PageRenderTime 61ms CodeModel.GetById 19ms RepoModel.GetById 1ms app.codeStats 0ms

/elgg/mod/dokuwiki/lib/dokuwiki/inc/utf8.php

https://bitbucket.org/rhizomatik/lorea_production/
PHP | 1657 lines | 1000 code | 151 blank | 506 comment | 172 complexity | f76f73f6788b1f64b985ee3eb451a0e3 MD5 | raw file
Possible License(s): GPL-3.0, GPL-2.0, BSD-3-Clause, LGPL-2.1

Large files files are truncated, but you can click here to view the full file

  1. <?php
  2. /**
  3. * UTF8 helper functions
  4. *
  5. * @license LGPL (http://www.gnu.org/copyleft/lesser.html)
  6. * @author Andreas Gohr <andi@splitbrain.org>
  7. */
  8. /**
  9. * check for mb_string support
  10. */
  11. if(!defined('UTF8_MBSTRING')){
  12. if(function_exists('mb_substr') && !defined('UTF8_NOMBSTRING')){
  13. define('UTF8_MBSTRING',1);
  14. }else{
  15. define('UTF8_MBSTRING',0);
  16. }
  17. }
  18. if(UTF8_MBSTRING){ mb_internal_encoding('UTF-8'); }
  19. if(!function_exists('utf8_encodeFN')){
  20. /**
  21. * URL-Encode a filename to allow unicodecharacters
  22. *
  23. * Slashes are not encoded
  24. *
  25. * When the second parameter is true the string will
  26. * be encoded only if non ASCII characters are detected -
  27. * This makes it safe to run it multiple times on the
  28. * same string (default is true)
  29. *
  30. * @author Andreas Gohr <andi@splitbrain.org>
  31. * @see urlencode
  32. */
  33. function utf8_encodeFN($file,$safe=true){
  34. if($safe && preg_match('#^[a-zA-Z0-9/_\-.%]+$#',$file)){
  35. return $file;
  36. }
  37. $file = urlencode($file);
  38. $file = str_replace('%2F','/',$file);
  39. return $file;
  40. }
  41. }
  42. if(!function_exists('utf8_decodeFN')){
  43. /**
  44. * URL-Decode a filename
  45. *
  46. * This is just a wrapper around urldecode
  47. *
  48. * @author Andreas Gohr <andi@splitbrain.org>
  49. * @see urldecode
  50. */
  51. function utf8_decodeFN($file){
  52. $file = urldecode($file);
  53. return $file;
  54. }
  55. }
  56. if(!function_exists('utf8_isASCII')){
  57. /**
  58. * Checks if a string contains 7bit ASCII only
  59. *
  60. * @author Andreas Haerter <netzmeister@andreas-haerter.de>
  61. */
  62. function utf8_isASCII($str){
  63. return (preg_match('/(?:[^\x00-\x7F])/', $str) !== 1);
  64. }
  65. }
  66. if(!function_exists('utf8_strip')){
  67. /**
  68. * Strips all highbyte chars
  69. *
  70. * Returns a pure ASCII7 string
  71. *
  72. * @author Andreas Gohr <andi@splitbrain.org>
  73. */
  74. function utf8_strip($str){
  75. $ascii = '';
  76. $len = strlen($str);
  77. for($i=0; $i<$len; $i++){
  78. if(ord($str{$i}) <128){
  79. $ascii .= $str{$i};
  80. }
  81. }
  82. return $ascii;
  83. }
  84. }
  85. if(!function_exists('utf8_check')){
  86. /**
  87. * Tries to detect if a string is in Unicode encoding
  88. *
  89. * @author <bmorel@ssi.fr>
  90. * @link http://www.php.net/manual/en/function.utf8-encode.php
  91. */
  92. function utf8_check($Str) {
  93. $len = strlen($Str);
  94. for ($i=0; $i<$len; $i++) {
  95. $b = ord($Str[$i]);
  96. if ($b < 0x80) continue; # 0bbbbbbb
  97. elseif (($b & 0xE0) == 0xC0) $n=1; # 110bbbbb
  98. elseif (($b & 0xF0) == 0xE0) $n=2; # 1110bbbb
  99. elseif (($b & 0xF8) == 0xF0) $n=3; # 11110bbb
  100. elseif (($b & 0xFC) == 0xF8) $n=4; # 111110bb
  101. elseif (($b & 0xFE) == 0xFC) $n=5; # 1111110b
  102. else return false; # Does not match any model
  103. for ($j=0; $j<$n; $j++) { # n bytes matching 10bbbbbb follow ?
  104. if ((++$i == $len) || ((ord($Str[$i]) & 0xC0) != 0x80))
  105. return false;
  106. }
  107. }
  108. return true;
  109. }
  110. }
  111. if(!function_exists('utf8_strlen')){
  112. /**
  113. * Unicode aware replacement for strlen()
  114. *
  115. * utf8_decode() converts characters that are not in ISO-8859-1
  116. * to '?', which, for the purpose of counting, is alright - It's
  117. * even faster than mb_strlen.
  118. *
  119. * @author <chernyshevsky at hotmail dot com>
  120. * @see strlen()
  121. * @see utf8_decode()
  122. */
  123. function utf8_strlen($string){
  124. return strlen(utf8_decode($string));
  125. }
  126. }
  127. if(!function_exists('utf8_substr')){
  128. /**
  129. * UTF-8 aware alternative to substr
  130. *
  131. * Return part of a string given character offset (and optionally length)
  132. *
  133. * @author Harry Fuecks <hfuecks@gmail.com>
  134. * @author Chris Smith <chris@jalakai.co.uk>
  135. * @param string
  136. * @param integer number of UTF-8 characters offset (from left)
  137. * @param integer (optional) length in UTF-8 characters from offset
  138. * @return mixed string or false if failure
  139. */
  140. function utf8_substr($str, $offset, $length = null) {
  141. if(UTF8_MBSTRING){
  142. if( $length === null ){
  143. return mb_substr($str, $offset);
  144. }else{
  145. return mb_substr($str, $offset, $length);
  146. }
  147. }
  148. /*
  149. * Notes:
  150. *
  151. * no mb string support, so we'll use pcre regex's with 'u' flag
  152. * pcre only supports repetitions of less than 65536, in order to accept up to MAXINT values for
  153. * offset and length, we'll repeat a group of 65535 characters when needed (ok, up to MAXINT-65536)
  154. *
  155. * substr documentation states false can be returned in some cases (e.g. offset > string length)
  156. * mb_substr never returns false, it will return an empty string instead.
  157. *
  158. * calculating the number of characters in the string is a relatively expensive operation, so
  159. * we only carry it out when necessary. It isn't necessary for +ve offsets and no specified length
  160. */
  161. // cast parameters to appropriate types to avoid multiple notices/warnings
  162. $str = (string)$str; // generates E_NOTICE for PHP4 objects, but not PHP5 objects
  163. $offset = (int)$offset;
  164. if (!is_null($length)) $length = (int)$length;
  165. // handle trivial cases
  166. if ($length === 0) return '';
  167. if ($offset < 0 && $length < 0 && $length < $offset) return '';
  168. $offset_pattern = '';
  169. $length_pattern = '';
  170. // normalise -ve offsets (we could use a tail anchored pattern, but they are horribly slow!)
