|
|
$endPos = 0; |
|
166
|
|
|
} |
|
167
|
|
|
$currentLen += $endPos; |
|
168
|
|
|
$entity = substr($workingText, 0, $currentLen); |
|
169
|
|
|
if (strtolower($entity) === ' ') { |
|
170
|
|
|
$entity .= ' '; // Ensure consistent case for this entity |
|
171
|
|
|
} |
|
172
|
|
|
if (self::$waitingText === '') { |
|
173
|
|
|
$result .= $entity; |
|
174
|
|
|
} else { |
|
175
|
|
|
self::$waitingText .= $entity; |
|
176
|
|
|
} |
|
177
|
|
|
$workingText = substr($workingText, $currentLen); |
|
178
|
|
|
break; |
|
179
|
|
|
case '{': |
|
180
|
|
|
if (substr($workingText, 1, 1) === '{') { |
|
181
|
|
|
// Assume this '{{' starts a TCPDF directive |
|
182
|
|
|
$endPos = strpos($workingText, '}}'); // look for the terminating '}}' |
|
183
|
|
|
if ($endPos === false) { |
|
184
|
|
|
$endPos = 0; |
|
185
|
|
|
} |
|
186
|
|
|
$currentLen = $endPos + 2; |
|
187
|
|
|
$directive = substr($workingText, 0, $currentLen); |
|
188
|
|
|
$workingText = substr($workingText, $currentLen); |
|
189
|
|
|
$result .= self::$waitingText . $directive; |
|
190
|
|
|
self::$waitingText = ''; |
|
191
|
|
|
break; |
|
192
|
|
|
} |
|
193
|
|
|
// no break |
|
194
|
|
|
default: |
|
195
|
|
|
// Look for strings of numbers with optional leading or trailing + or - |
|
196
|
|
|
// and with optional embedded numeric punctuation |
|
197
|
|
|
if ($numberState) { |
|
198
|
|
|
// If we're inside a numeric string, look for reasons to end it |
|
199
|
|
|
$offset = 0; // Be sure to look at the current character first |
|
200
|
|
|
$charArray = self::getChar($workingText . "\n", $offset); |
|
201
|
|
|
if (!str_contains(self::NUMBERS, $charArray['letter'])) { |
|
202
|
|
|
// This is not a digit. Is it numeric punctuation? |
|
203
|
|
|
if (substr($workingText . "\n", $offset, 6) === ' ') { |
|
204
|
|
|
$offset += 6; // This could be numeric punctuation |
|
205
|
|
|
} elseif (str_contains(self::NUMBER_PUNCTUATION, $charArray['letter'])) { |
|
206
|
|
|
$offset += $charArray['length']; // This could be numeric punctuation |
|
207
|
|
|
} |
|
208
|
|
|
// If the next character is a digit, the current character is numeric punctuation |
|
209
|
|
|
$charArray = self::getChar($workingText . "\n", $offset); |
|
210
|
|
|
if (!str_contains(self::NUMBERS, $charArray['letter'])) { |
|
211
|
|
|
// This is not a digit. End the run of digits and punctuation. |
|
212
|
|
|
$numberState = false; |
|
213
|
|
|
if (self::$currentState === 'RTL') { |
|
214
|
|
|
if (!str_contains(self::NUMBER_PREFIX, $currentLetter)) { |
|
215
|
|
|
$currentLetter = self::UTF8_PDF . $currentLetter; |
|
216
|
|
|
} else { |
|
217
|
|
|
$currentLetter .= self::UTF8_PDF; // Include a trailing + or - in the run |
|
218
|
|
|
} |
|
219
|
|
|
} |
|
220
|
|
|
} |
|
221
|
|
|
} |
|
222
|
|
|
} else { |
|
223
|
|
|
// If we're outside a numeric string, look for reasons to start it |
|
224
|
|
|
if (str_contains(self::NUMBER_PREFIX, $currentLetter)) { |
|
225
|
|
|
// This might be a number lead-in |
|
226
|
|
|
$offset = $currentLen; |
|
227
|
|
|
$nextChar = substr($workingText . "\n", $offset, 1); |
|
228
|
|
|
if (str_contains(self::NUMBERS, $nextChar)) { |
|
229
|
|
|
$numberState = true; // We found a digit: the lead-in is therefore numeric |
|
230
|
|
|
if (self::$currentState === 'RTL') { |
|
231
|
|
|
$currentLetter = self::UTF8_LRE . $currentLetter; |
|
232
|
|
|
} |
|
233
|
|
|
} |
|
234
|
|
|
} elseif (str_contains(self::NUMBERS, $currentLetter)) { |
|
235
|
|
|
$numberState = true; // The current letter is a digit |
|
236
|
|
|
if (self::$currentState === 'RTL') { |
|
237
|
|
|
$currentLetter = self::UTF8_LRE . $currentLetter; |
|
238
|
|
|
} |
|
239
|
|
|
} |
|
240
|
|
|
} |
|
241
|
|
|
|
|
242
|
|
|
// Determine the directionality of the current UTF-8 character |
|
243
|
|
|
$newState = self::$currentState; |
|
244
|
|
|
|
|
245
|
|
|
while (true) { |
|
246
|
|
|
if (I18N::scriptDirection(I18N::textScript($currentLetter)) === 'rtl') { |
|
247
|
|
|
if (self::$currentState === '') { |
|
248
|
|
|
$newState = 'RTL'; |
|
249
|
|
|
break; |
|
250
|
|
|
} |
|
251
|
|
|
|
|
252
|
|
|
if (self::$currentState === 'RTL') { |
|
253
|
|
|
break; |
|
254
|
|
|
} |
|
255
|
|
|
// Switch to RTL only if this isn't a solitary RTL letter |
|
256
|
|
|
$tempText = substr($workingText, $currentLen); |
|
257
|
|
|
while ($tempText !== '') { |
|
258
|
|
|
$nextCharArray = self::getChar($tempText, 0); |
|
259
|
|
|
$nextLetter = $nextCharArray['letter']; |
|
260
|
|
|
$nextLen = $nextCharArray['length']; |
|
261
|
|
|
$tempText = substr($tempText, $nextLen); |
|
262
|
|
|
|
|
263
|
|
|
if (I18N::scriptDirection(I18N::textScript($nextLetter)) === 'rtl') { |
|
264
|
|
|
$newState = 'RTL'; |
|
265
|
|
|
break 2; |
|
266
|
|
|
} |
|
267
|
|
|
|
|
268
|
|
|
if (str_contains(self::PUNCTUATION, $nextLetter) || str_contains(self::OPEN_PARENTHESES, $nextLetter)) { |
|
269
|
|
|
$newState = 'RTL'; |
|
270
|
|
|
break 2; |
|
271
|
|
|
} |
|
272
|
|
|
|
|
273
|
|
|
if ($nextLetter === ' ') { |
|
274
|
|
|
break; |
|
275
|
|
|
} |
|
276
|
|
|
$nextLetter .= substr($tempText . "\n", 0, 5); |
|
277
|
|
|
if ($nextLetter === ' ') { |
|
278
|
|
|
break; |
|
279
|
|
|
} |
|
280
|
|
|
} |
|
281
|
|
|
// This is a solitary RTL letter : wrap it in UTF8 control codes to force LTR directionality |
|
282
|
|
|
$currentLetter = self::UTF8_LRO . $currentLetter . self::UTF8_PDF; |
|
283
|
|
|
$newState = 'LTR'; |
|
284
|
|
|
break; |
|
285
|
|
|
} |
|
286
|
|
|
