vendor/symfony/polyfill-mbstring/Mbstring.php

   1 <?php
   2
   3 /*
   4  * This file is part of the Symfony package.
   5  *
   6  * (c) Fabien Potencier <fabien@symfony.com>
   7  *
   8  * For the full copyright and license information, please view the LICENSE
   9  * file that was distributed with this source code.
  10  */
  11
  12 namespace Symfony\Polyfill\Mbstring;
  13
  14 /**
  15  * Partial mbstring implementation in PHP, iconv based, UTF-8 centric.
  16  *
  17  * Implemented:
  18  * - mb_chr                  - Returns a specific character from its Unicode code point
  19  * - mb_convert_encoding     - Convert character encoding
  20  * - mb_convert_variables    - Convert character code in variable(s)
  21  * - mb_decode_mimeheader    - Decode string in MIME header field
  22  * - mb_encode_mimeheader    - Encode string for MIME header XXX NATIVE IMPLEMENTATION IS REALLY BUGGED
  23  * - mb_decode_numericentity - Decode HTML numeric string reference to character
  24  * - mb_encode_numericentity - Encode character to HTML numeric string reference
  25  * - mb_convert_case         - Perform case folding on a string
  26  * - mb_detect_encoding      - Detect character encoding
  27  * - mb_get_info             - Get internal settings of mbstring
  28  * - mb_http_input           - Detect HTTP input character encoding
  29  * - mb_http_output          - Set/Get HTTP output character encoding
  30  * - mb_internal_encoding    - Set/Get internal character encoding
  31  * - mb_list_encodings       - Returns an array of all supported encodings
  32  * - mb_ord                  - Returns the Unicode code point of a character
  33  * - mb_output_handler       - Callback function converts character encoding in output buffer
  34  * - mb_scrub                - Replaces ill-formed byte sequences with substitute characters
  35  * - mb_strlen               - Get string length
  36  * - mb_strpos               - Find position of first occurrence of string in a string
  37  * - mb_strrpos              - Find position of last occurrence of a string in a string
  38  * - mb_strtolower           - Make a string lowercase
  39  * - mb_strtoupper           - Make a string uppercase
  40  * - mb_substitute_character - Set/Get substitution character
  41  * - mb_substr               - Get part of string
  42  * - mb_stripos              - Finds position of first occurrence of a string within another, case insensitive
  43  * - mb_stristr              - Finds first occurrence of a string within another, case insensitive
  44  * - mb_strrchr              - Finds the last occurrence of a character in a string within another
  45  * - mb_strrichr             - Finds the last occurrence of a character in a string within another, case insensitive
  46  * - mb_strripos             - Finds position of last occurrence of a string within another, case insensitive
  47  * - mb_strstr               - Finds first occurrence of a string within anothers
  48  * - mb_strwidth             - Return width of string
  49  * - mb_substr_count         - Count the number of substring occurrences
  50  *
  51  * Not implemented:
  52  * - mb_convert_kana         - Convert "kana" one from another ("zen-kaku", "han-kaku" and more)
  53  * - mb_ereg_*               - Regular expression with multibyte support
  54  * - mb_parse_str            - Parse GET/POST/COOKIE data and set global variable
  55  * - mb_preferred_mime_name  - Get MIME charset string
  56  * - mb_regex_encoding       - Returns current encoding for multibyte regex as string
  57  * - mb_regex_set_options    - Set/Get the default options for mbregex functions
  58  * - mb_send_mail            - Send encoded mail
  59  * - mb_split                - Split multibyte string using regular expression
  60  * - mb_strcut               - Get part of string
  61  * - mb_strimwidth           - Get truncated string with specified width
  62  *