  171. if ($offset < 0) {
  172. $strlen = strlen(utf8_decode($str)); // see notes
  173. $offset = $strlen + $offset;
  174. if ($offset < 0) $offset = 0;
  175. }
  176. // establish a pattern for offset, a non-captured group equal in length to offset
  177. if ($offset > 0) {
  178. $Ox = (int)($offset/65535);
  179. $Oy = $offset%65535;
  180. if ($Ox) $offset_pattern = '(?:.{65535}){'.$Ox.'}';
  181. $offset_pattern = '^(?:'.$offset_pattern.'.{'.$Oy.'})';
  182. } else {
  183. $offset_pattern = '^'; // offset == 0; just anchor the pattern
  184. }
  185. // establish a pattern for length
  186. if (is_null($length)) {
  187. $length_pattern = '(.*)$'; // the rest of the string
  188. } else {
  189. if (!isset($strlen)) $strlen = strlen(utf8_decode($str)); // see notes
  190. if ($offset > $strlen) return ''; // another trivial case
  191. if ($length > 0) {
  192. $length = min($strlen-$offset, $length); // reduce any length that would go passed the end of the string
  193. $Lx = (int)($length/65535);
  194. $Ly = $length%65535;
  195. // +ve length requires ... a captured group of length characters
  196. if ($Lx) $length_pattern = '(?:.{65535}){'.$Lx.'}';
  197. $length_pattern = '('.$length_pattern.'.{'.$Ly.'})';
  198. } else if ($length < 0) {
  199. if ($length < ($offset - $strlen)) return '';
  200. $Lx = (int)((-$length)/65535);
  201. $Ly = (-$length)%65535;
  202. // -ve length requires ... capture everything except a group of -length characters
  203. // anchored at the tail-end of the string
  204. if ($Lx) $length_pattern = '(?:.{65535}){'.$Lx.'}';
  205. $length_pattern = '(.*)(?:'.$length_pattern.'.{'.$Ly.'})$';
  206. }
  207. }
  208. if (!preg_match('#'.$offset_pattern.$length_pattern.'#us',$str,$match)) return '';
  209. return $match[1];
  210. }
  211. }
  212. if(!function_exists('utf8_substr_replace')){
  213. /**
  214. * Unicode aware replacement for substr_replace()
  215. *
  216. * @author Andreas Gohr <andi@splitbrain.org>
  217. * @see substr_replace()
  218. */
  219. function utf8_substr_replace($string, $replacement, $start , $length=0 ){
  220. $ret = '';
  221. if($start>0) $ret .= utf8_substr($string, 0, $start);
  222. $ret .= $replacement;
  223. $ret .= utf8_substr($string, $start+$length);
  224. return $ret;
  225. }
  226. }
  227. if(!function_exists('utf8_ltrim')){
  228. /**
  229. * Unicode aware replacement for ltrim()
  230. *
  231. * @author Andreas Gohr <andi@splitbrain.org>
  232. * @see ltrim()
  233. * @return string
  234. */
  235. function utf8_ltrim($str,$charlist=''){
  236. if($charlist == '') return ltrim($str);
  237. //quote charlist for use in a characterclass
  238. $charlist = preg_replace('!([\\\\\\-\\]\\[/])!','\\\${1}',$charlist);
  239. return preg_replace('/^['.$charlist.']+/u','',$str);
  240. }
  241. }
  242. if(!function_exists('utf8_rtrim')){
  243. /**
  244. * Unicode aware replacement for rtrim()
  245. *
  246. * @author Andreas Gohr <andi@splitbrain.org>
  247. * @see rtrim()
  248. * @return string
  249. */
  250. function utf8_rtrim($str,$charlist=''){
  251. if($charlist == '') return rtrim($str);
  252. //quote charlist for use in a characterclass
  253. $charlist = preg_replace('!([\\\\\\-\\]\\[/])!','\\\${1}',$charlist);
  254. return preg_replace('/['.$charlist.']+$/u','',$str);
  255. }
  256. }
  257. if(!function_exists('utf8_trim')){
  258. /**
  259. * Unicode aware replacement for trim()
  260. *
  261. * @author Andreas Gohr <andi@splitbrain.org>
  262. * @see trim()
  263. * @return string
  264. */
  265. function utf8_trim($str,$charlist='') {
  266. if($charlist == '') return trim($str);
  267. return utf8_ltrim(utf8_rtrim($str,$charlist),$charlist);
  268. }
  269. }
  270. if(!function_exists('utf8_strtolower')){
  271. /**
  272. * This is a unicode aware replacement for strtolower()
  273. *
  274. * Uses mb_string extension if available
  275. *
  276. * @author Leo Feyer <leo@typolight.org>
  277. * @see strtolower()
  278. * @see utf8_strtoupper()
  279. */
  280. function utf8_strtolower($string){
  281. if(UTF8_MBSTRING) return mb_strtolower($string,'utf-8');
  282. global $UTF8_UPPER_TO_LOWER;
  283. return strtr($string,$UTF8_UPPER_TO_LOWER);
  284. }
  285. }
  286. if(!function_exists('utf8_strtoupper')){
  287. /**
  288. * This is a unicode aware replacement for strtoupper()
  289. *
  290. * Uses mb_string extension if available
  291. *
  292. * @author Leo Feyer <leo@typolight.org>
  293. * @see strtoupper()
  294. * @see utf8_strtoupper()
  295. */
  296. function utf8_strtoupper($string){
  297. if(UTF8_MBSTRING) return mb_strtoupper($string,'utf-8');
  298. global $UTF8_LOWER_TO_UPPER;
  299. return strtr($string,$UTF8_LOWER_TO_UPPER);
  300. }
  301. }
  302. if(!function_exists('utf8_ucfirst')){
  303. /**
  304. * UTF-8 aware alternative to ucfirst
  305. * Make a string's first character uppercase
  306. *
  307. * @author Harry Fuecks
  308. * @param string
  309. * @return string with first character as upper case (if applicable)
  310. */
  311. function utf8_ucfirst($str){
  312. switch ( utf8_strlen($str) ) {
  313. case 0:
  314. return '';
  315. case 1:
  316. return utf8_strtoupper($str);
  317. default:
  318. preg_match('/^(.{1})(.*)$/us', $str, $matches);
  319. return utf8_strtoupper($matches[1]).$matches[2];
  320. }
  321. }
  322. }
  323. if(!function_exists('utf8_ucwords')){
  324. /**
  325. * UTF-8 aware alternative to ucwords
  326. * Uppercase the first character of each word in a string
  327. *