if (($currentLen !== 1) || ($currentLetter >= 'A' && $currentLetter <= 'Z') || ($currentLetter >= 'a' && $currentLetter <= 'z')) { |
|
287
|
|
|
// Since it’s neither Hebrew nor Arabic, this UTF-8 character or ASCII letter must be LTR |
|
288
|
|
|
$newState = 'LTR'; |
|
289
|
|
|
break; |
|
290
|
|
|
} |
|
291
|
|
|
if ($closeParIndex !== false) { |
|
292
|
|
|
// This closing parenthesis has to inherit the matching opening parenthesis' directionality |
|
293
|
|
|
if (!empty($openParDirection[$closeParIndex]) && $openParDirection[$closeParIndex] !== '?') { |
|
294
|
|
|
$newState = $openParDirection[$closeParIndex]; |
|
295
|
|
|
} |
|
296
|
|
|
$openParDirection[$closeParIndex] = ''; |
|
297
|
|
|
break; |
|
298
|
|
|
} |
|
299
|
|
|
if ($openParIndex !== false) { |
|
300
|
|
|
// Opening parentheses always inherit the following directionality |
|
301
|
|
|
self::$waitingText .= $currentLetter; |
|
302
|
|
|
$workingText = substr($workingText, $currentLen); |
|
303
|
|
|
while (true) { |
|
304
|
|
|
if ($workingText === '') { |
|
305
|
|
|
break; |
|
306
|
|
|
} |
|
307
|
|
|
if (substr($workingText, 0, 1) === ' ') { |
|
308
|
|
|
// Spaces following this left parenthesis inherit the following directionality too |
|
309
|
|
|
self::$waitingText .= ' '; |
|
310
|
|
|
$workingText = substr($workingText, 1); |
|
311
|
|
|
continue; |
|
312
|
|
|
} |
|
313
|
|
|
if (substr($workingText, 0, 6) === ' ') { |
|
314
|
|
|
// Spaces following this left parenthesis inherit the following directionality too |
|
315
|
|
|
self::$waitingText .= ' '; |
|
316
|
|
|
$workingText = substr($workingText, 6); |
|
317
|
|
|
continue; |
|
318
|
|
|
} |
|
319
|
|
|
break; |
|
320
|
|
|
} |
|
321
|
|
|
$openParDirection[$openParIndex] = '?'; |
|
322
|
|
|
break 2; // double break because we're waiting for more information |
|
323
|
|
|
} |
|
324
|
|
|
|
|
325
|
|
|
// We have a digit or a "normal" special character. |
|
326
|
|
|
// |
|
327
|
|
|
// When this character is not at the start of the input string, it inherits the preceding directionality; |
|
328
|
|
|
// at the start of the input string, it assumes the following directionality. |
|
329
|
|
|
// |
|
330
|
|
|
// Exceptions to this rule will be handled later during final clean-up. |
|
331
|
|
|
// |
|
332
|
|
|
self::$waitingText .= $currentLetter; |
|
333
|
|
|
$workingText = substr($workingText, $currentLen); |
|
334
|
|
|
if (self::$currentState !== '') { |
|
335
|
|
|
$result .= self::$waitingText; |
|
336
|
|
|
self::$waitingText = ''; |
|
337
|
|
|
} |
|
338
|
|
|
break 2; // double break because we're waiting for more information |
|
339
|
|
|
} |
|
340
|
|
|
if ($newState !== self::$currentState) { |
|
341
|
|
|
// A direction change has occurred |
|
342
|
|
|
self::finishCurrentSpan($result); |
|
343
|
|
|
self::$previousState = self::$currentState; |
|
344
|
|
|
self::$currentState = $newState; |
|
345
|
|
|
self::beginCurrentSpan($result); |
|
346
|
|
|
} |
|
347
|
|
|
self::$waitingText .= $currentLetter; |
|
348
|
|
|
$workingText = substr($workingText, $currentLen); |
|
349
|
|
|
$result .= self::$waitingText; |
|
350
|
|
|
self::$waitingText = ''; |
|
351
|
|
|
|
|
352
|
|
|
foreach ($openParDirection as $index => $value) { |
|
353
|
|
|
// Since we now know the proper direction, remember it for all waiting opening parentheses |
|
354
|
|
|
if ($value === '?') { |
|
355
|
|
|
$openParDirection[$index] = self::$currentState; |
|
356
|
|
|
} |
|
357
|
|
|
} |
|
358
|
|
|
|
|
359
|
|
|
break; |
|
360
|
|
|
} |
|
361
|
|
|
} |
|
362
|
|
|
|
|
363
|
|
|
// We're done. Finish last <span> if necessary |
|
364
|
|
|
if ($numberState) { |
|
365
|
|
|
if (self::$waitingText === '') { |
|
366
|
|
|
if (self::$currentState === 'RTL') { |
|
367
|
|
|
$result .= self::UTF8_PDF; |
|
368
|
|
|
} |
|
369
|
|
|
} else { |
|
370
|
|
|
if (self::$currentState === 'RTL') { |
|
371
|
|
|
self::$waitingText .= self::UTF8_PDF; |
|
372
|
|
|
} |
|
373
|
|
|
} |
|
374
|
|
|
} |
|
375
|
|
|
self::finishCurrentSpan($result, true); |
|
376
|
|
|
|
|
377
|
|
|
// Get rid of any waiting text |
|
378
|
|
|
if (self::$waitingText !== '') { |
|
379
|
|
|
if (I18N::direction() === 'rtl' && self::$currentState === 'LTR') { |
|
380
|
|
|
$result .= self::START_RTL; |
|
381
|
|
|
$result .= self::$waitingText; |
|
382
|
|
|
$result .= self::END_RTL; |
|
383
|
|
|
} else { |
|
384
|
|
|
$result .= self::START_LTR; |
|
385
|
|
|
$result .= self::$waitingText; |
|
386
|
|
|
$result .= self::END_LTR; |
|
387
|
|
|
} |
|
388
|
|
|
self::$waitingText = ''; |
|
389
|
|
|
} |
|
390
|
|
|
|
|
391
|
|
|
// Lastly, do some more cleanups |
|
392
|
|
|
|
|
393
|
|
|
// Move leading RTL numeric strings to following LTR text |
|
394
|
|
|
// (this happens when the page direction is RTL and the original text begins with a number and is followed by LTR text) |
|
395
|
|
|
while (substr($result, 0, self::LENGTH_START + 3) === self::START_RTL . self::UTF8_LRE) { |
|
396
|
|
|
$spanEnd = strpos($result, self::END_RTL . self::START_LTR); |
|
397
|
|
|
if ($spanEnd === false) { |
|
398
|
|
|
break; |
|
399
|
|
|
} |
|
400
|
|
|
$textSpan = self::stripLrmRlm(substr($result, self::LENGTH_START + 3, $spanEnd - self::LENGTH_START - 3)); |
|
401
|
|
|
if (I18N::scriptDirection(I18N::textScript($textSpan)) === 'rtl') { |
|
402
|
|
|
break; |
|
403
|
|
|
} |
|
404
|
|
|
$result = self::START_LTR . substr($result, self::LENGTH_START, $spanEnd - self::LENGTH_START) . substr($result, $spanEnd + self::LENGTH_START + self::LENGTH_END); |