  63  * @author Nicolas Grekas <p@tchwork.com>
  64  *
  65  * @internal
  66  */
  67 final class Mbstring
  68 {
  69     const MB_CASE_FOLD = PHP_INT_MAX;
  70
  71     private static $encodingList = array('ASCII', 'UTF-8');
  72     private static $language = 'neutral';
  73     private static $internalEncoding = 'UTF-8';
  74     private static $caseFold = array(
  75         array('µ','ſ',"\xCD\x85",'ς',"\xCF\x90","\xCF\x91","\xCF\x95","\xCF\x96","\xCF\xB0","\xCF\xB1","\xCF\xB5","\xE1\xBA\x9B","\xE1\xBE\xBE"),
  76         array('μ','s','ι',       'σ','β',       'θ',       'φ',       'π',       'κ',       'ρ',       'ε',       "\xE1\xB9\xA1",'ι'),
  77     );
  78
  79     public static function mb_convert_encoding($s, $toEncoding, $fromEncoding = null)
  80     {
  81         if (\is_array($fromEncoding) || false !== strpos($fromEncoding, ',')) {
  82             $fromEncoding = self::mb_detect_encoding($s, $fromEncoding);
  83         } else {
  84             $fromEncoding = self::getEncoding($fromEncoding);
  85         }
  86
  87         $toEncoding = self::getEncoding($toEncoding);
  88
  89         if ('BASE64' === $fromEncoding) {
  90             $s = base64_decode($s);
  91             $fromEncoding = $toEncoding;
  92         }
  93
  94         if ('BASE64' === $toEncoding) {
  95             return base64_encode($s);
  96         }
  97
  98         if ('HTML-ENTITIES' === $toEncoding || 'HTML' === $toEncoding) {
  99             if ('HTML-ENTITIES' === $fromEncoding || 'HTML' === $fromEncoding) {
 100                 $fromEncoding = 'Windows-1252';
 101             }
 102             if ('UTF-8' !== $fromEncoding) {
 103                 $s = iconv($fromEncoding, 'UTF-8//IGNORE', $s);
 104             }
 105
 106             return preg_replace_callback('/[\x80-\xFF]+/', array(__CLASS__, 'html_encoding_callback'), $s);
 107         }
 108
 109         if ('HTML-ENTITIES' === $fromEncoding) {
 110             $s = html_entity_decode($s, ENT_COMPAT, 'UTF-8');
 111             $fromEncoding = 'UTF-8';
 112         }
 113
 114         return iconv($fromEncoding, $toEncoding.'//IGNORE', $s);
 115     }
 116
 117     public static function mb_convert_variables($toEncoding, $fromEncoding, &$a = null, &$b = null, &$c = null, &$d = null, &$e = null, &$f = null)
 118     {
 119         $vars = array(&$a, &$b, &$c, &$d, &$e, &$f);
 120
 121         $ok = true;
 122         array_walk_recursive($vars, function (&$v) use (&$ok, $toEncoding, $fromEncoding) {
 123             if (false === $v = Mbstring::mb_convert_encoding($v, $toEncoding, $fromEncoding)) {
 124                 $ok = false;
 125             }
 126         });
 127
 128         return $ok ? $fromEncoding : false;
 129     }
 130
 131     public static function mb_decode_mimeheader($s)
 132     {
 133         return iconv_mime_decode($s, 2, self::$internalEncoding);
 134     }
 135
 136     public static function mb_encode_mimeheader($s, $charset = null, $transferEncoding = null, $linefeed = null, $indent = null)
 137     {
 138         trigger_error('mb_encode_mimeheader() is bugged. Please use iconv_mime_encode() instead', E_USER_WARNING);
 139     }
 140
 141     public static function mb_decode_numericentity($s, $convmap, $encoding = null)
 142     {
 143         if (null !== $s && !\is_scalar($s) && !(\is_object($s) && \method_exists($s, '__toString'))) {
 144             trigger_error('mb_decode_numericentity() expects parameter 1 to be string, '.gettype($s).' given', E_USER_WARNING);
 145             return null;
 146         }
 147
 148         if (!\is_array($convmap) || !$convmap) {
 149             return false;
 150         }
 151
 152         if (null !== $encoding && !\is_scalar($encoding)) {
 153             trigger_error('mb_decode_numericentity() expects parameter 3 to be string, '.gettype($s).' given', E_USER_WARNING);
 154             return '';  // Instead of null (cf. mb_encode_numericentity).
