| 1 |
<?php |
| 2 |
|
| 3 |
/** |
| 4 |
* File: /helper/e2pdf-rtl.php |
| 5 |
* |
| 6 |
* @package E2Pdf |
| 7 |
* @license GPLv3 <https://www.gnu.org/licenses/gpl-3.0.html> |
| 8 |
* @link https://e2pdf.com |
| 9 |
* |
| 10 |
* Portions of this class (functions: arabic(), getGlyphs(), preConvert(), a4MaxChars(), |
| 11 |
* a4Lines(), utf8Glyphs(), decodeEntities(), decodeEntities2()) are derived from the |
| 12 |
* I18N/Arabic library, which is licensed under the GNU Lesser General Public License (LGPL). |
| 13 |
* |
| 14 |
* Original Author: Khaled Al-Sham'aa <khaled@ar-php.org> |
| 15 |
* Copyright: 2006-2016 Khaled Al-Sham'aa |
| 16 |
* License: LGPL v3 <https://www.gnu.org/licenses/lgpl-3.0.html> |
| 17 |
* Source: http://www.ar-php.org |
| 18 |
*/ |
| 19 |
if (!defined('ABSPATH')) { |
| 20 |
die('Access denied.'); |
| 21 |
} |
| 22 |
|
| 23 |
class Helper_E2pdf_Rtl { |
| 24 |
|
| 25 |
private $_glyphs = null; |
| 26 |
private $_hex = null; |
| 27 |
private $_prevLink = null; |
| 28 |
private $_nextLink = null; |
| 29 |
private $_vowel = null; |
| 30 |
|
| 31 |
public function rtl($value) { |
| 32 |
if (preg_match('/\p{Bidi_Class=R}|\p{Bidi_Class=AL}/u', $value)) { |
| 33 |
$reorderedText = $this->apply_bidi_algorithm($value); |
| 34 |
return $this->process_characters($reorderedText); |
| 35 |
} else { |
| 36 |
return $value; |
| 37 |
} |
| 38 |
} |
| 39 |
|
| 40 |
private function apply_bidi_algorithm($text) { |
| 41 |
$runs = $this->analyze_runs($text); |
| 42 |
return $this->reorder_runs($runs); |
| 43 |
} |
| 44 |
|
| 45 |
private function analyze_runs($text) { |
| 46 |
$chars = preg_split('//u', $text, -1, PREG_SPLIT_NO_EMPTY); |
| 47 |
$runs = []; |
| 48 |
$currentRun = ''; |
| 49 |
$currentDir = null; |
| 50 |
foreach ($chars as $char) { |
| 51 |
$charDir = $this->get_char_direction($char); |
| 52 |
if ($currentDir === null) { |
| 53 |
$currentDir = $charDir; |
| 54 |
$currentRun = $char; |
| 55 |
} elseif ($currentDir === $charDir) { |
| 56 |
$currentRun .= $char; |
| 57 |
} else { |
| 58 |
if ($currentRun !== '') { |
| 59 |
$runs[] = ['text' => $currentRun, 'dir' => $currentDir]; |
| 60 |
} |
| 61 |
$currentRun = $char; |
| 62 |
$currentDir = $charDir; |
| 63 |
} |
| 64 |
} |
| 65 |
if ($currentRun !== '') { |
| 66 |
$runs[] = ['text' => $currentRun, 'dir' => $currentDir]; |
| 67 |
} |
| 68 |
return $runs; |
| 69 |
} |
| 70 |
|
| 71 |
private function get_char_direction($char) { |
| 72 |
if (preg_match('/\p{Bidi_Class=R}|\p{Bidi_Class=AL}/u', $char)) { |
| 73 |
return 'RTL'; |
| 74 |
} |
| 75 |
if (preg_match('/\p{N}/u', $char)) { |
| 76 |
return 'NEUTRAL'; |
| 77 |
} |
| 78 |
|
| 79 |
// spaces and punctuation |
| 80 |
if (preg_match('/\s|[.!?،؟؛()[\]{}"\']/u', $char)) { |
| 81 |
return 'NEUTRAL'; |
| 82 |
} |
| 83 |
|
| 84 |
// everything else (Latin, etc.) is LTR |
| 85 |
return 'LTR'; |
| 86 |
} |
| 87 |
|
| 88 |
private function reorder_runs($runs) { |
| 89 |
if (empty($runs)) { |
| 90 |
return ''; |
| 91 |
} |
| 92 |
|
| 93 |
// find the main paragraph direction (first strong character) |