|
405
|
|
|
break; |
|
406
|
|
|
} |
|
407
|
|
|
|
|
408
|
|
|
// On RTL pages, put trailing "." in RTL numeric strings into its own RTL span |
|
409
|
|
|
if (I18N::direction() === 'rtl') { |
|
410
|
|
|
$result = str_replace(self::UTF8_PDF . '.' . self::END_RTL, self::UTF8_PDF . self::END_RTL . self::START_RTL . '.' . self::END_RTL, $result); |
|
411
|
|
|
} |
|
412
|
|
|
|
|
413
|
|
|
// Trim trailing blanks preceding <br> in LTR text |
|
414
|
|
|
while (self::$previousState !== 'RTL') { |
|
415
|
|
|
if (str_contains($result, ' <LTRbr>')) { |
|
416
|
|
|
$result = str_replace(' <LTRbr>', '<LTRbr>', $result); |
|
417
|
|
|
continue; |
|
418
|
|
|
} |
|
419
|
|
|
if (str_contains($result, ' <LTRbr>')) { |
|
420
|
|
|
$result = str_replace(' <LTRbr>', '<LTRbr>', $result); |
|
421
|
|
|
continue; |
|
422
|
|
|
} |
|
423
|
|
|
if (str_contains($result, ' <br>')) { |
|
424
|
|
|
$result = str_replace(' <br>', '<br>', $result); |
|
425
|
|
|
continue; |
|
426
|
|
|
} |
|
427
|
|
|
if (str_contains($result, ' <br>')) { |
|
428
|
|
|
$result = str_replace(' <br>', '<br>', $result); |
|
429
|
|
|
continue; |
|
430
|
|
|
} |
|
431
|
|
|
break; // Neither space nor : we're done |
|
432
|
|
|
} |
|
433
|
|
|
|
|
434
|
|
|
// Trim trailing blanks preceding <br> in RTL text |
|
435
|
|
|
while (true) { |
|
436
|
|
|
if (str_contains($result, ' <RTLbr>')) { |
|
437
|
|
|
$result = str_replace(' <RTLbr>', '<RTLbr>', $result); |
|
438
|
|
|
continue; |
|
439
|
|
|
} |
|
440
|
|
|
if (str_contains($result, ' <RTLbr>')) { |
|
441
|
|
|
$result = str_replace(' <RTLbr>', '<RTLbr>', $result); |
|
442
|
|
|
continue; |
|
443
|
|
|
} |
|
444
|
|
|
break; // Neither space nor : we're done |
|
445
|
|
|
} |
|
446
|
|
|
|
|
447
|
|
|
// Convert '<LTRbr>' and '<RTLbr' |
|
448
|
|
|
$result = str_replace([ |
|
449
|
|
|
'<LTRbr>', |
|
450
|
|
|
'<RTLbr>', |
|
451
|
|
|
], [ |
|
452
|
|
|
self::END_LTR . '<br>' . self::START_LTR, |
|
453
|
|
|
self::END_RTL . '<br>' . self::START_RTL, |
|
454
|
|
|
], $result); |
|
455
|
|
|
|
|
456
|
|
|
// Include leading indeterminate directional text in whatever follows |
|
457
|
|
|
if (substr($result . "\n", 0, self::LENGTH_START) !== self::START_LTR && substr($result . "\n", 0, self::LENGTH_START) !== self::START_RTL && substr($result . "\n", 0, 4) !== '<br>') { |
|
458
|
|
|
$leadingText = ''; |
|
459
|
|
|
while (true) { |
|
460
|
|
|
if ($result === '') { |
|
461
|
|
|
$result = $leadingText; |
|
462
|
|
|
break; |
|
463
|
|
|
} |
|
464
|
|
|
if (substr($result . "\n", 0, self::LENGTH_START) !== self::START_LTR && substr($result . "\n", 0, self::LENGTH_START) !== self::START_RTL) { |
|
465
|
|
|
$leadingText .= substr($result, 0, 1); |
|
466
|
|
|
$result = substr($result, 1); |
|
467
|
|
|
continue; |
|
468
|
|
|
} |
|
469
|
|
|
$result = substr($result, 0, self::LENGTH_START) . $leadingText . substr($result, self::LENGTH_START); |
|
470
|
|
|
break; |
|
471
|
|
|
} |
|
472
|
|
|
} |
|
473
|
|
|
|
|
474
|
|
|
// Include solitary "-" and "+" in surrounding RTL text |
|
475
|
|
|
$result = str_replace([ |
|
476
|
|
|
self::END_RTL . self::START_LTR . '-' . self::END_LTR . self::START_RTL, |
|
477
|
|
|
self::END_RTL . self::START_LTR . '-' . self::END_LTR . self::START_RTL, |
|
478
|
|
|
], [ |
|
479
|
|
|
'-', |
|
480
|
|
|
'+', |
|
481
|
|
|
], $result); |
|
482
|
|
|
|
|
483
|
|
|
// Remove empty spans |
|
484
|
|
|
$result = str_replace([ |
|
485
|
|
|
self::START_LTR . self::END_LTR, |
|
486
|
|
|
self::START_RTL . self::END_RTL, |
|
487
|
|
|
], '', $result); |
|
488
|
|
|
|
|
489
|
|
|
// Finally, correct '<LTR>', '</LTR>', '<RTL>', and '</RTL>' |
|
490
|
|
|
// LTR text: <span dir="ltr"> text </span> |
|
491
|
|
|
// RTL text: <span dir="rtl"> text </span> |
|
492
|
|
|
|
|
493
|
|
|
$result = str_replace([ |
|
494
|
|
|
self::START_LTR, |
|
495
|
|
|
self::END_LTR, |
|
496
|
|
|
self::START_RTL, |
|
497
|
|
|
self::END_RTL, |
|
498
|
|
|
], [ |
|
499
|
|
|
'<span dir="ltr">', |
|
500
|
|
|
'</span>', |
|
501
|
|
|
'<span dir="rtl">', |
|
502
|
|
|
'</span>', |
|
503
|
|
|
], $result); |
|
504
|
|
|
|
|
505
|
|
|
return $result; |
|
506
|
|
|
} |
|
507
|
|
|
|
|
508
|
|
|
/** |
|
509
|
|
|
* Wrap words that have an asterisk suffix in <u> and </u> tags. |
|
510
|
|
|
* This should underline starred names to show the preferred name. |
|
511
|
|
|
* |
|
512
|
|
|
* @param string $textSpan |
|
513
|
|
|
* @param string $direction |
|
514
|
|
|
* |
|
515
|
|
|
* @return string |
|
516
|
|
|
*/ |
|
517
|
|
|
public static function starredName(string $textSpan, string $direction): string |
|
518
|
|
|
{ |
|
519
|
|
|
// To avoid a TCPDF bug that mixes up the word order, insert those <u> and </u> tags |
|
520
|
|
|
// only when page and span directions are identical. |
|
521
|
|
|
if ($direction === strtoupper(I18N::direction())) { |
|
522
|
|
|
while (true) { |
|
523
|
|
|
$starPos = strpos($textSpan, '*'); |
|
524
|
|
|
if ($starPos === false) { |
|
525
|
|
|
break; |
|
526
|
|
|
} |
|
527
|
|
|
$trailingText = substr($textSpan, $starPos + 1); |
|
528
|
|
|
$textSpan = substr($textSpan, 0, $starPos); |
|
529
|
|
|
$wordStart = strrpos($textSpan, ' '); // Find the start of the word |
|
530
|
|
|
if ($wordStart !== false) { |
|
531
|
|
|
$leadingText = substr($textSpan, 0, $wordStart + 1); |
|
532
|
|
|
$wordText = substr($textSpan, $wordStart + 1); |
|
533
|
|
|
} else { |
|
534
|
|
|
$leadingText = ''; |
|
535
|
|
|