 155         }
 156
 157         $s = (string) $s;
 158         if ('' === $s) {
 159             return '';
 160         }
 161
 162         $encoding = self::getEncoding($encoding);
 163
 164         if ('UTF-8' === $encoding) {
 165             $encoding = null;
 166             if (!preg_match('//u', $s)) {
 167                 $s = @iconv('UTF-8', 'UTF-8//IGNORE', $s);
 168             }
 169         } else {
 170             $s = iconv($encoding, 'UTF-8//IGNORE', $s);
 171         }
 172
 173         $cnt = floor(\count($convmap) / 4) * 4;
 174
 175         for ($i = 0; $i < $cnt; $i += 4) {
 176             // collector_decode_htmlnumericentity ignores $convmap[$i + 3]
 177             $convmap[$i] += $convmap[$i + 2];
 178             $convmap[$i + 1] += $convmap[$i + 2];
 179         }
 180
 181         $s = preg_replace_callback('/&#(?:0*([0-9]+)|x0*([0-9a-fA-F]+))(?!&);?/', function (array $m) use ($cnt, $convmap) {
 182             $c = isset($m[2]) ? (int) hexdec($m[2]) : $m[1];
 183             for ($i = 0; $i < $cnt; $i += 4) {
 184                 if ($c >= $convmap[$i] && $c <= $convmap[$i + 1]) {
 185                     return Mbstring::mb_chr($c - $convmap[$i + 2]);
 186                 }
 187             }
 188             return $m[0];
 189         }, $s);
 190
 191         if (null === $encoding) {
 192             return $s;
 193         }
 194
 195         return iconv('UTF-8', $encoding.'//IGNORE', $s);
 196     }
 197
 198     public static function mb_encode_numericentity($s, $convmap, $encoding = null, $is_hex = false)
 199     {
 200         if (null !== $s && !\is_scalar($s) && !(\is_object($s) && \method_exists($s, '__toString'))) {
 201             trigger_error('mb_encode_numericentity() expects parameter 1 to be string, '.gettype($s).' given', E_USER_WARNING);
 202             return null;
 203         }
 204
 205         if (!\is_array($convmap) || !$convmap) {
 206             return false;
 207         }
 208
 209         if (null !== $encoding && !\is_scalar($encoding)) {
 210             trigger_error('mb_encode_numericentity() expects parameter 3 to be string, '.gettype($s).' given', E_USER_WARNING);
 211             return null;  // Instead of '' (cf. mb_decode_numericentity).
 212         }
 213
 214         if (null !== $is_hex && !\is_scalar($is_hex)) {
 215             trigger_error('mb_encode_numericentity() expects parameter 4 to be boolean, '.gettype($s).' given', E_USER_WARNING);
 216             return null;
 217         }
 218
 219         $s = (string) $s;
 220         if ('' === $s) {
 221             return '';
 222         }
 223
 224         $encoding = self::getEncoding($encoding);
 225
 226         if ('UTF-8' === $encoding) {
 227             $encoding = null;
 228             if (!preg_match('//u', $s)) {
 229                 $s = @iconv('UTF-8', 'UTF-8//IGNORE', $s);
 230             }
 231         } else {
 232             $s = iconv($encoding, 'UTF-8//IGNORE', $s);
 233         }
 234
 235         static $ulenMask = array("\xC0" => 2, "\xD0" => 2, "\xE0" => 3, "\xF0" => 4);
 236
 237         $cnt = floor(\count($convmap) / 4) * 4;
 238         $i = 0;
 239         $len = \strlen($s);
 240         $result = '';
 241
 242         while ($i < $len) {
 243             $ulen = $s[$i] < "\x80" ? 1 : $ulenMask[$s[$i] & "\xF0"];
 244             $uchr = substr($s, $i, $ulen);
 245             $i += $ulen;
 246             $c = self::mb_ord($uchr);
 247
 248             for ($j = 0; $j < $cnt; $j += 4) {
 249                 if ($c >= $convmap[$j] && $c <= $convmap[$j + 1]) {
 250                     $cOffset = ($c + $convmap[$j + 2]) & $convmap[$j + 3];
 251                     $result .= $is_hex ? sprintf('&#x%X;', $cOffset) : '&#'.$cOffset.';';