| 94 |
$mainDir = 'LTR'; // Default |
| 95 |
foreach ($runs as $run) { |
| 96 |
if ($run['dir'] === 'RTL') { |
| 97 |
$mainDir = 'RTL'; |
| 98 |
break; |
| 99 |
} elseif ($run['dir'] === 'LTR') { |
| 100 |
break; |
| 101 |
} |
| 102 |
} |
| 103 |
|
| 104 |
$result = ''; |
| 105 |
$rtlSequence = []; |
| 106 |
|
| 107 |
foreach ($runs as $run) { |
| 108 |
if ($run['dir'] === 'RTL') { |
| 109 |
// collect RTL runs |
| 110 |
$rtlSequence[] = $run; |
| 111 |
} else { |
| 112 |
// process any pending RTL sequence |
| 113 |
if (!empty($rtlSequence)) { |
| 114 |
// Reverse the RTL sequence and add to result |
| 115 |
$rtlSequence = array_reverse($rtlSequence); |
| 116 |
foreach ($rtlSequence as $rtlRun) { |
| 117 |
$result .= $rtlRun['text']; |
| 118 |
} |
| 119 |
$rtlSequence = []; |
| 120 |
} |
| 121 |
|
| 122 |
// add LTR/NEUTRAL run as-is |
| 123 |
$result .= $run['text']; |
| 124 |
} |
| 125 |
} |
| 126 |
|
| 127 |
// process any remaining RTL sequence |
| 128 |
if (!empty($rtlSequence)) { |
| 129 |
$rtlSequence = array_reverse($rtlSequence); |
| 130 |
foreach ($rtlSequence as $rtlRun) { |
| 131 |
$result .= $rtlRun['text']; |
| 132 |
} |
| 133 |
} |
| 134 |
|
| 135 |
// if main direction is RTL, reverse the entire result |
| 136 |
if ($mainDir === 'RTL') { |
| 137 |
$result = $this->reverse_by_words($result); |
| 138 |
} |
| 139 |
|
| 140 |
return $result; |
| 141 |
} |
| 142 |
|
| 143 |
private function reverse_by_words($text) { |
| 144 |
// split by spaces while preserving them |
| 145 |
$parts = preg_split('/(\s+)/u', $text, -1, PREG_SPLIT_DELIM_CAPTURE); |
| 146 |
return implode('', array_reverse($parts)); |
| 147 |
} |
| 148 |
|
| 149 |
private function process_characters($text) { |
| 150 |
// split text to handle different scripts separately |
| 151 |
$result = ''; |
| 152 |
$i = 0; |
| 153 |
$length = mb_strlen($text, 'UTF-8'); |
| 154 |
|
| 155 |
while ($i < $length) { |
| 156 |
$char = mb_substr($text, $i, 1, 'UTF-8'); |
| 157 |
|
| 158 |
if (preg_match('/\p{Arabic}/u', $char)) { |
| 159 |
// collect arabic sequence |
| 160 |
$arabicText = ''; |
| 161 |
while ($i < $length && preg_match('/\p{Arabic}/u', mb_substr($text, $i, 1, 'UTF-8'))) { |
| 162 |
$arabicText .= mb_substr($text, $i, 1, 'UTF-8'); |
| 163 |
$i++; |
| 164 |
} |
| 165 |
$result .= $this->arabic($arabicText); |
| 166 |
} elseif ($this->is_non_arabic_char(mb_substr($text, $i, 1, 'UTF-8'))) { |
| 167 |
// collect non-arabic RTL sequence |
| 168 |
$rtlText = ''; |
| 169 |
while ($i < $length && $this->is_non_arabic_char(mb_substr($text, $i, 1, 'UTF-8'))) { |
| 170 |
$rtlText .= mb_substr($text, $i, 1, 'UTF-8'); |
| 171 |
$i++; |
| 172 |
} |
| 173 |
$result .= $this->non_arabic($rtlText); |
| 174 |
} else { |
| 175 |
// Regular character |
| 176 |
$result .= $char; |
| 177 |
$i++; |
| 178 |
} |
| 179 |
} |
| 180 |
return $result; |
| 181 |
} |
| 182 |
|
| 183 |
function non_arabic($text) { |
| 184 |
$chars = preg_split('//u', $text, -1, PREG_SPLIT_NO_EMPTY); |
| 185 |
return implode('', array_reverse($chars)); |
| 186 |
} |