$wordText = $textSpan; |
|
536
|
|
|
} |
|
537
|
|
|
$textSpan = $leadingText . '<u>' . $wordText . '</u>' . $trailingText; |
|
538
|
|
|
} |
|
539
|
|
|
$textSpan = preg_replace('~<span class="starredname">(.*)</span>~', '<u>\1</u>', $textSpan); |
|
540
|
|
|
// The is a work-around for a TCPDF bug eating blanks. |
|
541
|
|
|
$textSpan = str_replace([ |
|
542
|
|
|
' <u>', |
|
543
|
|
|
'</u> ', |
|
544
|
|
|
], [ |
|
545
|
|
|
' <u>', |
|
546
|
|
|
'</u> ', |
|
547
|
|
|
], $textSpan); |
|
548
|
|
|
} else { |
|
549
|
|
|
// Text and page directions differ: remove the <span> and </span> |
|
550
|
|
|
$textSpan = preg_replace('~(.*)\*~', '\1', $textSpan); |
|
551
|
|
|
$textSpan = preg_replace('~<span class="starredname">(.*)</span>~', '\1', $textSpan); |
|
552
|
|
|
} |
|
553
|
|
|
|
|
554
|
|
|
return $textSpan; |
|
555
|
|
|
} |
|
556
|
|
|
|
|
557
|
|
|
/** |
|
558
|
|
|
* Get the next character from an input string |
|
559
|
|
|
* |
|
560
|
|
|
* @param string $text |
|
561
|
|
|
* @param int $offset |
|
562
|
|
|
* |
|
563
|
|
|
* @return array{'letter':string,'length':int} |
|
564
|
|
|
*/ |
|
565
|
|
|
public static function getChar(string $text, int $offset): array |
|
566
|
|
|
{ |
|
567
|
|
|
if ($text === '') { |
|
568
|
|
|
return [ |
|
569
|
|
|
'letter' => '', |
|
570
|
|
|
'length' => 0, |
|
571
|
|
|
]; |
|
572
|
|
|
} |
|
573
|
|
|
|
|
574
|
|
|
$char = substr($text, $offset, 1); |
|
575
|
|
|
$length = 1; |
|
576
|
|
|
if ((ord($char) & 0xE0) === 0xC0) { |
|
577
|
|
|
$length = 2; |
|
578
|
|
|
} |
|
579
|
|
|
if ((ord($char) & 0xF0) === 0xE0) { |
|
580
|
|
|
$length = 3; |
|
581
|
|
|
} |
|
582
|
|
|
if ((ord($char) & 0xF8) === 0xF0) { |
|
583
|
|
|
$length = 4; |
|
584
|
|
|
} |
|
585
|
|
|
$letter = substr($text, $offset, $length); |
|
586
|
|
|
|
|
587
|
|
|
return [ |
|
588
|
|
|
'letter' => $letter, |
|
589
|
|
|
'length' => $length, |
|
590
|
|
|
]; |
|
591
|
|
|
} |
|
592
|
|
|
|
|
593
|
|
|
/** |
|
594
|
|
|
* Insert <br> into current span |
|
595
|
|
|
* |
|
596
|
|
|
* @param string $result |
|
597
|
|
|
* |
|
598
|
|
|
* @return void |
|
599
|
|
|
*/ |
|
600
|
|
|
public static function breakCurrentSpan(string &$result): void |
|
601
|
|
|
{ |
|
602
|
|
|
// Interrupt the current span, insert that <br>, and then continue the current span |
|
603
|
|
|
$result .= self::$waitingText; |
|
604
|
|
|
self::$waitingText = ''; |
|
605
|
|
|
|
|
606
|
|
|
$breakString = '<' . self::$currentState . 'br>'; |
|
607
|
|
|
$result .= $breakString; |
|
608
|
|
|
} |
|
609
|
|
|
|
|
610
|
|
|
/** |
|
611
|
|
|
* Begin current span |
|
612
|
|
|
* |
|
613
|
|
|
* @param string $result |
|
614
|
|
|
* |
|
615
|
|
|
* @return void |
|
616
|
|
|
*/ |
|
617
|
|
|
public static function beginCurrentSpan(string &$result): void |
|
618
|
|
|
{ |
|
619
|
|
|
if (self::$currentState === 'LTR') { |
|
620
|
|
|
$result .= self::START_LTR; |
|
621
|
|
|
} |
|
622
|
|
|
if (self::$currentState === 'RTL') { |
|
623
|
|
|
$result .= self::START_RTL; |
|
624
|
|
|
} |
|
625
|
|
|
|
|
626
|
|
|
self::$posSpanStart = strlen($result); |
|
627
|
|
|
} |
|
628
|
|
|
|
|
629
|
|
|
/** |
|
630
|
|
|
* Finish current span |
|
631
|
|
|
* |
|
632
|
|
|
* @param string $result |
|
633
|
|
|
* @param bool $theEnd |
|
634
|
|
|
* |
|
635
|
|
|
* @return void |
|
636
|
|
|
*/ |
|
637
|
|
|
public static function finishCurrentSpan(string &$result, bool $theEnd = false): void |
|
638
|
|
|
{ |
|
639
|
|
|
$textSpan = substr($result, self::$posSpanStart); |
|
640
|
|
|
$result = substr($result, 0, self::$posSpanStart); |
|
641
|
|
|
|
|
642
|
|
|
// Get rid of empty spans, so that our check for presence of RTL will work |
|
643
|
|
|
$result = str_replace([ |
|
644
|
|
|
self::START_LTR . self::END_LTR, |
|
645
|
|
|
self::START_RTL . self::END_RTL, |
|
646
|
|
|
], '', $result); |
|
647
|
|
|
|
|
648
|
|
|
// Look for numeric strings that are times (hh:mm:ss). These have to be separated from surrounding numbers. |
|
649
|
|
|
$tempResult = ''; |
|
650
|
|
|
while ($textSpan !== '') { |
|
651
|
|
|
$posColon = strpos($textSpan, ':'); |
|
652
|
|
|
if ($posColon === false) { |
|
653
|
|
|
break; |
|
654
|
|
|
} // No more possible time strings |
|
655
|
|
|
$posLRE = strpos($textSpan, self::UTF8_LRE); |
|
656
|
|
|
if ($posLRE === false) { |
|
657
|
|
|
break; |
|
658
|
|
|
} // No more numeric strings |
|
659
|
|
|
$posPDF = strpos($textSpan, self::UTF8_PDF, $posLRE); |
|
660
|
|
|
if ($posPDF === false) { |
|
661
|
|
|
break; |
|
662
|
|
|
} // No more numeric strings |
|
663
|
|
|
|
|
664
|
|
|
$tempResult .= substr($textSpan, 0, $posLRE + 3); // Copy everything preceding the numeric string |
|
665
|
|
|
$numericString = substr($textSpan, $posLRE + 3, $posPDF - $posLRE); // Separate the entire numeric string |
|
666
|
|
|
$textSpan = substr($textSpan, $posPDF + 3); |
|
667
|
|
|
$posColon = strpos($numericString, ':'); |
|
668
|
|
|
if ($posColon === false) { |
|
669
|
|
|
// Nothing that looks like a time here |
|
670
|
|
|
$tempResult .= $numericString; |
|
671
|
|
|
continue; |
|
672
|
|
|
} |
|
673
|
|
|
$posBlank = strpos($numericString . ' ', ' '); |
|
674
|
|
|
$posNbsp = strpos($numericString . ' ', ' '); |
|
675
|
|
|
if ($posBlank < $posNbsp) { |
|
676
|
|
|
$posSeparator = $posBlank; |
|
677
|
|
|
$lengthSeparator = 1; |
|
678
|
|
|
} else { |
|
679
|
|
|
$posSeparator = $posNbsp; |
|
680
|
|
|
$lengthSeparator = 6; |
|
681
|
|
|
} |
|
682
|
|
|
if ($posColon > $posSeparator) { |
|