 252                     continue 2;
 253                 }
 254             }
 255             $result .= $uchr;
 256         }
 257
 258         if (null === $encoding) {
 259             return $result;
 260         }
 261
 262         return iconv('UTF-8', $encoding.'//IGNORE', $result);
 263     }
 264
 265     public static function mb_convert_case($s, $mode, $encoding = null)
 266     {
 267         $s = (string) $s;
 268         if ('' === $s) {
 269             return '';
 270         }
 271
 272         $encoding = self::getEncoding($encoding);
 273
 274         if ('UTF-8' === $encoding) {
 275             $encoding = null;
 276             if (!preg_match('//u', $s)) {
 277                 $s = @iconv('UTF-8', 'UTF-8//IGNORE', $s);
 278             }
 279         } else {
 280             $s = iconv($encoding, 'UTF-8//IGNORE', $s);
 281         }
 282
 283         if (MB_CASE_TITLE == $mode) {
 284             $s = preg_replace_callback('/\b\p{Ll}/u', array(__CLASS__, 'title_case_upper'), $s);
 285             $s = preg_replace_callback('/\B[\p{Lu}\p{Lt}]+/u', array(__CLASS__, 'title_case_lower'), $s);
 286         } else {
 287             if (MB_CASE_UPPER == $mode) {
 288                 static $upper = null;
 289                 if (null === $upper) {
 290                     $upper = self::getData('upperCase');
 291                 }
 292                 $map = $upper;
 293             } else {
 294                 if (self::MB_CASE_FOLD === $mode) {
 295                     $s = str_replace(self::$caseFold[0], self::$caseFold[1], $s);
 296                 }
 297
 298                 static $lower = null;
 299                 if (null === $lower) {
 300                     $lower = self::getData('lowerCase');
 301                 }
 302                 $map = $lower;
 303             }
 304
 305             static $ulenMask = array("\xC0" => 2, "\xD0" => 2, "\xE0" => 3, "\xF0" => 4);
 306
 307             $i = 0;
 308             $len = \strlen($s);
 309
 310             while ($i < $len) {
 311                 $ulen = $s[$i] < "\x80" ? 1 : $ulenMask[$s[$i] & "\xF0"];
 312                 $uchr = substr($s, $i, $ulen);
 313                 $i += $ulen;
 314
 315                 if (isset($map[$uchr])) {
 316                     $uchr = $map[$uchr];
 317                     $nlen = \strlen($uchr);
 318
 319                     if ($nlen == $ulen) {
 320                         $nlen = $i;
 321                         do {
 322                             $s[--$nlen] = $uchr[--$ulen];
 323                         } while ($ulen);
 324                     } else {
 325                         $s = substr_replace($s, $uchr, $i - $ulen, $ulen);
 326                         $len += $nlen - $ulen;
 327                         $i   += $nlen - $ulen;
 328                     }
 329                 }
 330             }
 331         }
 332
 333         if (null === $encoding) {
 334             return $s;
 335         }
 336
 337         return iconv('UTF-8', $encoding.'//IGNORE', $s);
 338     }
 339
 340     public static function mb_internal_encoding($encoding = null)
 341     {
 342         if (null === $encoding) {
 343             return self::$internalEncoding;
 344         }
 345
 346         $encoding = self::getEncoding($encoding);
 347
 348         if ('UTF-8' === $encoding || false !== @iconv($encoding, $encoding, ' ')) {
 349             self::$internalEncoding = $encoding;
 350
 351             return true;
 352         }
 353
 354         return false;
 355     }
 356
 357     public static function mb_language($lang = null)
 358     {
 359         if (null === $lang) {
 360             return self::$language;
 361         }
 362