| 187 |
|
| 188 |
function is_non_arabic_char($char) { |
| 189 |
return preg_match('/\p{Bidi_Class=R}|\p{Bidi_Class=AL}/u', $char); |
| 190 |
} |
| 191 |
|
| 192 |
public function arabic($value, $max_chars = 99999, $hindo = true) { |
| 193 |
$this->_prevLink = '،؟؛ـئبتثجحخسشصضطظعغفقكل� |
| 194 |
نهي'; |
| 195 |
$this->_nextLink = 'ـآأؤإائبةتثجحخدذرز'; |
| 196 |
$this->_nextLink .= 'سشصضطظعغفقكل� |
| 197 |
نهوىي'; |
| 198 |
$this->_vowel = 'ًٌٍَُِّْ'; |
| 199 |
|
| 200 |
$this->_glyphs = 'ًٌٍَُِّْٰ'; |
| 201 |
$this->_hex = '064B064B064B064B064C064C064C064C064D064D064D064D064E064E'; |
| 202 |
$this->_hex .= '064E064E064F064F064F064F06500650065006500651065106510651'; |
| 203 |
$this->_hex .= '06520652065206520670067006700670'; |
| 204 |
|
| 205 |
$this->_glyphs .= 'ءآأؤإئاب'; |
| 206 |
$this->_hex .= 'FE80FE80FE80FE80FE81FE82FE81FE82FE83FE84FE83FE84FE85FE86'; |
| 207 |
$this->_hex .= 'FE85FE86FE87FE88FE87FE88FE89FE8AFE8BFE8CFE8DFE8EFE8DFE8E'; |
| 208 |
$this->_hex .= 'FE8FFE90FE91FE92'; |
| 209 |
|
| 210 |
$this->_glyphs .= 'ةتثجحخدذ'; |
| 211 |
$this->_hex .= 'FE93FE94FE93FE94FE95FE96FE97FE98FE99FE9AFE9BFE9CFE9DFE9E'; |
| 212 |
$this->_hex .= 'FE9FFEA0FEA1FEA2FEA3FEA4FEA5FEA6FEA7FEA8FEA9FEAAFEA9FEAA'; |
| 213 |
$this->_hex .= 'FEABFEACFEABFEAC'; |
| 214 |
|
| 215 |
$this->_glyphs .= 'رزسشصضطظ'; |
| 216 |
$this->_hex .= 'FEADFEAEFEADFEAEFEAFFEB0FEAFFEB0FEB1FEB2FEB3FEB4FEB5FEB6'; |
| 217 |
$this->_hex .= 'FEB7FEB8FEB9FEBAFEBBFEBCFEBDFEBEFEBFFEC0FEC1FEC2FEC3FEC4'; |
| 218 |
$this->_hex .= 'FEC5FEC6FEC7FEC8'; |
| 219 |
|
| 220 |
$this->_glyphs .= 'عغفقكل� |
| 221 |
ن'; |
| 222 |
$this->_hex .= 'FEC9FECAFECBFECCFECDFECEFECFFED0FED1FED2FED3FED4FED5FED6'; |
| 223 |
$this->_hex .= 'FED7FED8FED9FEDAFEDBFEDCFEDDFEDEFEDFFEE0FEE1FEE2FEE3FEE4'; |
| 224 |
$this->_hex .= 'FEE5FEE6FEE7FEE8'; |
| 225 |
|
| 226 |
$this->_glyphs .= 'هوىيـ،؟؛'; |
| 227 |
$this->_hex .= 'FEE9FEEAFEEBFEECFEEDFEEEFEEDFEEEFEEFFEF0FEEFFEF0FEF1FEF2'; |
| 228 |
$this->_hex .= 'FEF3FEF40640064006400640060C060C060C060C061F061F061F061F'; |
| 229 |
$this->_hex .= '061B061B061B061B'; |
| 230 |
|
| 231 |
// Support the extra 4 Persian letters (p), (ch), (zh) and (g) |
| 232 |
// This needs value in getGlyphs function to be 52 instead of 48 |
| 233 |
// $this->_glyphs .= chr(129).chr(141).chr(142).chr(144); |
| 234 |
// $this->_hex .= 'FB56FB57FB58FB59FB7AFB7BFB7CFB7DFB8AFB8BFB8AFB8BFB92'; |
| 235 |
// $this->_hex .= 'FB93FB94FB95'; |
| 236 |
// |
| 237 |
// $this->_prevLink .= chr(129).chr(141).chr(142).chr(144); |
| 238 |
// $this->_nextLink .= chr(129).chr(141).chr(142).chr(144); |
| 239 |
// |
| 240 |
// Example: $text = 'ن� |
| 241 |
ونة قل� |
| 242 |
: لاگچ ژافپ'; |
| 243 |
// Email Yossi Beck <yosbeck@gmail.com> ask him to save that example |
| 244 |
// string using ANSI encoding in Notepad |
| 245 |
$this->_glyphs .= ''; |
| 246 |
$this->_hex .= ''; |
| 247 |
|
| 248 |
$this->_glyphs .= 'لآلألإلا'; |
| 249 |
$this->_hex .= 'FEF5FEF6FEF5FEF6FEF7FEF8FEF7FEF8FEF9FEFAFEF9FEFAFEFBFEFC'; |
| 250 |
$this->_hex .= 'FEFBFEFC'; |
| 251 |
|
| 252 |