  328. * @author Harry Fuecks
  329. * @param string
  330. * @return string with first char of each word uppercase
  331. * @see http://www.php.net/ucwords
  332. */
  333. function utf8_ucwords($str) {
  334. // Note: [\x0c\x09\x0b\x0a\x0d\x20] matches;
  335. // form feeds, horizontal tabs, vertical tabs, linefeeds and carriage returns
  336. // This corresponds to the definition of a "word" defined at http://www.php.net/ucwords
  337. $pattern = '/(^|([\x0c\x09\x0b\x0a\x0d\x20]+))([^\x0c\x09\x0b\x0a\x0d\x20]{1})[^\x0c\x09\x0b\x0a\x0d\x20]*/u';
  338. return preg_replace_callback($pattern, 'utf8_ucwords_callback',$str);
  339. }
  340. /**
  341. * Callback function for preg_replace_callback call in utf8_ucwords
  342. * You don't need to call this yourself
  343. *
  344. * @author Harry Fuecks
  345. * @param array of matches corresponding to a single word
  346. * @return string with first char of the word in uppercase
  347. * @see utf8_ucwords
  348. * @see utf8_strtoupper
  349. */
  350. function utf8_ucwords_callback($matches) {
  351. $leadingws = $matches[2];
  352. $ucfirst = utf8_strtoupper($matches[3]);
  353. $ucword = utf8_substr_replace(ltrim($matches[0]),$ucfirst,0,1);
  354. return $leadingws . $ucword;
  355. }
  356. }
  357. if(!function_exists('utf8_deaccent')){
  358. /**
  359. * Replace accented UTF-8 characters by unaccented ASCII-7 equivalents
  360. *
  361. * Use the optional parameter to just deaccent lower ($case = -1) or upper ($case = 1)
  362. * letters. Default is to deaccent both cases ($case = 0)
  363. *
  364. * @author Andreas Gohr <andi@splitbrain.org>
  365. */
  366. function utf8_deaccent($string,$case=0){
  367. if($case <= 0){
  368. global $UTF8_LOWER_ACCENTS;
  369. $string = strtr($string,$UTF8_LOWER_ACCENTS);
  370. }
  371. if($case >= 0){
  372. global $UTF8_UPPER_ACCENTS;
  373. $string = strtr($string,$UTF8_UPPER_ACCENTS);
  374. }
  375. return $string;
  376. }
  377. }
  378. if(!function_exists('utf8_romanize')){
  379. /**
  380. * Romanize a non-latin string
  381. *
  382. * @author Andreas Gohr <andi@splitbrain.org>
  383. */
  384. function utf8_romanize($string){
  385. if(utf8_isASCII($string)) return $string; //nothing to do
  386. global $UTF8_ROMANIZATION;
  387. return strtr($string,$UTF8_ROMANIZATION);
  388. }
  389. }
  390. if(!function_exists('utf8_stripspecials')){
  391. /**
  392. * Removes special characters (nonalphanumeric) from a UTF-8 string
  393. *
  394. * This function adds the controlchars 0x00 to 0x19 to the array of
  395. * stripped chars (they are not included in $UTF8_SPECIAL_CHARS)
  396. *
  397. * @author Andreas Gohr <andi@splitbrain.org>
  398. * @param string $string The UTF8 string to strip of special chars
  399. * @param string $repl Replace special with this string
  400. * @param string $additional Additional chars to strip (used in regexp char class)
  401. */
  402. function utf8_stripspecials($string,$repl='',$additional=''){
  403. global $UTF8_SPECIAL_CHARS;
  404. global $UTF8_SPECIAL_CHARS2;
  405. static $specials = null;
  406. if(is_null($specials)){
  407. #$specials = preg_quote(unicode_to_utf8($UTF8_SPECIAL_CHARS), '/');
  408. $specials = preg_quote($UTF8_SPECIAL_CHARS2, '/');
  409. }
  410. return preg_replace('/['.$additional.'\x00-\x19'.$specials.']/u',$repl,$string);
  411. }
  412. }
  413. if(!function_exists('utf8_strpos')){
  414. /**
  415. * This is an Unicode aware replacement for strpos
  416. *
  417. * @author Leo Feyer <leo@typolight.org>
  418. * @see strpos()
  419. * @param string
  420. * @param string
  421. * @param integer
  422. * @return integer
  423. */
  424. function utf8_strpos($haystack, $needle, $offset=0){
  425. $comp = 0;
  426. $length = null;
  427. while (is_null($length) || $length < $offset) {
  428. $pos = strpos($haystack, $needle, $offset + $comp);
  429. if ($pos === false)
  430. return false;
  431. $length = utf8_strlen(substr($haystack, 0, $pos));
  432. if ($length < $offset)
  433. $comp = $pos - $length;
  434. }
  435. return $length;
  436. }
  437. }
  438. if(!function_exists('utf8_tohtml')){
  439. /**
  440. * Encodes UTF-8 characters to HTML entities
  441. *
  442. * @author Tom N Harris <tnharris@whoopdedo.org>
  443. * @author <vpribish at shopping dot com>
  444. * @link http://www.php.net/manual/en/function.utf8-decode.php
  445. */
  446. function utf8_tohtml ($str) {
  447. $ret = '';
  448. foreach (utf8_to_unicode($str) as $cp) {
  449. if ($cp < 0x80)
  450. $ret .= chr($cp);
  451. elseif ($cp < 0x100)
  452. $ret .= "&#$cp;";
  453. else
  454. $ret .= '&#x'.dechex($cp).';';
  455. }
  456. return $ret;
  457. }
  458. }
  459. if(!function_exists('utf8_unhtml')){
  460. /**
  461. * Decodes HTML entities to UTF-8 characters
  462. *
  463. * Convert any &#..; entity to a codepoint,
  464. * The entities flag defaults to only decoding numeric entities.
  465. * Pass HTML_ENTITIES and named entities, including &amp; &lt; etc.
  466. * are handled as well. Avoids the problem that would occur if you
  467. * had to decode "&amp;#38;&#38;amp;#38;"
  468. *
  469. * unhtmlspecialchars(utf8_unhtml($s)) -> "&#38;&#38;"
  470. * utf8_unhtml(unhtmlspecialchars($s)) -> "&&amp#38;"
  471. * what it should be -> "&#38;&amp#38;"
  472. *
  473. * @author Tom N Harris <tnharris@whoopdedo.org>
  474. * @param string $str UTF-8 encoded string
  475. * @param boolean $entities Flag controlling decoding of named entities.
  476. * @return UTF-8 encoded string with numeric (and named) entities replaced.