 363         switch ($lang = strtolower($lang)) {
 364             case 'uni':
 365             case 'neutral':
 366                 self::$language = $lang;
 367
 368                 return true;
 369         }
 370
 371         return false;
 372     }
 373
 374     public static function mb_list_encodings()
 375     {
 376         return array('UTF-8');
 377     }
 378
 379     public static function mb_encoding_aliases($encoding)
 380     {
 381         switch (strtoupper($encoding)) {
 382             case 'UTF8':
 383             case 'UTF-8':
 384                 return array('utf8');
 385         }
 386
 387         return false;
 388     }
 389
 390     public static function mb_check_encoding($var = null, $encoding = null)
 391     {
 392         if (null === $encoding) {
 393             if (null === $var) {
 394                 return false;
 395             }
 396             $encoding = self::$internalEncoding;
 397         }
 398
 399         return self::mb_detect_encoding($var, array($encoding)) || false !== @iconv($encoding, $encoding, $var);
 400     }
 401
 402     public static function mb_detect_encoding($str, $encodingList = null, $strict = false)
 403     {
 404         if (null === $encodingList) {
 405             $encodingList = self::$encodingList;
 406         } else {
 407             if (!\is_array($encodingList)) {
 408                 $encodingList = array_map('trim', explode(',', $encodingList));
 409             }
 410             $encodingList = array_map('strtoupper', $encodingList);
 411         }
 412
 413         foreach ($encodingList as $enc) {
 414             switch ($enc) {
 415                 case 'ASCII':
 416                     if (!preg_match('/[\x80-\xFF]/', $str)) {
 417                         return $enc;
 418                     }
 419                     break;
 420
 421                 case 'UTF8':
 422                 case 'UTF-8':
 423                     if (preg_match('//u', $str)) {
 424                         return 'UTF-8';
 425                     }
 426                     break;
 427
 428                 default:
 429                     if (0 === strncmp($enc, 'ISO-8859-', 9)) {
 430                         return $enc;
 431                     }
 432             }
 433         }
 434
 435         return false;
 436     }
 437
 438     public static function mb_detect_order($encodingList = null)
 439     {
 440         if (null === $encodingList) {
 441             return self::$encodingList;
 442         }
 443
 444         if (!\is_array($encodingList)) {
 445             $encodingList = array_map('trim', explode(',', $encodingList));
 446         }
 447         $encodingList = array_map('strtoupper', $encodingList);
 448
 449         foreach ($encodingList as $enc) {
 450             switch ($enc) {
 451                 default:
 452                     if (strncmp($enc, 'ISO-8859-', 9)) {
 453                         return false;
 454                     }
 455                 case 'ASCII':
 456                 case 'UTF8':
 457                 case 'UTF-8':
 458             }
 459         }
 460
 461         self::$encodingList = $encodingList;
 462
 463         return true;
 464     }
 465
 466     public static function mb_strlen($s, $encoding = null)
 467     {
 468         $encoding = self::getEncoding($encoding);
 469         if ('CP850' === $encoding || 'ASCII' === $encoding) {
 470             return \strlen($s);
 471         }
 472
 473         return @iconv_strlen($s, $encoding);
 474     }
 475
 476     public static function mb_strpos($haystack, $needle, $offset = 0, $encoding = null)
 477     {
 478         $encoding = self::getEncoding($encoding);
 479         if ('CP850' === $encoding || 'ASCII' === $encoding) {