return $this->utf8Glyphs($value, $max_chars, $hindo); |
| 253 |
} |
| 254 |
|
| 255 |
/** |
| 256 |
* Get glyphs |
| 257 |
* |
| 258 |
* @param string $char Char |
| 259 |
* @param integer $type Type |
| 260 |
* |
| 261 |
* @return string |
| 262 |
*/ |
| 263 |
protected function getGlyphs($char, $type) { |
| 264 |
|
| 265 |
$pos = mb_strpos($this->_glyphs, $char); |
| 266 |
|
| 267 |
if ($pos > 49) { |
| 268 |
$pos = ($pos - 49) / 2 + 49; |
| 269 |
} |
| 270 |
|
| 271 |
$pos = $pos * 16 + $type * 4; |
| 272 |
|
| 273 |
return substr($this->_hex, $pos, 4); |
| 274 |
} |
| 275 |
|
| 276 |
/** |
| 277 |
* Convert Arabic Windows-1256 charset string into glyph joining in UTF-8 |
| 278 |
* hexadecimals stream |
| 279 |
* |
| 280 |
* @param string $str Arabic string in Windows-1256 charset |
| 281 |
* |
| 282 |
* @return string Arabic glyph joining in UTF-8 hexadecimals stream |
| 283 |
* @author Khaled Al-Sham'aa <khaled@ar-php.org> |
| 284 |
*/ |
| 285 |
protected function preConvert($str) { |
| 286 |
$crntChar = null; |
| 287 |
$prevChar = null; |
| 288 |
$nextChar = null; |
| 289 |
$output = ''; |
| 290 |
|
| 291 |
$_temp = mb_strlen($str); |
| 292 |
|
| 293 |
for ($i = 0; $i < $_temp; $i++) { |
| 294 |
$chars[] = mb_substr($str, $i, 1); |
| 295 |
} |
| 296 |
|
| 297 |
$max = count($chars); |
| 298 |
|
| 299 |
for ($i = $max - 1; $i >= 0; $i--) { |
| 300 |
$crntChar = $chars[$i]; |
| 301 |
$prevChar = ' '; |
| 302 |
|
| 303 |
if ($i > 0) { |
| 304 |
$prevChar = $chars[$i - 1]; |
| 305 |
} |
| 306 |
|
| 307 |
if ($prevChar && mb_strpos($this->_vowel, $prevChar) !== false) { |
| 308 |
$prevChar = $chars[$i - 2]; |
| 309 |
if ($prevChar && mb_strpos($this->_vowel, $prevChar) !== false) { |
| 310 |
$prevChar = $chars[$i - 3]; |
| 311 |
} |
| 312 |
} |
| 313 |
|
| 314 |
$Reversed = false; |
| 315 |
$flip_arr = ')]>}'; |
| 316 |
$ReversedChr = '([<{'; |
| 317 |
|
| 318 |
if ($crntChar && mb_strpos($flip_arr, $crntChar) !== false) { |
| 319 |
$crntChar = $ReversedChr[mb_strpos($flip_arr, $crntChar)]; |
| 320 |
$Reversed = true; |
| 321 |
} else { |
| 322 |
$Reversed = false; |
| 323 |
} |
| 324 |
|
| 325 |
if ($crntChar && !$Reversed && (mb_strpos($ReversedChr, $crntChar) !== false) |
| 326 |
) { |
| 327 |
$crntChar = $flip_arr[mb_strpos($ReversedChr, $crntChar)]; |
| 328 |
} |
| 329 |
|
| 330 |
if (ord($crntChar) < 128) { |
| 331 |
$output .= $crntChar; |
| 332 |
$nextChar = $crntChar; |
| 333 |
continue; |
| 334 |
} |
| 335 |
|
| 336 |
if ($crntChar == 'ل' && isset($chars[$i + 1]) && (mb_strpos('آأإا', $chars[$i + 1]) !== false) |
| 337 |
) { |
| 338 |
continue; |
| 339 |
} |
| 340 |
|
| 341 |
if ($crntChar && mb_strpos($this->_vowel, $crntChar) !== false) { |
| 342 |
if (isset($chars[$i + 1]) && (mb_strpos($this->_nextLink, $chars[$i + 1]) !== false) && (mb_strpos($this->_prevLink, $prevChar) !== false) |
| 343 |
) { |
| 344 |
$output .= '&#x' . $this->getGlyphs($crntChar, 1) . ';'; |
| 345 |
} else { |
| 346 |
$output .= '&#x' . $this->getGlyphs($crntChar, 0) . ';'; |
| 347 |
} |
| 348 |
continue; |
| 349 |