  477. */
  478. function utf8_unhtml($str, $entities=null) {
  479. static $decoder = null;
  480. if (is_null($decoder))
  481. $decoder = new utf8_entity_decoder();
  482. if (is_null($entities))
  483. return preg_replace_callback('/(&#([Xx])?([0-9A-Za-z]+);)/m',
  484. 'utf8_decode_numeric', $str);
  485. else
  486. return preg_replace_callback('/&(#)?([Xx])?([0-9A-Za-z]+);/m',
  487. array(&$decoder, 'decode'), $str);
  488. }
  489. }
  490. if(!function_exists('utf8_decode_numeric')){
  491. function utf8_decode_numeric($ent) {
  492. switch ($ent[2]) {
  493. case 'X':
  494. case 'x':
  495. $cp = hexdec($ent[3]);
  496. break;
  497. default:
  498. $cp = intval($ent[3]);
  499. break;
  500. }
  501. return unicode_to_utf8(array($cp));
  502. }
  503. }
  504. if(!class_exists('utf8_entity_decoder')){
  505. class utf8_entity_decoder {
  506. var $table;
  507. function utf8_entity_decoder() {
  508. $table = get_html_translation_table(HTML_ENTITIES);
  509. $table = array_flip($table);
  510. $this->table = array_map(array(&$this,'makeutf8'), $table);
  511. }
  512. function makeutf8($c) {
  513. return unicode_to_utf8(array(ord($c)));
  514. }
  515. function decode($ent) {
  516. if ($ent[1] == '#') {
  517. return utf8_decode_numeric($ent);
  518. } elseif (array_key_exists($ent[0],$this->table)) {
  519. return $this->table[$ent[0]];
  520. } else {
  521. return $ent[0];
  522. }
  523. }
  524. }
  525. }
  526. if(!function_exists('utf8_to_unicode')){
  527. /**
  528. * Takes an UTF-8 string and returns an array of ints representing the
  529. * Unicode characters. Astral planes are supported ie. the ints in the
  530. * output can be > 0xFFFF. Occurrances of the BOM are ignored. Surrogates
  531. * are not allowed.
  532. *
  533. * If $strict is set to true the function returns false if the input
  534. * string isn't a valid UTF-8 octet sequence and raises a PHP error at
  535. * level E_USER_WARNING
  536. *
  537. * Note: this function has been modified slightly in this library to
  538. * trigger errors on encountering bad bytes
  539. *
  540. * @author <hsivonen@iki.fi>
  541. * @author Harry Fuecks <hfuecks@gmail.com>
  542. * @param string UTF-8 encoded string
  543. * @param boolean Check for invalid sequences?
  544. * @return mixed array of unicode code points or false if UTF-8 invalid
  545. * @see unicode_to_utf8
  546. * @link http://hsivonen.iki.fi/php-utf8/
  547. * @link http://sourceforge.net/projects/phputf8/
  548. */
  549. function utf8_to_unicode($str,$strict=false) {
  550. $mState = 0; // cached expected number of octets after the current octet
  551. // until the beginning of the next UTF8 character sequence
  552. $mUcs4 = 0; // cached Unicode character
  553. $mBytes = 1; // cached expected number of octets in the current sequence
  554. $out = array();
  555. $len = strlen($str);
  556. for($i = 0; $i < $len; $i++) {
  557. $in = ord($str{$i});
  558. if ( $mState == 0) {
  559. // When mState is zero we expect either a US-ASCII character or a
  560. // multi-octet sequence.
  561. if (0 == (0x80 & ($in))) {
  562. // US-ASCII, pass straight through.
  563. $out[] = $in;
  564. $mBytes = 1;
  565. } else if (0xC0 == (0xE0 & ($in))) {
  566. // First octet of 2 octet sequence
  567. $mUcs4 = ($in);
  568. $mUcs4 = ($mUcs4 & 0x1F) << 6;
  569. $mState = 1;
  570. $mBytes = 2;
  571. } else if (0xE0 == (0xF0 & ($in))) {
  572. // First octet of 3 octet sequence
  573. $mUcs4 = ($in);
  574. $mUcs4 = ($mUcs4 & 0x0F) << 12;
  575. $mState = 2;
  576. $mBytes = 3;
  577. } else if (0xF0 == (0xF8 & ($in))) {
  578. // First octet of 4 octet sequence
  579. $mUcs4 = ($in);
  580. $mUcs4 = ($mUcs4 & 0x07) << 18;
  581. $mState = 3;
  582. $mBytes = 4;
  583. } else if (0xF8 == (0xFC & ($in))) {
  584. /* First octet of 5 octet sequence.
  585. *
  586. * This is illegal because the encoded codepoint must be either
  587. * (a) not the shortest form or
  588. * (b) outside the Unicode range of 0-0x10FFFF.
  589. * Rather than trying to resynchronize, we will carry on until the end
  590. * of the sequence and let the later error handling code catch it.
  591. */
  592. $mUcs4 = ($in);
  593. $mUcs4 = ($mUcs4 & 0x03) << 24;
  594. $mState = 4;
  595. $mBytes = 5;
  596. } else if (0xFC == (0xFE & ($in))) {
  597. // First octet of 6 octet sequence, see comments for 5 octet sequence.
  598. $mUcs4 = ($in);
  599. $mUcs4 = ($mUcs4 & 1) << 30;
  600. $mState = 5;
  601. $mBytes = 6;
  602. } elseif($strict) {
  603. /* Current octet is neither in the US-ASCII range nor a legal first
  604. * octet of a multi-octet sequence.
  605. */
  606. trigger_error(
  607. 'utf8_to_unicode: Illegal sequence identifier '.
  608. 'in UTF-8 at byte '.$i,
  609. E_USER_WARNING
  610. );
  611. return false;
  612. }
  613. } else {
  614. // When mState is non-zero, we expect a continuation of the multi-octet
  615. // sequence
  616. if (0x80 == (0xC0 & ($in))) {
  617. // Legal continuation.
  618. $shift = ($mState - 1) * 6;
  619. $tmp = $in;
  620. $tmp = ($tmp & 0x0000003F) << $shift;
  621. $mUcs4 |= $tmp;
  622. /**
  623. * End of the multi-octet sequence. mUcs4 now contains the final
  624. * Unicode codepoint to be output
  625. */
  626. if (0 == --$mState) {
  627. /*
  628. * Check for illegal sequences and codepoints.
  629. */
  630. // From Unicode 3.1, non-shortest form is illegal
  631. if (((2 == $mBytes) && ($mUcs4 < 0x0080)) ||
  632. ((3 == $mBytes) && ($mUcs4 < 0x0800)) ||
  633. ((4 == $mBytes) && ($mUcs4 < 0x10000)) ||
  634. (4 < $mBytes) ||
  635. // From Unicode 3.2, surrogate characters are illegal
  636. (($mUcs4 & 0xFFFFF800) == 0xD800) ||
  637. // Codepoints outside the Unicode range are illegal
  638. ($mUcs4 > 0x10FFFF)) {
  639. if($strict){
  640. trigger_error(
  641. 'utf8_to_unicode: Illegal sequence or codepoint '.
  642. 'in UTF-8 at byte '.$i,
  643. E_USER_WARNING
  644. );
  645. return false;
  646. }
  647. }
  648. if (0xFEFF != $mUcs4) {
  649. // BOM is legal but we don't want to output it
  650. $out[] = $mUcs4;
  651. }
  652. //initialize UTF8 cache
  653. $mState = 0;
  654. $mUcs4 = 0;
  655. $mBytes = 1;
  656. }
  657. } elseif($strict) {
  658. /**
  659. *((0xC0 & (*in) != 0x80) && (mState != 0))
  660. * Incomplete multi-octet sequence.
  661. */
  662. trigger_error(
  663. 'utf8_to_unicode: Incomplete multi-octet '.