683
|
|
|
// We have a time string preceded by a blank: Exclude that blank from the numeric string |
|
684
|
|
|
$tempResult .= substr($numericString, 0, $posSeparator); |
|
685
|
|
|
$tempResult .= self::UTF8_PDF; |
|
686
|
|
|
$tempResult .= substr($numericString, $posSeparator, $lengthSeparator); |
|
687
|
|
|
$tempResult .= self::UTF8_LRE; |
|
688
|
|
|
$numericString = substr($numericString, $posSeparator + $lengthSeparator); |
|
689
|
|
|
} |
|
690
|
|
|
|
|
691
|
|
|
$posBlank = strpos($numericString, ' '); |
|
692
|
|
|
$posNbsp = strpos($numericString, ' '); |
|
693
|
|
|
if ($posBlank === false && $posNbsp === false) { |
|
694
|
|
|
// The time string isn't followed by a blank |
|
695
|
|
|
$textSpan = $numericString . $textSpan; |
|
696
|
|
|
continue; |
|
697
|
|
|
} |
|
698
|
|
|
|
|
699
|
|
|
// We have a time string followed by a blank: Exclude that blank from the numeric string |
|
700
|
|
|
if ($posBlank === false) { |
|
701
|
|
|
$posSeparator = $posNbsp; |
|
702
|
|
|
$lengthSeparator = 6; |
|
703
|
|
|
} elseif ($posNbsp === false) { |
|
704
|
|
|
$posSeparator = $posBlank; |
|
705
|
|
|
$lengthSeparator = 1; |
|
706
|
|
|
} elseif ($posBlank < $posNbsp) { |
|
707
|
|
|
$posSeparator = $posBlank; |
|
708
|
|
|
$lengthSeparator = 1; |
|
709
|
|
|
} else { |
|
710
|
|
|
$posSeparator = $posNbsp; |
|
711
|
|
|
$lengthSeparator = 6; |
|
712
|
|
|
} |
|
713
|
|
|
$tempResult .= substr($numericString, 0, $posSeparator); |
|
714
|
|
|
$tempResult .= self::UTF8_PDF; |
|
715
|
|
|
$tempResult .= substr($numericString, $posSeparator, $lengthSeparator); |
|
716
|
|
|
$posSeparator += $lengthSeparator; |
|
717
|
|
|
$numericString = substr($numericString, $posSeparator); |
|
718
|
|
|
$textSpan = self::UTF8_LRE . $numericString . $textSpan; |
|
719
|
|
|
} |
|
720
|
|
|
$textSpan = $tempResult . $textSpan; |
|
721
|
|
|
$trailingBlanks = ''; |
|
722
|
|
|
$trailingBreaks = ''; |
|
723
|
|
|
|
|
724
|
|
|
/* ****************************** LTR text handling ******************************** */ |
|
725
|
|
|
|
|
726
|
|
|
if (self::$currentState === 'LTR') { |
|
727
|
|
|
// Move trailing numeric strings to the following RTL text. Include any blanks preceding or following the numeric text too. |
|
728
|
|
|
if (I18N::direction() === 'rtl' && self::$previousState === 'RTL' && !$theEnd) { |
|
729
|
|
|
$trailingString = ''; |
|
730
|
|
|
$savedSpan = $textSpan; |
|
731
|
|
|
while ($textSpan !== '') { |
|
732
|
|
|
// Look for trailing spaces and tentatively move them |
|
733
|
|
|
if (substr($textSpan, -1) === ' ') { |
|
734
|
|
|
$trailingString = ' ' . $trailingString; |
|
735
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
736
|
|
|
continue; |
|
737
|
|
|
} |
|
738
|
|
|
if (substr($textSpan, -6) === ' ') { |
|
739
|
|
|
$trailingString = ' ' . $trailingString; |
|
740
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
741
|
|
|
continue; |
|
742
|
|
|
} |
|
743
|
|
|
if (substr($textSpan, -3) !== self::UTF8_PDF) { |
|
744
|
|
|
// There is no trailing numeric string |
|
745
|
|
|
$textSpan = $savedSpan; |
|
746
|
|
|
break; |
|
747
|
|
|
} |
|
748
|
|
|
|
|
749
|
|
|
// We have a numeric string |
|
750
|
|
|
$posStartNumber = strrpos($textSpan, self::UTF8_LRE); |
|
751
|
|
|
if ($posStartNumber === false) { |
|
752
|
|
|
$posStartNumber = 0; |
|
753
|
|
|
} |
|
754
|
|
|
$trailingString = substr($textSpan, $posStartNumber) . $trailingString; |
|
755
|
|
|
$textSpan = substr($textSpan, 0, $posStartNumber); |
|
756
|
|
|
|
|
757
|
|
|
// Look for more spaces and move them too |
|
758
|
|
|
while ($textSpan !== '') { |
|
759
|
|
|
if (substr($textSpan, -1) === ' ') { |
|
760
|
|
|
$trailingString = ' ' . $trailingString; |
|
761
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
762
|
|
|
continue; |
|
763
|
|
|
} |
|
764
|
|
|
if (substr($textSpan, -6) === ' ') { |
|
765
|
|
|
$trailingString = ' ' . $trailingString; |
|
766
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
767
|
|
|
continue; |
|
768
|
|
|
} |
|
769
|
|
|
break; |
|
770
|
|
|
} |
|
771
|
|
|
|
|
772
|
|
|
self::$waitingText = $trailingString . self::$waitingText; |
|
773
|
|
|
break; |
|
774
|
|
|
} |
|
775
|
|
|
} |
|
776
|
|
|
|
|
777
|
|
|
$savedSpan = $textSpan; |
|
778
|
|
|
// Move any trailing <br>, optionally preceded or followed by blanks, outside this LTR span |
|
779
|
|
|
while ($textSpan !== '') { |
|
780
|
|
|
if (substr($textSpan, -1) === ' ') { |
|
781
|
|
|
$trailingBlanks = ' ' . $trailingBlanks; |
|
782
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
783
|
|
|
continue; |
|
784
|
|
|
} |
|
785
|
|
|
if (substr('......' . $textSpan, -6) === ' ') { |
|
786
|
|
|
$trailingBlanks = ' ' . $trailingBlanks; |
|
787
|
|
|
$textSpan = substr($textSpan, 0, -6); |
|
788
|
|
|
continue; |
|
789
|
|
|
} |
|
790
|
|
|
break; |
|
791
|
|
|
} |
|
792
|
|
|
while (substr($textSpan, -7) === '<LTRbr>') { |
|
793
|
|
|
$trailingBreaks = '<br>' . $trailingBreaks; // Plain <br> because it’s outside a span |
|
794
|
|
|
$textSpan = substr($textSpan, 0, -7); |
|
795
|
|
|
} |
|
796
|
|
|
if ($trailingBreaks !== '') { |
|
797
|
|
|
while ($textSpan !== '') { |
|
798
|
|
|
if (substr($textSpan, -1) === ' ') { |
|
799
|
|
|
$trailingBreaks = ' ' . $trailingBreaks; |
|
800
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
801
|
|
|
continue; |
|
802
|
|
|
} |
|
803
|
|
|
if (substr($textSpan, -6) === ' ') { |
|
804
|
|
|
$trailingBreaks = ' ' . $trailingBreaks; |
|
805
|
|
|