} |
| 350 |
|
| 351 |
$form = 0; |
| 352 |
|
| 353 |
if (($prevChar == 'لا' || $prevChar == 'لآ' || $prevChar == 'لأ' || $prevChar == 'لإ' || $prevChar == 'ل') && (mb_strpos('آأإا', $crntChar) !== false) |
| 354 |
) { |
| 355 |
if (mb_strpos($this->_prevLink, $chars[$i - 2]) !== false) { |
| 356 |
$form++; |
| 357 |
} |
| 358 |
|
| 359 |
if (mb_strpos($this->_vowel, $chars[$i - 1])) { |
| 360 |
$output .= '&#x'; |
| 361 |
$output .= $this->getGlyphs($crntChar, $form) . ';'; |
| 362 |
} else { |
| 363 |
$output .= '&#x'; |
| 364 |
$output .= $this->getGlyphs($prevChar . $crntChar, $form) . ';'; |
| 365 |
} |
| 366 |
$nextChar = $prevChar; |
| 367 |
continue; |
| 368 |
} |
| 369 |
|
| 370 |
if ($prevChar && mb_strpos($this->_prevLink, $prevChar) !== false) { |
| 371 |
$form++; |
| 372 |
} |
| 373 |
|
| 374 |
if ($nextChar && mb_strpos($this->_nextLink, $nextChar) !== false) { |
| 375 |
$form += 2; |
| 376 |
} |
| 377 |
|
| 378 |
$output .= '&#x' . $this->getGlyphs($crntChar, $form) . ';'; |
| 379 |
$nextChar = $crntChar; |
| 380 |
} |
| 381 |
|
| 382 |
// from Arabic Presentation Forms-B, Range: FE70-FEFF, |
| 383 |
// file "UFE70.pdf" (in reversed order) |
| 384 |
// into Arabic Presentation Forms-A, Range: FB50-FDFF, file "UFB50.pdf" |
| 385 |
// Example: $output = str_replace('ﺠﻟ', 'ﳉ', $output); |
| 386 |
// Lam Jeem |
| 387 |
|
| 388 |
$output = $this->decodeEntities($output, $exclude = array('&')); |
| 389 |
return $output; |
| 390 |
} |
| 391 |
|
| 392 |
/** |
| 393 |
* Regression analysis calculate roughly the max number of character fit in |
| 394 |
* one A4 page line for a given font size. |
| 395 |
* |
| 396 |
* @param integer $font Font size |
| 397 |
* |
| 398 |
* @return integer Maximum number of characters per line |
| 399 |
* @author Khaled Al-Sham'aa <khaled@ar-php.org> |
| 400 |
*/ |
| 401 |
public function a4MaxChars($font) { |
| 402 |
$x = 381.6 - 31.57 * $font + 1.182 * pow($font, 2) - 0.02052 * |
| 403 |
pow($font, 3) + 0.0001342 * pow($font, 4); |
| 404 |
return floor($x - 2); |
| 405 |
} |
| 406 |
|
| 407 |
/** |
| 408 |
* Calculate the lines number of given Arabic text and font size that will |
| 409 |
* fit in A4 page size |
| 410 |
* |
| 411 |
* @param string $str Arabic string you would like to split it into lines |
| 412 |
* @param integer $font Font size |
| 413 |
* |
| 414 |
* @return integer Number of lines for a given Arabic string in A4 page size |
| 415 |
* @author Khaled Al-Sham'aa <khaled@ar-php.org> |
| 416 |
*/ |
| 417 |
public function a4Lines($str, $font) { |
| 418 |
$str = str_replace(array("\r\n", "\n", "\r"), "\n", $str); |
| 419 |
|
| 420 |
$lines = 0; |
| 421 |
$chars = 0; |
| 422 |
$words = explode(' ', $str); |
| 423 |
$w_count = count($words); |
| 424 |
$max_chars = $this->a4MaxChars($font); |
| 425 |
|
| 426 |
for ($i = 0; $i < $w_count; $i++) { |
| 427 |
$w_len = mb_strlen($words[$i]) + 1; |
| 428 |
|
| 429 |
if ($chars + $w_len < $max_chars) { |
| 430 |
if (mb_strpos($words[$i], "\n") !== false) { |
| 431 |
$words_nl = explode("\n", $words[$i]); |
