Changeset 2123 for trunk/include/functions.inc.php
- Timestamp:
- Oct 6, 2007, 12:56:22 AM (17 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/include/functions.inc.php
r2122 r2123 300 300 } 301 301 302 /* Returns true if the string appears to be encoded in UTF-8. (from wordpress) 303 * @param string Str 304 */ 305 function seems_utf8($Str) { # by bmorel at ssi dot fr 306 for ($i=0; $i<strlen($Str); $i++) { 307 if (ord($Str[$i]) < 0x80) continue; # 0bbbbbbb 308 elseif ((ord($Str[$i]) & 0xE0) == 0xC0) $n=1; # 110bbbbb 309 elseif ((ord($Str[$i]) & 0xF0) == 0xE0) $n=2; # 1110bbbb 310 elseif ((ord($Str[$i]) & 0xF8) == 0xF0) $n=3; # 11110bbb 311 elseif ((ord($Str[$i]) & 0xFC) == 0xF8) $n=4; # 111110bb 312 elseif ((ord($Str[$i]) & 0xFE) == 0xFC) $n=5; # 1111110b 313 else return false; # Does not match any model 314 for ($j=0; $j<$n; $j++) { # n bytes matching 10bbbbbb follow ? 315 if ((++$i == strlen($Str)) || ((ord($Str[$i]) & 0xC0) != 0x80)) 316 return false; 317 } 318 } 319 return true; 320 } 321 322 /* Remove accents from a UTF-8 or ISO-859-1 string (from wordpress) 323 * @param string sstring - an UTF-8 or ISO-8859-1 string 324 */ 325 function remove_accents($string) 326 { 327 if ( !preg_match('/[\x80-\xff]/', $string) ) 328 return $string; 329 330 if (seems_utf8($string)) { 331 $chars = array( 332 // Decompositions for Latin-1 Supplement 333 chr(195).chr(128) => 'A', chr(195).chr(129) => 'A', 334 chr(195).chr(130) => 'A', chr(195).chr(131) => 'A', 335 chr(195).chr(132) => 'A', chr(195).chr(133) => 'A', 336 chr(195).chr(135) => 'C', chr(195).chr(136) => 'E', 337 chr(195).chr(137) => 'E', chr(195).chr(138) => 'E', 338 chr(195).chr(139) => 'E', chr(195).chr(140) => 'I', 339 chr(195).chr(141) => 'I', chr(195).chr(142) => 'I', 340 chr(195).chr(143) => 'I', chr(195).chr(145) => 'N', 341 chr(195).chr(146) => 'O', chr(195).chr(147) => 'O', 342 chr(195).chr(148) => 'O', chr(195).chr(149) => 'O', 343 chr(195).chr(150) => 'O', chr(195).chr(153) => 'U', 344 chr(195).chr(154) => 'U', chr(195).chr(155) => 'U', 345 chr(195).chr(156) => 'U', chr(195).chr(157) => 'Y', 346 chr(195).chr(159) => 's', chr(195).chr(160) => 'a', 347 chr(195).chr(161) => 'a', chr(195).chr(162) => 'a', 348 chr(195).chr(163) => 'a', chr(195).chr(164) => 'a', 349 chr(195).chr(165) => 'a', chr(195).chr(167) => 'c', 350 chr(195).chr(168) => 'e', chr(195).chr(169) => 'e', 351 chr(195).chr(170) => 'e', chr(195).chr(171) => 'e', 352 chr(195).chr(172) => 'i', chr(195).chr(173) => 'i', 353 chr(195).chr(174) => 'i', chr(195).chr(175) => 'i', 354 chr(195).chr(177) => 'n', chr(195).chr(178) => 'o', 355 chr(195).chr(179) => 'o', chr(195).chr(180) => 'o', 356 chr(195).chr(181) => 'o', chr(195).chr(182) => 'o', 357 chr(195).chr(182) => 'o', chr(195).chr(185) => 'u', 358 chr(195).chr(186) => 'u', chr(195).chr(187) => 'u', 359 chr(195).chr(188) => 'u', chr(195).chr(189) => 'y', 360 chr(195).chr(191) => 'y', 361 // Decompositions for Latin Extended-A 362 chr(196).chr(128) => 'A', chr(196).chr(129) => 'a', 363 chr(196).chr(130) => 'A', chr(196).chr(131) => 'a', 364 chr(196).chr(132) => 'A', chr(196).chr(133) => 'a', 365 chr(196).chr(134) => 'C', chr(196).chr(135) => 'c', 366 chr(196).chr(136) => 'C', chr(196).chr(137) => 'c', 367 chr(196).chr(138) => 'C', chr(196).chr(139) => 'c', 368 chr(196).chr(140) => 'C', chr(196).chr(141) => 'c', 369 chr(196).chr(142) => 'D', chr(196).chr(143) => 'd', 370 chr(196).chr(144) => 'D', chr(196).chr(145) => 