 480             return strpos($haystack, $needle, $offset);
 481         }
 482
 483         $needle = (string) $needle;
 484         if ('' === $needle) {
 485             trigger_error(__METHOD__.': Empty delimiter', E_USER_WARNING);
 486
 487             return false;
 488         }
 489
 490         return iconv_strpos($haystack, $needle, $offset, $encoding);
 491     }
 492
 493     public static function mb_strrpos($haystack, $needle, $offset = 0, $encoding = null)
 494     {
 495         $encoding = self::getEncoding($encoding);
 496         if ('CP850' === $encoding || 'ASCII' === $encoding) {
 497             return strrpos($haystack, $needle, $offset);
 498         }
 499
 500         if ($offset != (int) $offset) {
 501             $offset = 0;
 502         } elseif ($offset = (int) $offset) {
 503             if ($offset < 0) {
 504                 $haystack = self::mb_substr($haystack, 0, $offset, $encoding);
 505                 $offset = 0;
 506             } else {
 507                 $haystack = self::mb_substr($haystack, $offset, 2147483647, $encoding);
 508             }
 509         }
 510
 511         $pos = iconv_strrpos($haystack, $needle, $encoding);
 512
 513         return false !== $pos ? $offset + $pos : false;
 514     }
 515
 516     public static function mb_strtolower($s, $encoding = null)
 517     {
 518         return self::mb_convert_case($s, MB_CASE_LOWER, $encoding);
 519     }
 520
 521     public static function mb_strtoupper($s, $encoding = null)
 522     {
 523         return self::mb_convert_case($s, MB_CASE_UPPER, $encoding);
 524     }
 525
 526     public static function mb_substitute_character($c = null)
 527     {
 528         if (0 === strcasecmp($c, 'none')) {
 529             return true;
 530         }
 531
 532         return null !== $c ? false : 'none';
 533     }
 534
 535     public static function mb_substr($s, $start, $length = null, $encoding = null)
 536     {
 537         $encoding = self::getEncoding($encoding);
 538         if ('CP850' === $encoding || 'ASCII' === $encoding) {
 539             return substr($s, $start, null === $length ? 2147483647 : $length);
 540         }
 541
 542         if ($start < 0) {
 543             $start = iconv_strlen($s, $encoding) + $start;
 544             if ($start < 0) {
 545                 $start = 0;
 546             }
 547         }
 548
 549         if (null === $length) {
 550             $length = 2147483647;
 551         } elseif ($length < 0) {
 552             $length = iconv_strlen($s, $encoding) + $length - $start;
 553             if ($length < 0) {
 554                 return '';
 555             }
 556         }
 557
 558         return (string) iconv_substr($s, $start, $length, $encoding);
 559     }
 560
 561     public static function mb_stripos($haystack, $needle, $offset = 0, $encoding = null)
 562     {
 563         $haystack = self::mb_convert_case($haystack, self::MB_CASE_FOLD, $encoding);
 564         $needle = self::mb_convert_case($needle, self::MB_CASE_FOLD, $encoding);
 565
 566         return self::mb_strpos($haystack, $needle, $offset, $encoding);
 567     }
 568
 569     public static function mb_stristr($haystack, $needle, $part = false, $encoding = null)
 570     {
 571         $pos = self::mb_stripos($haystack, $needle, 0, $encoding);
 572
 573         return self::getSubpart($pos, $part, $haystack, $encoding);
 574     }
 575
 576     public static function mb_strrchr($haystack, $needle, $part = false, $encoding = null)
 577     {
 578         $encoding = self::getEncoding($encoding);
 579         if ('CP850' === $encoding || 'ASCII' === $encoding) {
 580             return strrchr($haystack, $needle, $part);
 581         }