| 432 |
|
| 433 |
$nl_num = count($words_nl) - 1; |
| 434 |
for ($j = 1; $j < $nl_num; $j++) { |
| 435 |
$lines++; |
| 436 |
} |
| 437 |
|
| 438 |
$chars = mb_strlen($words_nl[$nl_num]) + 1; |
| 439 |
} else { |
| 440 |
$chars += $w_len; |
| 441 |
} |
| 442 |
} else { |
| 443 |
$lines++; |
| 444 |
$chars = $w_len; |
| 445 |
} |
| 446 |
} |
| 447 |
$lines++; |
| 448 |
|
| 449 |
return $lines; |
| 450 |
} |
| 451 |
|
| 452 |
/** |
| 453 |
* Convert Arabic Windows-1256 charset string into glyph joining in UTF-8 |
| 454 |
* hexadecimals stream (take care of whole the document including English |
| 455 |
* sections as well as numbers and arcs etc...) |
| 456 |
* |
| 457 |
* @param string $str Arabic string in Windows-1256 charset |
| 458 |
* @param integer $max_chars Max number of chars you can fit in one line |
| 459 |
* @param boolean $hindo If true use Hindo digits else use Arabic digits |
| 460 |
* |
| 461 |
* @return string Arabic glyph joining in UTF-8 hexadecimals stream (take |
| 462 |
* care of whole document including English sections as well |
| 463 |
* as numbers and arcs etc...) |
| 464 |
* @author Khaled Al-Sham'aa <khaled@ar-php.org> |
| 465 |
*/ |
| 466 |
public function utf8Glyphs($str, $max_chars = 50, $hindo = true) { |
| 467 |
$str = str_replace(array("\r\n", "\n", "\r"), " \n ", $str); |
| 468 |
$str = str_replace("\t", ' ', $str); |
| 469 |
|
| 470 |
$lines = array(); |
| 471 |
$words = explode(' ', $str); |
| 472 |
$w_count = count($words); |
| 473 |
$c_chars = 0; |
| 474 |
$c_words = array(); |
| 475 |
|
| 476 |
$english = array(); |
| 477 |
$en_index = -1; |
| 478 |
|
| 479 |
$en_words = array(); |
| 480 |
$en_stack = array(); |
| 481 |
|
| 482 |
for ($i = 0; $i < $w_count; $i++) { |
| 483 |
$pattern = '/^(\n?)'; |
| 484 |
$pattern .= '[a-z\d\\/\@\#\$\%\^\&\*\(\)\_\~\"\'\[\]\{\}\;\,\|\-\.\:!]*'; |
| 485 |
$pattern .= '([\.\:\+\=\-\!،؟]?)$/i'; |
| 486 |
|
| 487 |
if (preg_match($pattern, $words[$i], $matches)) { |
| 488 |
if ($matches[1]) { |
| 489 |
$words[$i] = mb_substr($words[$i], 1) . $matches[1]; |
| 490 |
} |
| 491 |
if ($matches[2]) { |
| 492 |
$words[$i] = $matches[2] . mb_substr($words[$i], 0, -1); |
| 493 |
} |
| 494 |
$words[$i] = strrev($words[$i]); |
| 495 |
array_push($english, $words[$i]); |
| 496 |
if ($en_index == -1) { |
| 497 |
$en_index = $i; |
| 498 |
} |
| 499 |
$en_words[] = true; |
| 500 |
} elseif ($en_index != -1) { |
| 501 |
$en_count = count($english); |
| 502 |
|
| 503 |
for ($j = 0; $j < $en_count; $j++) { |
| 504 |
$words[$en_index + $j] = $english[$en_count - 1 - $j]; |
| 505 |
} |
| 506 |
|
| 507 |
$en_index = -1; |
| 508 |
$english = array(); |
| 509 |
|
| 510 |
$en_words[] = false; |
| 511 |
} else { |
| 512 |
$en_words[] = false; |
| 513 |
} |
| 514 |
} |
| 515 |
|
| 516 |
if ($en_index != -1) { |
| 517 |
$en_count = count($english); |
| 518 |
|
| 519 |
for ($j = 0; $j < $en_count; $j++) { |
| 520 |
$words[$en_index + $j] = $english[$en_count - 1 - $j]; |
| 521 |
} |
| 522 |
} |
| 523 |
|
| 524 |
for ($i = 0; $i < $w_count; $i++) { |
| 525 |