$textSpan = substr($textSpan, 0, -6); |
|
806
|
|
|
continue; |
|
807
|
|
|
} |
|
808
|
|
|
break; |
|
809
|
|
|
} |
|
810
|
|
|
self::$waitingText = $trailingBlanks . self::$waitingText; // Put those trailing blanks inside the following span |
|
811
|
|
|
} else { |
|
812
|
|
|
$textSpan = $savedSpan; |
|
813
|
|
|
} |
|
814
|
|
|
|
|
815
|
|
|
$trailingBlanks = ''; |
|
816
|
|
|
$trailingPunctuation = ''; |
|
817
|
|
|
$trailingID = ''; |
|
818
|
|
|
$trailingSeparator = ''; |
|
819
|
|
|
$leadingSeparator = ''; |
|
820
|
|
|
|
|
821
|
|
|
while (I18N::direction() === 'rtl') { |
|
822
|
|
|
if (str_contains($result, self::START_RTL)) { |
|
823
|
|
|
// Remove trailing blanks for inclusion in a separate LTR span |
|
824
|
|
|
while ($textSpan !== '') { |
|
825
|
|
|
if (substr($textSpan, -1) === ' ') { |
|
826
|
|
|
$trailingBlanks = ' ' . $trailingBlanks; |
|
827
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
828
|
|
|
continue; |
|
829
|
|
|
} |
|
830
|
|
|
if (substr($textSpan, -6) === ' ') { |
|
831
|
|
|
$trailingBlanks = ' ' . $trailingBlanks; |
|
832
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
833
|
|
|
continue; |
|
834
|
|
|
} |
|
835
|
|
|
break; |
|
836
|
|
|
} |
|
837
|
|
|
|
|
838
|
|
|
// Remove trailing punctuation for inclusion in a separate LTR span |
|
839
|
|
|
if ($textSpan === '') { |
|
840
|
|
|
$trailingChar = "\n"; |
|
841
|
|
|
} else { |
|
842
|
|
|
$trailingChar = substr($textSpan, -1); |
|
843
|
|
|
} |
|
844
|
|
|
if (str_contains(self::PUNCTUATION, $trailingChar)) { |
|
845
|
|
|
$trailingPunctuation = $trailingChar; |
|
846
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
847
|
|
|
} |
|
848
|
|
|
} |
|
849
|
|
|
|
|
850
|
|
|
// Remove trailing ID numbers that look like "(xnnn)" for inclusion in a separate LTR span |
|
851
|
|
|
while (true) { |
|
852
|
|
|
if (substr($textSpan, -1) !== ')') { |
|
853
|
|
|
break; |
|
854
|
|
|
} // There is no trailing ')' |
|
855
|
|
|
$posLeftParen = strrpos($textSpan, '('); |
|
856
|
|
|
if ($posLeftParen === false) { |
|
857
|
|
|
break; |
|
858
|
|
|
} // There is no leading '(' |
|
859
|
|
|
$temp = self::stripLrmRlm(substr($textSpan, $posLeftParen)); // Get rid of UTF8 control codes |
|
860
|
|
|
|
|
861
|
|
|
// If the parenthesized text doesn't look like an ID number, |
|
862
|
|
|
// we don't want to touch it. |
|
863
|
|
|
// This check won’t work if somebody uses ID numbers with an unusual format. |
|
864
|
|
|
$offset = 1; |
|
865
|
|
|
$charArray = self::getChar($temp, $offset); // Get 1st character of parenthesized text |
|
866
|
|
|
if (str_contains(self::NUMBERS, $charArray['letter'])) { |
|
867
|
|
|
break; |
|
868
|
|
|
} |
|
869
|
|
|
$offset += $charArray['length']; // Point at 2nd character of parenthesized text |
|
870
|
|
|
if (!str_contains(self::NUMBERS, substr($temp, $offset, 1))) { |
|
871
|
|
|
break; |
|
872
|
|
|
} |
|
873
|
|
|
// 1st character of parenthesized text is alpha, 2nd character is a digit; last has to be a digit too |
|
874
|
|
|
if (!str_contains(self::NUMBERS, substr($temp, -2, 1))) { |
|
875
|
|
|
break; |
|
876
|
|
|
} |
|
877
|
|
|
|
|
878
|
|
|
$trailingID = substr($textSpan, $posLeftParen); |
|
879
|
|
|
$textSpan = substr($textSpan, 0, $posLeftParen); |
|
880
|
|
|
break; |
|
881
|
|
|
} |
|
882
|
|
|
|
|
883
|
|
|
// Look for " - " or blank preceding the ID number and remove it for inclusion in a separate LTR span |
|
884
|
|
|
if ($trailingID !== '') { |
|
885
|
|
|
while ($textSpan !== '') { |
|
886
|
|
|
if (substr($textSpan, -1) === ' ') { |
|
887
|
|
|
$trailingSeparator = ' ' . $trailingSeparator; |
|
888
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
889
|
|
|
continue; |
|
890
|
|
|
} |
|
891
|
|
|
if (substr($textSpan, -6) === ' ') { |
|
892
|
|
|
$trailingSeparator = ' ' . $trailingSeparator; |
|
893
|
|
|
$textSpan = substr($textSpan, 0, -6); |
|
894
|
|
|
continue; |
|
895
|
|
|
} |
|
896
|
|
|
if (substr($textSpan, -1) === '-') { |
|
897
|
|
|
$trailingSeparator = '-' . $trailingSeparator; |
|
898
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
899
|
|
|
continue; |
|
900
|
|
|
} |
|
901
|
|
|
break; |
|
902
|
|
|
} |
|
903
|
|
|
} |
|
904
|
|
|
|
|
905
|
|
|
// Look for " - " preceding the text and remove it for inclusion in a separate LTR span |
|
906
|
|
|
$foundSeparator = false; |
|
907
|
|
|
$savedSpan = $textSpan; |
|
908
|
|
|
while ($textSpan !== '') { |
|
909
|
|
|
if (substr($textSpan, 0, 1) === ' ') { |
|
910
|
|
|
$leadingSeparator = ' ' . $leadingSeparator; |
|
911
|
|
|
$textSpan = substr($textSpan, 1); |
|
912
|
|
|
continue; |
|
913
|
|
|
} |
|
914
|
|
|
if (substr($textSpan, 0, 6) === ' ') { |
|
915
|
|
|
$leadingSeparator = ' ' . $leadingSeparator; |
|
916
|
|
|
$textSpan = substr($textSpan, 6); |
|
917
|
|
|
continue; |
|
918
|
|
|
} |
|
919
|
|
|
if (substr($textSpan, 0, 1) === '-') { |
|
920
|
|
|
$leadingSeparator = '-' . $leadingSeparator; |
|
921
|
|
|
$textSpan = substr($textSpan, 1); |
|
922
|
|
|
$foundSeparator = true; |
|
923
|
|
|
continue; |
|
924
|
|
|
} |
|
925
|
|
|
break; |
|
926
|
|
|
} |
|
927
|
|
|
if (!$foundSeparator) { |
|
928
|
|
|
$textSpan = $savedSpan; |
|
929
|
|
|
$leadingSeparator = ''; |
|
930
|
|
|
} |
|
931
|
|
|
break; |
|
932
|
|
|
} |
|
933
|
|
|
|
|
934
|
|
|
// We're done: finish the span |
|
935
|
|
|
$textSpan = self::starredName($textSpan, 'LTR'); // Wrap starred name in <u> and </u> tags |
|
936
|
|
|
while (true) { |
|
937
|
|
|
// Remove blanks that precede <LTRbr> |