  664. ' sequence in UTF-8 at byte '.$i,
  665. E_USER_WARNING
  666. );
  667. return false;
  668. }
  669. }
  670. }
  671. return $out;
  672. }
  673. }
  674. if(!function_exists('unicode_to_utf8')){
  675. /**
  676. * Takes an array of ints representing the Unicode characters and returns
  677. * a UTF-8 string. Astral planes are supported ie. the ints in the
  678. * input can be > 0xFFFF. Occurrances of the BOM are ignored. Surrogates
  679. * are not allowed.
  680. *
  681. * If $strict is set to true the function returns false if the input
  682. * array contains ints that represent surrogates or are outside the
  683. * Unicode range and raises a PHP error at level E_USER_WARNING
  684. *
  685. * Note: this function has been modified slightly in this library to use
  686. * output buffering to concatenate the UTF-8 string (faster) as well as
  687. * reference the array by it's keys
  688. *
  689. * @param array of unicode code points representing a string
  690. * @param boolean Check for invalid sequences?
  691. * @return mixed UTF-8 string or false if array contains invalid code points
  692. * @author <hsivonen@iki.fi>
  693. * @author Harry Fuecks <hfuecks@gmail.com>
  694. * @see utf8_to_unicode
  695. * @link http://hsivonen.iki.fi/php-utf8/
  696. * @link http://sourceforge.net/projects/phputf8/
  697. */
  698. function unicode_to_utf8($arr,$strict=false) {
  699. if (!is_array($arr)) return '';
  700. ob_start();
  701. foreach (array_keys($arr) as $k) {
  702. if ( ($arr[$k] >= 0) && ($arr[$k] <= 0x007f) ) {
  703. # ASCII range (including control chars)
  704. echo chr($arr[$k]);
  705. } else if ($arr[$k] <= 0x07ff) {
  706. # 2 byte sequence
  707. echo chr(0xc0 | ($arr[$k] >> 6));
  708. echo chr(0x80 | ($arr[$k] & 0x003f));
  709. } else if($arr[$k] == 0xFEFF) {
  710. # Byte order mark (skip)
  711. // nop -- zap the BOM
  712. } else if ($arr[$k] >= 0xD800 && $arr[$k] <= 0xDFFF) {
  713. # Test for illegal surrogates
  714. // found a surrogate
  715. if($strict){
  716. trigger_error(
  717. 'unicode_to_utf8: Illegal surrogate '.
  718. 'at index: '.$k.', value: '.$arr[$k],
  719. E_USER_WARNING
  720. );
  721. return false;
  722. }
  723. } else if ($arr[$k] <= 0xffff) {
  724. # 3 byte sequence
  725. echo chr(0xe0 | ($arr[$k] >> 12));
  726. echo chr(0x80 | (($arr[$k] >> 6) & 0x003f));
  727. echo chr(0x80 | ($arr[$k] & 0x003f));
  728. } else if ($arr[$k] <= 0x10ffff) {
  729. # 4 byte sequence
  730. echo chr(0xf0 | ($arr[$k] >> 18));
  731. echo chr(0x80 | (($arr[$k] >> 12) & 0x3f));
  732. echo chr(0x80 | (($arr[$k] >> 6) & 0x3f));
  733. echo chr(0x80 | ($arr[$k] & 0x3f));
  734. } elseif($strict) {
  735. trigger_error(
  736. 'unicode_to_utf8: Codepoint out of Unicode range '.
  737. 'at index: '.$k.', value: '.$arr[$k],
  738. E_USER_WARNING
  739. );
  740. // out of range
  741. return false;
  742. }
  743. }
  744. $result = ob_get_contents();
  745. ob_end_clean();
  746. return $result;
  747. }
  748. }
  749. if(!function_exists('utf8_to_utf16be')){
  750. /**
  751. * UTF-8 to UTF-16BE conversion.
  752. *
  753. * Maybe really UCS-2 without mb_string due to utf8_to_unicode limits
  754. */
  755. function utf8_to_utf16be(&$str, $bom = false) {