 582         $needle = self::mb_substr($needle, 0, 1, $encoding);
 583         $pos = iconv_strrpos($haystack, $needle, $encoding);
 584
 585         return self::getSubpart($pos, $part, $haystack, $encoding);
 586     }
 587
 588     public static function mb_strrichr($haystack, $needle, $part = false, $encoding = null)
 589     {
 590         $needle = self::mb_substr($needle, 0, 1, $encoding);
 591         $pos = self::mb_strripos($haystack, $needle, $encoding);
 592
 593         return self::getSubpart($pos, $part, $haystack, $encoding);
 594     }
 595
 596     public static function mb_strripos($haystack, $needle, $offset = 0, $encoding = null)
 597     {
 598         $haystack = self::mb_convert_case($haystack, self::MB_CASE_FOLD, $encoding);
 599         $needle = self::mb_convert_case($needle, self::MB_CASE_FOLD, $encoding);
 600
 601         return self::mb_strrpos($haystack, $needle, $offset, $encoding);
 602     }
 603
 604     public static function mb_strstr($haystack, $needle, $part = false, $encoding = null)
 605     {
 606         $pos = strpos($haystack, $needle);
 607         if (false === $pos) {
 608             return false;
 609         }
 610         if ($part) {
 611             return substr($haystack, 0, $pos);
 612         }
 613
 614         return substr($haystack, $pos);
 615     }
 616
 617     public static function mb_get_info($type = 'all')
 618     {
 619         $info = array(
 620             'internal_encoding' => self::$internalEncoding,
 621             'http_output' => 'pass',
 622             'http_output_conv_mimetypes' => '^(text/|application/xhtml\+xml)',
 623             'func_overload' => 0,
 624             'func_overload_list' => 'no overload',
 625             'mail_charset' => 'UTF-8',
 626             'mail_header_encoding' => 'BASE64',
 627             'mail_body_encoding' => 'BASE64',
 628             'illegal_chars' => 0,
 629             'encoding_translation' => 'Off',
 630             'language' => self::$language,
 631             'detect_order' => self::$encodingList,
 632             'substitute_character' => 'none',
 633             'strict_detection' => 'Off',
 634         );
 635
 636         if ('all' === $type) {
 637             return $info;
 638         }
 639         if (isset($info[$type])) {
 640             return $info[$type];
 641         }
 642
 643         return false;
 644     }
 645
 646     public static function mb_http_input($type = '')
 647     {
 648         return false;
 649     }
 650
 651     public static function mb_http_output($encoding = null)
 652     {
 653         return null !== $encoding ? 'pass' === $encoding : 'pass';
 654     }
 655
 656     public static function mb_strwidth($s, $encoding = null)
 657     {
 658         $encoding = self::getEncoding($encoding);
 659
 660         if ('UTF-8' !== $encoding) {
 661             $s = iconv($encoding, 'UTF-8//IGNORE', $s);
 662         }
 663
 664         $s = preg_replace('/[\x{1100}-\x{115F}\x{2329}\x{232A}\x{2E80}-\x{303E}\x{3040}-\x{A4CF}\x{AC00}-\x{D7A3}\x{F900}-\x{FAFF}\x{FE10}-\x{FE19}\x{FE30}-\x{FE6F}\x{FF00}-\x{FF60}\x{FFE0}-\x{FFE6}\x{20000}-\x{2FFFD}\x{30000}-\x{3FFFD}]/u', '', $s, -1, $wide);
 665
 666         return ($wide << 1) + iconv_strlen($s, 'UTF-8');
 667     }
 668
 669     public static function mb_substr_count($haystack, $needle, $encoding = null)
 670     {
 671         return substr_count($haystack, $needle);
 672     }
 673
 674     public static function mb_output_handler($contents, $status)
 675     {
 676         return $contents;
 677     }
 678
 679     public static function mb_chr($code, $encoding = null)
 680     {
 681         if (0x80 > $code %= 0x200000) {
 682             $s = \chr($code);