$w_len = mb_strlen($words[$i]) + 1; |
| 526 |
|
| 527 |
if ($c_chars + $w_len < $max_chars) { |
| 528 |
if (mb_strpos($words[$i], "\n") !== false) { |
| 529 |
$words_nl = explode("\n", $words[$i]); |
| 530 |
|
| 531 |
array_push($c_words, $words_nl[0]); |
| 532 |
array_push($lines, implode(' ', $c_words)); |
| 533 |
|
| 534 |
$nl_num = count($words_nl) - 1; |
| 535 |
for ($j = 1; $j < $nl_num; $j++) { |
| 536 |
array_push($lines, $words_nl[$j]); |
| 537 |
} |
| 538 |
|
| 539 |
$c_words = array($words_nl[$nl_num]); |
| 540 |
$c_chars = mb_strlen($words_nl[$nl_num]) + 1; |
| 541 |
} else { |
| 542 |
array_push($c_words, $words[$i]); |
| 543 |
$c_chars += $w_len; |
| 544 |
} |
| 545 |
} else { |
| 546 |
array_push($lines, implode(' ', $c_words)); |
| 547 |
$c_words = array($words[$i]); |
| 548 |
$c_chars = $w_len; |
| 549 |
} |
| 550 |
} |
| 551 |
array_push($lines, implode(' ', $c_words)); |
| 552 |
|
| 553 |
$maxLine = count($lines); |
| 554 |
$output = ''; |
| 555 |
|
| 556 |
for ($j = $maxLine - 1; $j >= 0; $j--) { |
| 557 |
$output .= $lines[$j] . "\n"; |
| 558 |
} |
| 559 |
|
| 560 |
$output = rtrim($output); |
| 561 |
|
| 562 |
$output = $this->preConvert($output); |
| 563 |
if ($hindo) { |
| 564 |
$nums = array( |
| 565 |
'0', '1', '2', '3', '4', |
| 566 |
'5', '6', '7', '8', '9', |
| 567 |
); |
| 568 |
$arNums = array( |
| 569 |
'٠', '١', '٢', '٣', '٤', |
| 570 |
'٥', '٦', '٧', '٨', '٩', |
| 571 |
); |
| 572 |
|
| 573 |
foreach ($nums as $k => $v) { |
| 574 |
$p_nums[$k] = '/' . $v . '/ui'; |
| 575 |
} |
| 576 |
$output = preg_replace($p_nums, $arNums, $output); |
| 577 |
|
| 578 |
foreach ($arNums as $k => $v) { |
| 579 |
$p_arNums[$k] = '/([a-z-\d]+)' . $v . '/ui'; |
| 580 |
} |
| 581 |
foreach ($nums as $k => $v) { |
| 582 |
$r_nums[$k] = '${1}' . $v; |
| 583 |
} |
| 584 |
$output = preg_replace($p_arNums, $r_nums, $output); |
| 585 |
|
| 586 |
foreach ($arNums as $k => $v) { |
| 587 |
$p_arNums[$k] = '/' . $v . '([a-z-\d]+)/ui'; |
| 588 |
} |
| 589 |
foreach ($nums as $k => $v) { |
| 590 |
$r_nums[$k] = $v . '${1}'; |
| 591 |
} |
| 592 |
$output = preg_replace($p_arNums, $r_nums, $output); |
| 593 |
} |
| 594 |
|
| 595 |
return $output; |
| 596 |
} |
| 597 |
|
| 598 |
/** |
| 599 |
* Decode all HTML entities (including numerical ones) to regular UTF-8 bytes. |
| 600 |
* Double-escaped entities will only be decoded once |
| 601 |
* ("&lt;" becomes "<", not "<"). |
| 602 |
* |
| 603 |
* @param string $text The text to decode entities in. |
| 604 |
* @param array $exclude An array of characters which should not be decoded. |
| 605 |
* For example, array('<', '&', '"'). This affects |
| 606 |
* both named and numerical entities. |
| 607 |
* |
| 608 |
* @return string |
| 609 |
*/ |
| 610 |
protected function decodeEntities($text, $exclude = array()) { |
| 611 |
static $table; |
| 612 |
|
| 613 |
// We store named entities in a table for quick processing. |
| 614 |
if (!isset($table)) { |
| 615 |
// Get all named HTML entities. |
| 616 |
$table = array_flip(get_html_translation_table(HTML_ENTITIES)); |