  756. $out = $bom ? "\xFE\xFF" : '';
  757. if(UTF8_MBSTRING) return $out.mb_convert_encoding($str,'UTF-16BE','UTF-8');
  758. $uni = utf8_to_unicode($str);
  759. foreach($uni as $cp){
  760. $out .= pack('n',$cp);
  761. }
  762. return $out;
  763. }
  764. }
  765. if(!function_exists('utf16be_to_utf8')){
  766. /**
  767. * UTF-8 to UTF-16BE conversion.
  768. *
  769. * Maybe really UCS-2 without mb_string due to utf8_to_unicode limits
  770. */
  771. function utf16be_to_utf8(&$str) {
  772. $uni = unpack('n*',$str);
  773. return unicode_to_utf8($uni);
  774. }
  775. }
  776. if(!function_exists('utf8_bad_replace')){
  777. /**
  778. * Replace bad bytes with an alternative character
  779. *
  780. * ASCII character is recommended for replacement char
  781. *
  782. * PCRE Pattern to locate bad bytes in a UTF-8 string
  783. * Comes from W3 FAQ: Multilingual Forms
  784. * Note: modified to include full ASCII range including control chars
  785. *
  786. * @author Harry Fuecks <hfuecks@gmail.com>
  787. * @see http://www.w3.org/International/questions/qa-forms-utf-8
  788. * @param string to search
  789. * @param string to replace bad bytes with (defaults to '?') - use ASCII
  790. * @return string
  791. */
  792. function utf8_bad_replace($str, $replace = '') {
  793. $UTF8_BAD =
  794. '([\x00-\x7F]'. # ASCII (including control chars)
  795. '|[\xC2-\xDF][\x80-\xBF]'. # non-overlong 2-byte
  796. '|\xE0[\xA0-\xBF][\x80-\xBF]'. # excluding overlongs
  797. '|[\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}'. # straight 3-byte
  798. '|\xED[\x80-\x9F][\x80-\xBF]'. # excluding surrogates
  799. '|\xF0[\x90-\xBF][\x80-\xBF]{2}'. # planes 1-3
  800. '|[\xF1-\xF3][\x80-\xBF]{3}'. # planes 4-15
  801. '|\xF4[\x80-\x8F][\x80-\xBF]{2}'. # plane 16
  802. '|(.{1}))'; # invalid byte
  803. ob_start();
  804. while (preg_match('/'.$UTF8_BAD.'/S', $str, $matches)) {
  805. if ( !isset($matches[2])) {
  806. echo $matches[0];
  807. } else {
  808. echo $replace;
  809. }
  810. $str = substr($str,strlen($matches[0]));
  811. }
  812. $result = ob_get_contents();
  813. ob_end_clean();
  814. return $result;
  815. }
  816. }
  817. if(!function_exists('utf8_correctIdx')){
  818. /**
  819. * adjust a byte index into a utf8 string to a utf8 character boundary
  820. *
  821. * @param $str string utf8 character string
  822. * @param $i int byte index into $str
  823. * @param $next bool direction to search for boundary,
  824. * false = up (current character)
  825. * true = down (next character)
  826. *
  827. * @return int byte index into $str now pointing to a utf8 character boundary
  828. *
  829. * @author chris smith <chris@jalakai.co.uk>
  830. */
  831. function utf8_correctIdx(&$str,$i,$next=false) {
  832. if ($i <= 0) return 0;
  833. $limit = strlen($str);
  834. if ($i>=$limit) return $limit;
  835. if ($next) {
  836. while (($i<$limit) && ((ord($str[$i]) & 0xC0) == 0x80)) $i++;
  837. } else {
  838. while ($i && ((ord($str[$i]) & 0xC0) == 0x80)) $i--;
  839. }
  840. return $i;
  841. }
  842. }
  843. // only needed if no mb_string available
  844. if(!UTF8_MBSTRING){
  845. /**
  846. * UTF-8 Case lookup table
  847. *
  848. * This lookuptable defines the upper case letters to their correspponding