 683         } elseif (0x800 > $code) {
 684             $s = \chr(0xC0 | $code >> 6).\chr(0x80 | $code & 0x3F);
 685         } elseif (0x10000 > $code) {
 686             $s = \chr(0xE0 | $code >> 12).\chr(0x80 | $code >> 6 & 0x3F).\chr(0x80 | $code & 0x3F);
 687         } else {
 688             $s = \chr(0xF0 | $code >> 18).\chr(0x80 | $code >> 12 & 0x3F).\chr(0x80 | $code >> 6 & 0x3F).\chr(0x80 | $code & 0x3F);
 689         }
 690
 691         if ('UTF-8' !== $encoding = self::getEncoding($encoding)) {
 692             $s = mb_convert_encoding($s, $encoding, 'UTF-8');
 693         }
 694
 695         return $s;
 696     }
 697
 698     public static function mb_ord($s, $encoding = null)
 699     {
 700         if ('UTF-8' !== $encoding = self::getEncoding($encoding)) {
 701             $s = mb_convert_encoding($s, 'UTF-8', $encoding);
 702         }
 703
 704         $code = ($s = unpack('C*', substr($s, 0, 4))) ? $s[1] : 0;
 705         if (0xF0 <= $code) {
 706             return (($code - 0xF0) << 18) + (($s[2] - 0x80) << 12) + (($s[3] - 0x80) << 6) + $s[4] - 0x80;
 707         }
 708         if (0xE0 <= $code) {
 709             return (($code - 0xE0) << 12) + (($s[2] - 0x80) << 6) + $s[3] - 0x80;
 710         }
 711         if (0xC0 <= $code) {
 712             return (($code - 0xC0) << 6) + $s[2] - 0x80;
 713         }
 714
 715         return $code;
 716     }
 717
 718     private static function getSubpart($pos, $part, $haystack, $encoding)
 719     {
 720         if (false === $pos) {
 721             return false;
 722         }
 723         if ($part) {
 724             return self::mb_substr($haystack, 0, $pos, $encoding);
 725         }
 726
 727         return self::mb_substr($haystack, $pos, null, $encoding);
 728     }
 729
 730     private static function html_encoding_callback(array $m)
 731     {
 732         $i = 1;
 733         $entities = '';
 734         $m = unpack('C*', htmlentities($m[0], ENT_COMPAT, 'UTF-8'));
 735
 736         while (isset($m[$i])) {
 737             if (0x80 > $m[$i]) {
 738                 $entities .= \chr($m[$i++]);
 739                 continue;
 740             }
 741             if (0xF0 <= $m[$i]) {
 742                 $c = (($m[$i++] - 0xF0) << 18) + (($m[$i++] - 0x80) << 12) + (($m[$i++] - 0x80) << 6) + $m[$i++] - 0x80;
 743             } elseif (0xE0 <= $m[$i]) {
 744                 $c = (($m[$i++] - 0xE0) << 12) + (($m[$i++] - 0x80) << 6) + $m[$i++] - 0x80;
 745             } else {
 746                 $c = (($m[$i++] - 0xC0) << 6) + $m[$i++] - 0x80;
 747             }
 748
 749             $entities .= '&#'.$c.';';
 750         }
 751
 752         return $entities;
 753     }
 754
 755     private static function title_case_lower(array $s)
 756     {
 757         return self::mb_convert_case($s[0], MB_CASE_LOWER, 'UTF-8');
 758     }
 759
 760     private static function title_case_upper(array $s)
 761     {
 762         return self::mb_convert_case($s[0], MB_CASE_UPPER, 'UTF-8');
 763     }
 764
 765     private static function getData($file)
 766     {
 767         if (file_exists($file = __DIR__.'/Resources/unidata/'.$file.'.php')) {
 768             return require $file;
 769         }
 770
 771         return false;
 772     }
 773
 774     private static function getEncoding($encoding)
 775     {
 776         if (null === $encoding) {
 777             return self::$internalEncoding;
 778         }
 779
 780         $encoding = strtoupper($encoding);
 781
 782         if ('8BIT' === $encoding || 'BINARY' === $encoding) {
 783             return 'CP850';
 784         }
 785         if ('UTF8' === $encoding) {
 786             return 'UTF-8';
 787         }
 788
 789         return $encoding;
 790     }
 791 }