'd', 371 chr(196).chr(146) => 'E', chr(196).chr(147) => 'e', 372 chr(196).chr(148) => 'E', chr(196).chr(149) => 'e', 373 chr(196).chr(150) => 'E', chr(196).chr(151) => 'e', 374 chr(196).chr(152) => 'E', chr(196).chr(153) => 'e', 375 chr(196).chr(154) => 'E', chr(196).chr(155) => 'e', 376 chr(196).chr(156) => 'G', chr(196).chr(157) => 'g', 377 chr(196).chr(158) => 'G', chr(196).chr(159) => 'g', 378 chr(196).chr(160) => 'G', chr(196).chr(161) => 'g', 379 chr(196).chr(162) => 'G', chr(196).chr(163) => 'g', 380 chr(196).chr(164) => 'H', chr(196).chr(165) => 'h', 381 chr(196).chr(166) => 'H', chr(196).chr(167) => 'h', 382 chr(196).chr(168) => 'I', chr(196).chr(169) => 'i', 383 chr(196).chr(170) => 'I', chr(196).chr(171) => 'i', 384 chr(196).chr(172) => 'I', chr(196).chr(173) => 'i', 385 chr(196).chr(174) => 'I', chr(196).chr(175) => 'i', 386 chr(196).chr(176) => 'I', chr(196).chr(177) => 'i', 387 chr(196).chr(178) => 'IJ',chr(196).chr(179) => 'ij', 388 chr(196).chr(180) => 'J', chr(196).chr(181) => 'j', 389 chr(196).chr(182) => 'K', chr(196).chr(183) => 'k', 390 chr(196).chr(184) => 'k', chr(196).chr(185) => 'L', 391 chr(196).chr(186) => 'l', chr(196).chr(187) => 'L', 392 chr(196).chr(188) => 'l', chr(196).chr(189) => 'L', 393 chr(196).chr(190) => 'l', chr(196).chr(191) => 'L', 394 chr(197).chr(128) => 'l', chr(197).chr(129) => 'L', 395 chr(197).chr(130) => 'l', chr(197).chr(131) => 'N', 396 chr(197).chr(132) => 'n', chr(197).chr(133) => 'N', 397 chr(197).chr(134) => 'n', chr(197).chr(135) => 'N', 398 chr(197).chr(136) => 'n', chr(197).chr(137) => 'N', 399 chr(197).chr(138) => 'n', chr(197).chr(139) => 'N', 400 chr(197).chr(140) => 'O', chr(197).chr(141) => 'o', 401 chr(197).chr(142) => 'O', chr(197).chr(143) => 'o', 402 chr(197).chr(144) => 'O', chr(197).chr(145) => 'o', 403 chr(197).chr(146) => 'OE',chr(197).chr(147) => 'oe', 404 chr(197).chr(148) => 'R',chr(197).chr(149) => 'r', 405 chr(197).chr(150) => 'R',chr(197).chr(151) => 'r', 406 chr(197).chr(152) => 'R',chr(197).chr(153) => 'r', 407 chr(197).chr(154) => 'S',chr(197).chr(155) => 's', 408 chr(197).chr(156) => 'S',chr(197).chr(157) => 's', 409 chr(197).chr(158) => 'S',chr(197).chr(159) => 's', 410 chr(197).chr(160) => 'S', chr(197).chr(161) => 's', 411 chr(197).chr(162) => 'T', chr(197).chr(163) => 't', 412 chr(197).chr(164) => 'T', chr(197).chr(165) => 't', 413 chr(197).chr(166) => 'T', chr(197).chr(167) => 't', 414 chr(197).chr(168) => 'U', chr(197).chr(169) => 'u', 415 chr(197).chr(170) => 'U', chr(197).chr(171) => 'u', 416 chr(197).chr(172) => 'U', chr(197).chr(173) => 'u', 417 chr(197).chr(174) => 'U', chr(197).chr(175) => 'u', 418 chr(197).chr(176) => 'U', chr(197).chr(177) => 'u', 419 chr(197).chr(178) => 'U', chr(197).chr(179) => 'u', 420 chr(197).chr(180) => 'W', chr(197).chr(181) => 'w', 421 chr(197).chr(182) => 'Y', chr(197).chr(183) => 'y', 422 chr(197).chr(184) => 'Y', chr(197).chr(185) => 'Z', 423 chr(197).chr(186) => 'z', chr(197).chr(187) => 'Z', 424 chr(197).chr(188) => 'z', chr(197).chr(189) => 'Z', 425 chr(197).chr(190) => 'z', chr(197).chr(191) => 's', 426 // Euro Sign 427 chr(226).chr(130).chr(172) => 'E', 428 // GBP (Pound) Sign 429 chr(194).chr(163) => ''); 430 431 $string = strtr($string, $chars); 432 } else { 433 // Assume ISO-8859-1 if not UTF-8 434 $chars['in'] = chr(128).chr(131).chr(138).chr(142).chr(154).chr(158) 435 .chr(159).chr(162).chr(165).chr(181).chr(192).chr(193).chr(194) 436 .chr(195).chr(196).chr(197).chr(199).chr(200).chr(201).chr(202) 437 .chr(203).chr(204).chr(205).chr(206).chr(207).chr(209).chr(210) 438 .chr(211).chr(212).chr(213).chr(214).chr(216).chr(217).chr(218) 439 .chr(219).chr(220).chr(221).chr(224).chr(225).chr(226).chr(227) 440 .chr(228).chr(229).chr(231).chr(232).chr(233).chr(234).chr(235) 441 .chr(236).chr(237).chr(238).chr(239).chr(241).chr(242).chr(243) 442 .chr(244).chr(245).chr(246).chr(248).chr(249).chr(250).chr(251) 443 .chr(252).chr(253).chr(255); 444 445 $chars['out'] = "EfSZszYcYuAAAAAACEEEEIIIINOOOOOOUUUUYaaaaaaceeeeiiiinoooooouuuuyy"; 446 447 $string = strtr($string, $chars['in'], $chars['out']); 448 $double_chars['in'] = array(chr(140), chr(156), chr(198), chr(208), chr(222), chr(223), chr(230), chr(240), chr(254)); 449 $double_chars['out'] = array('OE', 'oe', 'AE', 'DH', 'TH', 'ss', 'ae', 'dh', 'th'); 450 $string = str_replace($double_chars['in'], $double_chars['out'], $string); 451 } 452 453 return $string; 454 } 455 302 456 /** 303 457 * simplify a string to insert it into an URL 304 *305 * based on str2url function from Dotclear306 458 * 307 459 * @param string … … 310 462 function str2url($str) 311 463 { 312 $str = strtr( 313 $str, 314 'ÀÁÂÃÄÅàáâãäåÇçÒÓÔÕÖØòóôõöøÈÉÊËèéêëÌÍÎÏìíîïÙÚÛÜùúûü¾ÝÿýÑñ', 315 'AAAAAAaaaaaaCcOOOOOOooooooEEEEeeeeIIIIiiiiUUUUuuuuYYyyNn' 316 ); 317 318 $str = str_replace('Æ', 'AE', $str); 319 $str = str_replace('æ', 'ae', $str); 320 $str = str_replace('¼', 'OE', $str); 321 $str = str_replace('½', 'oe', $str); 322 464 $str = remove_accents($str); 323 465 $str = preg_replace('/[^a-z0-9_\s\'\:\/\[\],-]/','',strtolower($str)); 324 466 $str = preg_replace('/[\s\'\:\/\[\],-]+/',' ',trim($str)); … … 354 496 355 497 return $languages; 356 }357 358 /**359 * replaces the $search into <span style="$style">$search</span> in the360 * given $string.361 *362 * case insensitive replacements, does not replace characters in HTML tags363 *364 * @param string $string365 * @param string $search366 * @param string $style367 * @return string368 */369 function add_style( $string, $search, $style )370 {371 //return $string;372 $return_string = '';373 $remaining = $string;374 375 $start = 0;376 $end = 0;377 $start = strpos ( $remaining, '<' );378 $end = strpos ( $remaining, '>' );379 while ( is_numeric( $start ) and is_numeric( $end ) )380 {381 $treatment = substr ( $remaining, 0, $start );382 $treatment = preg_replace( '/('.$search.')/i',383 '<span style="'.$style.'">\\0</span>',384 $treatment );385 $return_string.= $treatment.substr( $remaining, $start, $end-$start+1 );386 $remaining = substr ( $remaining, $end + 1, strlen( $remaining ) );387 $start = strpos ( $remaining, '<' );388 $end = strpos ( $remaining, '>' );389 }390 $treatment = preg_replace( '/('.$search.')/i',391 '<span style="'.$style.'">\\0</span>',392 $remaining );393 $return_string.= $treatment;394 395 return $return_string;396 }397 398 // replace_search replaces a searched words array string by the search in399 // another style for the given $string.400 function replace_search( $string, $search )401 {402 // FIXME : with new advanced search, this function needs a rewrite403 return $string;404 405 $words = explode( ',', $search );406 $style = 'background-color:white;color:red;';407 foreach ( $words as $word ) {408 $string = add_style( $string, $word, $style );409 }410 return $string;411 498 } 412 499
Note: See TracChangeset
for help on using the changeset viewer.