  849. * lower case letter in UTF-8
  850. *
  851. * @author Andreas Gohr <andi@splitbrain.org>
  852. */
  853. global $UTF8_LOWER_TO_UPPER;
  854. if(empty($UTF8_LOWER_TO_UPPER)) $UTF8_LOWER_TO_UPPER = array(
  855. "z"=>"Z","y"=>"Y","x"=>"X","w"=>"W","v"=>"V","u"=>"U","t"=>"T","s"=>"S","r"=>"R","q"=>"Q",
  856. "p"=>"P","o"=>"O","n"=>"N","m"=>"M","l"=>"L","k"=>"K","j"=>"J","i"=>"I","h"=>"H","g"=>"G",
  857. "f"=>"F","e"=>"E","d"=>"D","c"=>"C","b"=>"B","a"=>"A","ῳ"=>"ῼ","ῥ"=>"Ῥ","ῡ"=>"Ῡ","ῑ"=>"Ῑ",
  858. "ῐ"=>"Ῐ","ῃ"=>"ῌ","ι"=>"Ι","ᾳ"=>"ᾼ","ᾱ"=>"Ᾱ","ᾰ"=>"Ᾰ","ᾧ"=>"ᾯ","ᾦ"=>"ᾮ","ᾥ"=>"ᾭ","ᾤ"=>"ᾬ",
  859. "ᾣ"=>"ᾫ","ᾢ"=>"ᾪ","ᾡ"=>"ᾩ","ᾗ"=>"ᾟ","ᾖ"=>"ᾞ","ᾕ"=>"ᾝ","ᾔ"=>"ᾜ","ᾓ"=>"ᾛ","ᾒ"=>"ᾚ","ᾑ"=>"ᾙ",
  860. "ᾐ"=>"ᾘ","ᾇ"=>"ᾏ","ᾆ"=>"ᾎ","ᾅ"=>"ᾍ","ᾄ"=>"ᾌ","ᾃ"=>"ᾋ","ᾂ"=>"ᾊ","ᾁ"=>"ᾉ","ᾀ"=>"ᾈ","ώ"=>"Ώ",
  861. "ὼ"=>"Ὼ","ύ"=>"Ύ","ὺ"=>"Ὺ","ό"=>"Ό","ὸ"=>"Ὸ","ί"=>"Ί","ὶ"=>"Ὶ","ή"=>"Ή","ὴ"=>"Ὴ","έ"=>"Έ",
  862. "ὲ"=>"Ὲ","ό"=>"ឝ","ὰ"=>"ឺ","ὧ"=>"Ὧ","Ὄ"=>"὎","ὼ"=>"Ὥ","ὤ"=>"὏","ὣ"=>"Ὅ","ὢ"=>"Ὢ","ὥ"=>"Ὂ",
  863. "ὗ"=>"Ὗ","ὕ"=>"Ὕ","ὓ"=>"Ὓ","ὑ"=>"Ὑ","ὅ"=>"Ὅ","ὄ"=>"Ὄ","ὃ"=>"Ὃ","ὂ"=>"Ὂ","ὁ"=>"Ὁ","ὀ"=>"Ὀ",
  864. "១"=>"៿","ៜ"=>"៞","៾"=>"៽","៴"=>"៟","៳"=>"៝","៲"=>"៺","៹"=>"៚","៰"=>"៸","៧"=>"៯","៌"=>"៎",
  865. "ἥ"=>"Ἥ","ἤ"=>"Ἤ","ἣ"=>"Ἣ","ἢ"=>"Ἢ","ἡ"=>"Ἡ","ἕ"=>"Ἕ","ἔ"=>"Ἔ","ἓ"=>"Ἓ","ἒ"=>"Ἒ","ἑ"=>"Ἑ",
  866. "ἐ"=>"Ἐ","ἇ"=>"Ἇ","ἆ"=>"Ἆ","ἅ"=>"Ἅ","ἄ"=>"Ἄ","ἃ"=>"Ἃ","ἂ"=>"Ἂ","ἁ"=>"Ἁ","ἀ"=>"Ἀ","ỹ"=>"Ỹ",
  867. "ᝡ"=>"᝜","᝾"=>"᝴","ᝳ"=>"ᝲ","᝹"=>"ᝰ","ᝯ"=>"ᝎ","᝭"=>"ᝏ","ᝍ"=>"ᝪ","ᝊ"=>"ᝨ","ᝧ"=>"ᝌ","᝼"=>"ᝤ",
  868. "ợ"=>"Ợ","ỡ"=>"Ỡ","ở"=>"Ở","ờ"=>"Ờ","ớ"=>"Ớ","ộ"=>"Ộ","ỗ"=>"Ỗ","ổ"=>"Ổ","ồ"=>"Ồ","ố"=>"Ố",
  869. "ỏ"=>"Ỏ","ọ"=>"Ọ","ị"=>"Ị","ỉ"=>"Ỉ","ệ"=>"Ệ","ễ"=>"Ễ","ể"=>"Ể","ề"=>"Ề","ế"=>"Ế","ẽ"=>"Ẽ",
  870. "ẝ"=>"Ẻ","ẚ"=>"Ẹ","ạ"=>"ẜ","Ế"=>"Ẵ","ẳ"=>"Ẳ","ẹ"=>"Ằ","ắ"=>"Ẏ","ậ"=>"ẏ","ẍ"=>"Ẫ","Ẋ"=>"Ẩ",
  871. "ầ"=>"Ầ","ấ"=>"Ấ","ả"=>"Ả","ạ"=>"Ạ","ẛ"=>"Ṡ","ẕ"=>"Ẕ","ẓ"=>"Ẓ","ẑ"=>"Ẑ","ẏ"=>"Ẏ","ẍ"=>"Ẍ",
  872. "ẋ"=>"Ẋ","ẉ"=>"Ẉ","ẇ"=>"Ẇ","ẅ"=>"Ẅ","ẃ"=>"Ẃ","ẁ"=>"Ẁ","ṿ"=>"Ṿ","ṽ"=>"Ṽ","ṻ"=>"Ṻ","ṹ"=>"Ṹ",
  873. "ᚡ"=>"᚜","ᚾ"=>"ᚴ","ᚳ"=>"ᚲ","ᚹ"=>"ᚰ","ᚯ"=>"ᚎ","ᚭ"=>"ᚏ","ᚍ"=>"ᚪ","ᚊ"=>"ᚨ","ᚧ"=>"ᚌ","ᚼ"=>"ᚤ",
  874. "ṣ"=>"Ṣ","ṡ"=>"Ṡ","ṟ"=>"Ṟ","ṝ"=>"Ṝ","ṛ"=>"Ṛ","ṙ"=>"Ṙ","ṗ"=>"Ṗ","ṕ"=>"Ṕ","ṓ"=>"Ṓ","ṑ"=>"Ṑ",
  875. "ṏ"=>"Ṏ","ṍ"=>"Ṍ","ṋ"=>"Ṋ","ṉ"=>"Ṉ","ṇ"=>"Ṇ","ṅ"=>"Ṅ","ṃ"=>"Ṃ","ṁ"=>"Ṁ","ḿ"=>"Ḿ","ḽ"=>"Ḽ",
  876. "ḝ"=>"Ḻ","Ḛ"=>"Ḹ","ḡ"=>"Ḝ","Ḿ"=>"Ḵ","ḳ"=>"Ḳ","ḹ"=>"Ḱ","ḯ"=>"Ḏ","ḭ"=>"ḏ","ḍ"=>"Ḫ","Ḋ"=>"Ḩ",
  877. "ḧ"=>"Ḧ","ḥ"=>"Ḥ","ḣ"=>"Ḣ","ḡ"=>"Ḡ","ḟ"=>"Ḟ","ḝ"=>"Ḝ","ḛ"=>"Ḛ","ḙ"=>"Ḙ","ḗ"=>"Ḗ","ḕ"=>"Ḕ",
  878. "ḓ"=>"Ḓ","ḑ"=>"Ḑ","ḏ"=>"Ḏ","ḍ"=>"Ḍ","ḋ"=>"Ḋ","ḉ"=>"Ḉ","ḇ"=>"Ḇ","ḅ"=>"Ḅ","ḃ"=>"Ḃ","ḁ"=>"Ḁ",
  879. "ֆ"=>"Ֆ","օ"=>"Օ","ք"=>"Ք","փ"=>"Փ","ւ"=>"Ւ","ց"=>"Ց","ր"=>"Ր","տ"=>"Տ","վ"=>"Վ","ս"=>"Ս",
  880. "ռ"=>"Ռ","ջ"=>"Ջ","պ"=>"Պ","չ"=>"Չ","ո"=>"Ո","շ"=>"Շ","ն"=>"Ն","յ"=>"Յ","մ"=>"Մ","ճ"=>"Ճ",
  881. "ղ"=>"Ղ","ձ"=>"Ձ","հ"=>"Հ","կ"=>"Կ","ծ"=>"Ծ","խ"=>"Խ","լ"=>"Լ","ի"=>"Ի","ժ"=>"Ժ","թ"=>"Թ",
  882. "ը"=>"Ը","է"=>"Է","զ"=>"Զ","ե"=>"Ե","դ"=>"Դ","գ"=>"Գ","բ"=>"Բ","ա"=>"Ա","ԏ"=>"Ԏ","ԍ"=>"Ԍ",
  883. "ԋ"=>"Ԋ","ԉ"=>"Ԉ","ԇ"=>"Ԇ","ԅ"=>"Ԅ","ԃ"=>"Ԃ","ԁ"=>"Ԁ","ӹ"=>"Ӹ","ӵ"=>"Ӵ","ӳ"=>"Ӳ","ӱ"=>"Ӱ",
  884. "ӯ"=>"Ӯ","ӭ"=>"Ӭ","ӫ"=>"Ӫ","ө"=>"Ө","ӧ"=>"Ӧ","ӥ"=>"Ӥ","ӣ"=>"Ӣ","ӡ"=>"Ӡ","ӟ"=>"Ӟ","ӝ"=>"Ӝ",
  885. "ӛ"=>"Ӛ","ә"=>"Ә","ӗ"=>"Ӗ","ӕ"=>"Ӕ","ӓ"=>"Ӓ","ӑ"=>"Ӑ","ӎ"=>"Ӎ","ӌ"=>"Ӌ","ӊ"=>"Ӊ","ӈ"=>"Ӈ",