|
938
|
|
|
if (str_contains($textSpan, ' <LTRbr>')) { |
|
939
|
|
|
$textSpan = str_replace(' <LTRbr>', '<LTRbr>', $textSpan); |
|
940
|
|
|
continue; |
|
941
|
|
|
} |
|
942
|
|
|
if (str_contains($textSpan, ' <LTRbr>')) { |
|
943
|
|
|
$textSpan = str_replace(' <LTRbr>', '<LTRbr>', $textSpan); |
|
944
|
|
|
continue; |
|
945
|
|
|
} |
|
946
|
|
|
break; |
|
947
|
|
|
} |
|
948
|
|
|
if ($leadingSeparator !== '') { |
|
949
|
|
|
$result .= self::START_LTR . $leadingSeparator . self::END_LTR; |
|
950
|
|
|
} |
|
951
|
|
|
$result .= $textSpan . self::END_LTR; |
|
952
|
|
|
if ($trailingSeparator !== '') { |
|
953
|
|
|
$result .= self::START_LTR . $trailingSeparator . self::END_LTR; |
|
954
|
|
|
} |
|
955
|
|
|
if ($trailingID !== '') { |
|
956
|
|
|
$result .= self::START_LTR . $trailingID . self::END_LTR; |
|
957
|
|
|
} |
|
958
|
|
|
if ($trailingPunctuation !== '') { |
|
959
|
|
|
$result .= self::START_LTR . $trailingPunctuation . self::END_LTR; |
|
960
|
|
|
} |
|
961
|
|
|
if ($trailingBlanks !== '') { |
|
962
|
|
|
$result .= self::START_LTR . $trailingBlanks . self::END_LTR; |
|
963
|
|
|
} |
|
964
|
|
|
} |
|
965
|
|
|
|
|
966
|
|
|
/* ****************************** RTL text handling ******************************** */ |
|
967
|
|
|
|
|
968
|
|
|
if (self::$currentState === 'RTL') { |
|
969
|
|
|
$savedSpan = $textSpan; |
|
970
|
|
|
|
|
971
|
|
|
// Move any trailing <br>, optionally followed by blanks, outside this RTL span |
|
972
|
|
|
while ($textSpan !== '') { |
|
973
|
|
|
if (substr($textSpan, -1) === ' ') { |
|
974
|
|
|
$trailingBlanks = ' ' . $trailingBlanks; |
|
975
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
976
|
|
|
continue; |
|
977
|
|
|
} |
|
978
|
|
|
if (substr('......' . $textSpan, -6) === ' ') { |
|
979
|
|
|
$trailingBlanks = ' ' . $trailingBlanks; |
|
980
|
|
|
$textSpan = substr($textSpan, 0, -6); |
|
981
|
|
|
continue; |
|
982
|
|
|
} |
|
983
|
|
|
break; |
|
984
|
|
|
} |
|
985
|
|
|
while (substr($textSpan, -7) === '<RTLbr>') { |
|
986
|
|
|
$trailingBreaks = '<br>' . $trailingBreaks; // Plain <br> because it’s outside a span |
|
987
|
|
|
$textSpan = substr($textSpan, 0, -7); |
|
988
|
|
|
} |
|
989
|
|
|
if ($trailingBreaks !== '') { |
|
990
|
|
|
self::$waitingText = $trailingBlanks . self::$waitingText; // Put those trailing blanks inside the following span |
|
991
|
|
|
} else { |
|
992
|
|
|
$textSpan = $savedSpan; |
|
993
|
|
|
} |
|
994
|
|
|
|
|
995
|
|
|
// Move trailing numeric strings to the following LTR text. Include any blanks preceding or following the numeric text too. |
|
996
|
|
|
if (!$theEnd && I18N::direction() !== 'rtl') { |
|
997
|
|
|
$trailingString = ''; |
|
998
|
|
|
$savedSpan = $textSpan; |
|
999
|
|
|
while ($textSpan !== '') { |
|
1000
|
|
|
// Look for trailing spaces and tentatively move them |
|
1001
|
|
|
if (substr($textSpan, -1) === ' ') { |
|
1002
|
|
|
$trailingString = ' ' . $trailingString; |
|
1003
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
1004
|
|
|
continue; |
|
1005
|
|
|
} |
|
1006
|
|
|
if (substr($textSpan, -6) === ' ') { |
|
1007
|
|
|
$trailingString = ' ' . $trailingString; |
|
1008
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
1009
|
|
|
continue; |
|
1010
|
|
|
} |
|
1011
|
|
|
if (substr($textSpan, -3) !== self::UTF8_PDF) { |
|
1012
|
|
|
// There is no trailing numeric string |
|
1013
|
|
|
$textSpan = $savedSpan; |
|
1014
|
|
|
break; |
|
1015
|
|
|
} |
|
1016
|
|
|
|
|
1017
|
|
|
// We have a numeric string |
|
1018
|
|
|
$posStartNumber = strrpos($textSpan, self::UTF8_LRE); |
|
1019
|
|
|
if ($posStartNumber === false) { |
|
1020
|
|
|
$posStartNumber = 0; |
|
1021
|
|
|
} |
|
1022
|
|
|
$trailingString = substr($textSpan, $posStartNumber) . $trailingString; |
|
1023
|
|
|
$textSpan = substr($textSpan, 0, $posStartNumber); |
|
1024
|
|
|
|
|
1025
|
|
|
// Look for more spaces and move them too |
|
1026
|
|
|
while ($textSpan !== '') { |
|
1027
|
|
|
if (substr($textSpan, -1) === ' ') { |
|
1028
|
|
|
$trailingString = ' ' . $trailingString; |
|
1029
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
1030
|
|
|
continue; |
|
1031
|
|
|
} |
|
1032
|
|
|
if (substr($textSpan, -6) === ' ') { |
|
1033
|
|
|
$trailingString = ' ' . $trailingString; |
|
1034
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
1035
|
|
|
continue; |
|
1036
|
|
|
} |
|
1037
|
|
|
break; |
|
1038
|
|
|
} |
|
1039
|
|
|
|
|
1040
|
|
|
self::$waitingText = $trailingString . self::$waitingText; |
|
1041
|
|
|
break; |
|
1042
|
|
|
} |
|
1043
|
|
|
} |
|
1044
|
|
|
|
|
1045
|
|
|
// Trailing " - " needs to be prefixed to the following span |
|
1046
|
|
|
if (!$theEnd && substr('...' . $textSpan, -3) === ' - ') { |
|
1047
|
|
|
$textSpan = substr($textSpan, 0, -3); |
|
1048
|
|
|
self::$waitingText = ' - ' . self::$waitingText; |
|
1049
|
|
|
} |
|
1050
|
|
|
|
|
1051
|
|
|
while (I18N::direction() === 'rtl') { |
|
1052
|
|
|
// Look for " - " preceding <RTLbr> and relocate it to the front of the string |
|
1053
|
|
|
$posDashString = strpos($textSpan, ' - <RTLbr>'); |
|
1054
|
|
|
if ($posDashString === false) { |
|
1055
|
|
|
break; |
|
1056
|
|
|
} |
|
1057
|
|
|
$posStringStart = strrpos(substr($textSpan, 0, $posDashString), '<RTLbr>'); |
|
1058
|
|
|
if ($posStringStart === false) { |
|
1059
|
|
|
$posStringStart = 0; |
|
1060
|
|
|
} else { |
|
1061
|
|
|
$posStringStart += 9; |
|
1062
|
|
|
} // Point to the first char following the last <RTLbr> |
|
1063
|