| 617 |
|
| 618 |
// PHP gives us ISO-8859-1 data, we need UTF-8. |
| 619 |
$table = array_map('utf8_encode', $table); |
| 620 |
|
| 621 |
// Add apostrophe (XML) |
| 622 |
$table['''] = "'"; |
| 623 |
} |
| 624 |
$newtable = array_diff($table, $exclude); |
| 625 |
|
| 626 |
// Use a regexp to select all entities in one pass, to avoid decoding |
| 627 |
// double-escaped entities twice. |
| 628 |
//return preg_replace('/&(#x?)?([A-Za-z0-9]+);/e', |
| 629 |
// '$this->decodeEntities2("$1", "$2", "$0", $newtable, |
| 630 |
// $exclude)', $text); |
| 631 |
|
| 632 |
$pieces = explode('&', $text); |
| 633 |
$text = array_shift($pieces); |
| 634 |
foreach ($pieces as $piece) { |
| 635 |
if ($piece[0] == '#') { |
| 636 |
if ($piece[1] == 'x') { |
| 637 |
$one = '#x'; |
| 638 |
} else { |
| 639 |
$one = '#'; |
| 640 |
} |
| 641 |
} else { |
| 642 |
$one = ''; |
| 643 |
} |
| 644 |
$end = mb_strpos($piece, ';'); |
| 645 |
$start = mb_strlen($one); |
| 646 |
|
| 647 |
$two = mb_substr($piece, $start, $end - $start); |
| 648 |
$zero = '&' . $one . $two . ';'; |
| 649 |
$text .= $this->decodeEntities2($one, $two, $zero, $newtable, $exclude) . |
| 650 |
mb_substr($piece, $end + 1); |
| 651 |
} |
| 652 |
return $text; |
| 653 |
} |
| 654 |
|
| 655 |
/** |
| 656 |
* Helper function for decodeEntities |
| 657 |
* |
| 658 |
* @param string $prefix Prefix |
| 659 |
* @param string $codepoint Codepoint |
| 660 |
* @param string $original Original |
| 661 |
* @param array &$table Store named entities in a table |
| 662 |
* @param array &$exclude An array of characters which should not be decoded |
| 663 |
* |
| 664 |
* @return string |
| 665 |
*/ |
| 666 |
protected function decodeEntities2( |
| 667 |
$prefix, $codepoint, $original, &$table, &$exclude |
| 668 |
) { |
| 669 |
// Named entity |
| 670 |
if (!$prefix) { |
| 671 |
if (isset($table[$original])) { |
| 672 |
return $table[$original]; |
| 673 |
} else { |
| 674 |
return $original; |
| 675 |
} |
| 676 |
} |
| 677 |
|
| 678 |
// Hexadecimal numerical entity |
| 679 |
if ($prefix == '#x') { |
| 680 |
$codepoint = base_convert($codepoint, 16, 10); |
| 681 |
} |
| 682 |
|
| 683 |
// Encode codepoint as UTF-8 bytes |
| 684 |
if ($codepoint < 0x80) { |
| 685 |
$str = chr($codepoint); |
| 686 |
} elseif ($codepoint < 0x800) { |
| 687 |
$str = chr(0xC0 | ($codepoint >> 6)) . |
| 688 |
chr(0x80 | ($codepoint & 0x3F)); |
| 689 |
} elseif ($codepoint < 0x10000) { |
| 690 |
$str = chr(0xE0 | ($codepoint >> 12)) . |
| 691 |
chr(0x80 | (($codepoint >> 6) & 0x3F)) . |
| 692 |
chr(0x80 | ($codepoint & 0x3F)); |
| 693 |
} elseif ($codepoint < 0x200000) { |
| 694 |
$str = chr(0xF0 | ($codepoint >> 18)) . |
| 695 |
chr(0x80 | (($codepoint >> 12) & 0x3F)) . |
| 696 |
chr(0x80 | (($codepoint >> 6) & 0x3F)) . |
| 697 |
chr(0x80 | ($codepoint & 0x3F)); |
| 698 |
} |
| 699 |
|
| 700 |
// Check for excluded characters |
| 701 |
if (in_array($str, $exclude, false)) { |
| 702 |
return $original; |
| 703 |
} else { |
| 704 |
return $str; |
| 705 |
} |
| 706 |
} |
| 707 |
} |
| 708 |
|