|
|
|
|
1064
|
|
|
$textSpan = substr($textSpan, 0, $posStringStart) . ' - ' . substr($textSpan, $posStringStart, $posDashString - $posStringStart) . substr($textSpan, $posDashString + 3); |
|
1065
|
|
|
} |
|
1066
|
|
|
|
|
1067
|
|
|
// Strip leading spaces from the RTL text |
|
1068
|
|
|
$countLeadingSpaces = 0; |
|
1069
|
|
|
while ($textSpan !== '') { |
|
1070
|
|
|
if (substr($textSpan, 0, 1) === ' ') { |
|
1071
|
|
|
$countLeadingSpaces++; |
|
1072
|
|
|
$textSpan = substr($textSpan, 1); |
|
1073
|
|
|
continue; |
|
1074
|
|
|
} |
|
1075
|
|
|
if (substr($textSpan, 0, 6) === ' ') { |
|
1076
|
|
|
$countLeadingSpaces++; |
|
1077
|
|
|
$textSpan = substr($textSpan, 6); |
|
1078
|
|
|
continue; |
|
1079
|
|
|
} |
|
1080
|
|
|
break; |
|
1081
|
|
|
} |
|
1082
|
|
|
|
|
1083
|
|
|
// Strip trailing spaces from the RTL text |
|
1084
|
|
|
$countTrailingSpaces = 0; |
|
1085
|
|
|
while ($textSpan !== '') { |
|
1086
|
|
|
if (substr($textSpan, -1) === ' ') { |
|
1087
|
|
|
$countTrailingSpaces++; |
|
1088
|
|
|
$textSpan = substr($textSpan, 0, -1); |
|
1089
|
|
|
continue; |
|
1090
|
|
|
} |
|
1091
|
|
|
if (substr($textSpan, -6) === ' ') { |
|
1092
|
|
|
$countTrailingSpaces++; |
|
1093
|
|
|
$textSpan = substr($textSpan, 0, -6); |
|
1094
|
|
|
continue; |
|
1095
|
|
|
} |
|
1096
|
|
|
break; |
|
1097
|
|
|
} |
|
1098
|
|
|
|
|
1099
|
|
|
// Look for trailing " -", reverse it, and relocate it to the front of the string |
|
1100
|
|
|
if (substr($textSpan, -2) === ' -') { |
|
1101
|
|
|
$posDashString = strlen($textSpan) - 2; |
|
1102
|
|
|
$posStringStart = strrpos(substr($textSpan, 0, $posDashString), '<RTLbr>'); |
|
1103
|
|
|
if ($posStringStart === false) { |
|
1104
|
|
|
$posStringStart = 0; |
|
1105
|
|
|
} else { |
|
1106
|
|
|
$posStringStart += 9; |
|
1107
|
|
|
} // Point to the first char following the last <RTLbr> |
|
1108
|
|
|
|
|
1109
|
|
|
$textSpan = substr($textSpan, 0, $posStringStart) . '- ' . substr($textSpan, $posStringStart, $posDashString - $posStringStart) . substr($textSpan, $posDashString + 2); |
|
1110
|
|
|
} |
|
1111
|
|
|
|
|
1112
|
|
|
if ($countLeadingSpaces !== 0) { |
|
1113
|
|
|
$newLength = strlen($textSpan) + $countLeadingSpaces; |
|
1114
|
|
|
$textSpan = str_pad($textSpan, $newLength, ' ', (I18N::direction() === 'rtl' ? STR_PAD_LEFT : STR_PAD_RIGHT)); |
|
1115
|
|
|
} |
|
1116
|
|
|
if ($countTrailingSpaces !== 0) { |
|
1117
|
|
|
if (I18N::direction() === 'ltr') { |
|
1118
|
|
|
if ($trailingBreaks === '') { |
|
1119
|
|
|
// Move trailing RTL spaces to front of following LTR span |
|
1120
|
|
|
$newLength = strlen(self::$waitingText) + $countTrailingSpaces; |
|
1121
|
|
|
self::$waitingText = str_pad(self::$waitingText, $newLength, ' ', STR_PAD_LEFT); |
|
1122
|
|
|
} |
|
1123
|
|
|
} else { |
|
1124
|
|
|
$newLength = strlen($textSpan) + $countTrailingSpaces; |
|
1125
|
|
|
$textSpan = str_pad($textSpan, $newLength, ' '); |
|
1126
|
|
|
} |
|
1127
|
|
|
} |
|
1128
|
|
|
|
|
1129
|
|
|
// We're done: finish the span |
|
1130
|
|
|
$textSpan = self::starredName($textSpan, 'RTL'); // Wrap starred name in <u> and </u> tags |
|
1131
|
|
|
$result .= $textSpan . self::END_RTL; |
|
1132
|
|
|
} |
|
1133
|
|
|
|
|
1134
|
|
|
if (self::$currentState !== 'LTR' && self::$currentState !== 'RTL') { |
|
1135
|
|
|
$result .= $textSpan; |
|
1136
|
|
|
} |
|
1137
|
|
|
|
|
1138
|
|
|
$result .= $trailingBreaks; // Get rid of any waiting <br> |
|
1139
|
|
|
} |
|
1140
|
|
|
|
|
1141
|
|
|
/** |
|
1142
|
|
|
* Wrap text, similar to the PHP wordwrap() function. |
|
1143
|
|
|
* |
|
1144
|
|
|
* @param string $string |
|
1145
|
|
|
* @param int $width |
|
1146
|
|
|
* @param string $sep |
|
1147
|
|
|
* @param bool $cut |
|
1148
|
|
|
* |
|
1149
|
|
|
* @return string |
|
1150
|
|
|
*/ |
|
1151
|
|
|
public static function utf8WordWrap(string $string, int $width = 75, string $sep = "\n", bool $cut = false): string |
|
1152
|
|
|
{ |
|
1153
|
|
|
$out = ''; |
|
1154
|
|
|
while ($string) { |
|
1155
|
|
|
if (mb_strlen($string) <= $width) { |
|
1156
|
|
|
// Do not wrap any text that is less than the output area. |
|
1157
|
|
|
$out .= $string; |
|
1158
|
|
|
$string = ''; |
|
1159
|
|
|
} else { |
|
1160
|
|
|
$sub1 = mb_substr($string, 0, $width + 1); |
|
1161
|
|
|
if (mb_substr($string, mb_strlen($sub1) - 1, 1) === ' ') { |
|
1162
|
|
|
// include words that end by a space immediately after the area. |
|
1163
|
|
|
$sub = $sub1; |
|
1164
|
|
|
} else { |
|
1165
|
|
|
$sub = mb_substr($string, 0, $width); |
|
1166
|
|
|
} |
|
1167
|
|
|
$spacepos = strrpos($sub, ' '); |
|
1168
|
|
|
if ($spacepos === false) { |
|
1169
|
|
|
// No space on line? |
|
1170
|
|
|
if ($cut) { |
|
1171
|
|
|
$out .= $sub . $sep; |
|
1172
|
|
|
$string = mb_substr($string, mb_strlen($sub)); |
|
1173
|
|
|
} else { |
|
1174
|
|
|
$spacepos = strpos($string, ' '); |
|
1175
|
|
|
if ($spacepos === false) { |
|
1176
|
|
|
$out .= $string; |
|
1177
|
|
|
$string = ''; |
|
1178
|
|
|
} else { |
|
1179
|
|
|
$out .= substr($string, 0, $spacepos) . $sep; |
|
1180
|
|
|
$string = substr($string, $spacepos + 1); |
|
1181
|
|
|
} |
|
1182
|
|
|
} |
|
1183
|
|
|
} else { |
|
1184
|
|
|
// Split at space; |
|
1185
|
|
|
$out .= substr($string, 0, $spacepos) . $sep; |
|
1186
|
|
|
$string = substr($string, $spacepos + 1); |
|
1187
|
|
|
} |
|
1188
|
|
|
} |
|
1189
|
|
|
} |
|
1190
|
|
|
|
|
1191
|
|
|
return $out; |
|
1192
|
|
|
} |
|
1193
|
|
|
} |
|
1194
|
|
|
|
fisharebest /
webtrees