  886. "ӆ"=>"Ӆ","ӄ"=>"Ӄ","ӂ"=>"Ӂ","ҿ"=>"Ҿ","ҽ"=>"Ҽ","һ"=>"Һ","ҹ"=>"Ҹ","ҷ"=>"Ҷ","ҵ"=>"Ҵ","ҳ"=>"Ҳ",
  887. "ұ"=>"Ұ","ү"=>"Ү","ҭ"=>"Ҭ","ҫ"=>"Ҫ","ҩ"=>"Ҩ","ҧ"=>"Ҧ","ҥ"=>"Ҥ","ң"=>"Ң","ҡ"=>"Ҡ","ҟ"=>"Ҟ",
  888. "ҝ"=>"Ҝ","қ"=>"Қ","ҙ"=>"Ҙ","җ"=>"Җ","ҕ"=>"Ҕ","ғ"=>"Ғ","ґ"=>"Ґ","ҏ"=>"Ҏ","ҍ"=>"Ҍ","ҋ"=>"Ҋ",
  889. "ҁ"=>"Ҁ","ѿ"=>"Ѿ","ѽ"=>"Ѽ","ѻ"=>"Ѻ","ѹ"=>"Ѹ","ѷ"=>"Ѷ","ѵ"=>"Ѵ","ѳ"=>"Ѳ","ѱ"=>"Ѱ","ѯ"=>"Ѯ",
  890. "ѭ"=>"Ѭ","ѫ"=>"Ѫ","ѩ"=>"Ѩ","ѧ"=>"Ѧ","ѥ"=>"Ѥ","ѣ"=>"Ѣ","ѡ"=>"Ѡ","џ"=>"Џ","ў"=>"Ў","ѝ"=>"Ѝ",
  891. "ќ"=>"Ќ","ћ"=>"Ћ","њ"=>"Њ","љ"=>"Љ","ј"=>"Ј","ї"=>"Ї","і"=>"І","ѕ"=>"Ѕ","є"=>"Є","ѓ"=>"Ѓ",
  892. "ђ"=>"Ђ","ё"=>"Ё","ѐ"=>"Ѐ","я"=>"Я","ю"=>"Ю","э"=>"Э","ь"=>"Ь","ы"=>"Ы","ъ"=>"Ъ","щ"=>"Щ",
  893. "ш"=>"Ш","ч"=>"Ч","ц"=>"Ц","х"=>"Х","ф"=>"Ф","у"=>"У","т"=>"Т","с"=>"С","р"=>"Р","п"=>"П",
  894. "о"=>"О","н"=>"Н","м"=>"М","л"=>"Л","к"=>"К","й"=>"Й","и"=>"И","з"=>"З","ж"=>"Ж","е"=>"Е",
  895. "д"=>"Д","г"=>"Г","в"=>"В","б"=>"Б","а"=>"А","ϵ"=>"Ε","ϲ"=>"Σ","ϱ"=>"Ρ","ϰ"=>"Κ","ϯ"=>"Ϯ",
  896. "ϭ"=>"Ϭ","ϫ"=>"Ϫ","ϩ"=>"Ϩ","ϧ"=>"Ϧ","ϥ"=>"Ϥ","ϣ"=>"Ϣ","ϡ"=>"Ϡ","ϟ"=>"Ϟ","ϝ"=>"Ϝ","ϛ"=>"Ϛ",
  897. "ϙ"=>"Ϙ","ϖ"=>"Π","ϕ"=>"Φ","ϑ"=>"Θ","ϐ"=>"Β","ώ"=>"Ώ","ύ"=>"Ύ","ό"=>"Ό","ϋ"=>"Ϋ","ϊ"=>"Ϊ",
  898. "ω"=>"Ω","ψ"=>"Ψ","χ"=>"Χ","φ"=>"Φ","υ"=>"Υ","τ"=>"Τ","σ"=>"Σ","ς"=>"Σ","ρ"=>"Ρ","π"=>"Π",
  899. "ο"=>"Ο","ξ"=>"Ξ","ν"=>"Ν","μ"=>"Μ","λ"=>"Λ","κ"=>"Κ","ι"=>"Ι","θ"=>"Θ","η"=>"Η","ζ"=>"Ζ",
  900. "ε"=>"Ε","δ"=>"Δ","γ"=>"Γ","β"=>"Β","α"=>"Α","ί"=>"Ί","ή"=>"Ή","έ"=>"Έ","ά"=>"Ά","ʒ"=>"Ʒ",
  901. "ʋ"=>"Ʋ","ʊ"=>"Ʊ","ʈ"=>"Ʈ","ʃ"=>"Ʃ","ʀ"=>"Ʀ","ɵ"=>"Ɵ","ɲ"=>"Ɲ","ɯ"=>"Ɯ","ɩ"=>"Ɩ","ɨ"=>"Ɨ",
  902. "ɣ"=>"Ɣ","ɛ"=>"Ɛ","ə"=>"Ə","ɗ"=>"Ɗ","ɖ"=>"Ɖ","ɔ"=>"Ɔ","ɓ"=>"Ɓ","ȳ"=>"Ȳ","ȱ"=>"Ȱ","ȯ"=>"Ȯ",
  903. "ȭ"=>"Ȭ","ȫ"=>"Ȫ","ȩ"=>"Ȩ","ȧ"=>"Ȧ","ȥ"=>"Ȥ","ȣ"=>"Ȣ","ȟ"=>"Ȟ","ȝ"=>"Ȝ","ț"=>"Ț","ș"=>"Ș",
  904. "ȗ"=>"Ȗ","ȕ"=>"Ȕ","ȓ"=>"Ȓ","ȑ"=>"Ȑ","ȏ"=>"Ȏ","ȍ"=>"Ȍ","ȋ"=>"Ȋ","ȉ"=>"Ȉ","ȇ"=>"Ȇ","ȅ"=>"Ȅ",
  905. "ȃ"=>"Ȃ","ȁ"=>"Ȁ","ǿ"=>"Ǿ","ǽ"=>"Ǽ","ǻ"=>"Ǻ","ǹ"=>"Ǹ","ǵ"=>"Ǵ","dz"=>"Dz","ǯ"=>"Ǯ","ǭ"=>"Ǭ",
  906. "ǫ"=>"Ǫ","ǩ"=>"Ǩ","ǧ"=>"Ǧ","ǥ"=>"Ǥ","ǣ"=>"Ǣ","ǡ"=>"Ǡ","ǟ"=>"Ǟ","ǝ"=>"Ǝ","ǜ"=>"Ǜ","ǚ"=>"Ǚ",
  907. "ǘ"=>"Ǘ","ǖ"=>"Ǖ","ǔ"=>"Ǔ","ǒ"=>"Ǒ","ǐ"=>"Ǐ","ǎ"=>"Ǎ","nj"=>"Nj","lj"=>"Lj","dž"=>"Dž","ƿ"=>"Ƿ",
  908. "ƽ"=>"Ɵ","ƚ"=>"Ƹ","Ɯ"=>"ƾ","ƴ"=>"Ƴ","ư"=>"Ư","ƭ"=>"Ə","ƨ"=>"Ƨ","Ƽ"=>"Ƥ","ƣ"=>"Ƣ","ƥ"=>"Ơ",
  909. "ƞ"=>"Ƞ","ƙ"=>"Ƙ","ƕ"=>"Ƕ","ƒ"=>"Ƒ","ƌ"=>"Ƌ","ƈ"=>"Ƈ","ƅ"=>"Ƅ","ƃ"=>"Ƃ","ſ"=>"S","ž"=>"Ž",
  910. "ş"=>"ŝ","ź"=>"Ś","š"=>"Ŝ","ž"=>"Ŵ","ų"=>"Ų","Ź"=>"Ű","ů"=>"Ŏ","ŭ"=>"ŏ","ō"=>"Ū","Ŋ"=>"Ũ",
  911. "ŧ"=>"Ŧ","ť"=>"Ť","ţ"=>"Ţ","š"=>"Š","ş"=>"Ş","ŝ"=>"Ŝ","ś"=>"Ś","ř"=>"Ř","ŗ"=>"Ŗ","ŕ"=>"Ŕ",
  912. "œ"=>"Œ","ő"=>"Ő","ŏ"=>"Ŏ","ō"=>"Ō","ŋ"=>"Ŋ","ň"=>"Ň","ņ"=>"Ņ","ń"=>"Ń","ł"=>"Ł","ŀ"=>"Ŀ",
  913. "Ğ"=>"Ľ","ğ"=>"ĝ","ĺ"=>"Ě","ġ"=>"Ĝ","ľ"=>"Ĵ","ij"=>"IJ","Ĺ"=>"I","į"=>"Ď","ĭ"=>"ď","č"=>"Ī",
  914. "ĩ"=>"Ĩ","ħ"=>"Ħ","ĥ"=>"Ĥ","ģ"=>"Ģ","ġ"=>"Ġ","ğ"=>"Ğ","ĝ"=>"Ĝ","ě"=>"Ě","ę"=>"Ę","ė"=>"Ė",
  915. "ĕ"=>"Ĕ","ē"=>"Ē","đ"=>"Đ","ď"=>"Ď","č"=>"Č","ċ"=>"Ċ","ĉ"=>"Ĉ","ć"=>"Ć","ą"=>"Ą","ă"=>"Ă",
  916. "ā"=>"Ā","ÿ"=>"Ÿ","þ"=>"Þ","ý"=>"Ý","ü"=>"Ü","û"=>"Û","ú"=>"Ú","ù"=>"Ù","ø"=>"Ø","ö"=>"Ö",
  917. "õ"=>"Õ","ô"=>"Ô","ó"=>"Ó","ò"=>"Ò","ñ"=>"Ñ","ð"=>"Ð","ï"=>"Ï","î"=>"Î","í"=>"Í","ì"=>"Ì",
  918. "ë"=>"Ë","ê"=>"Ê","é"=>"É","è"=>"È","ç"=>"Ç","æ"=>"Æ","å"=>"Å","ä"=>"Ä","ã"=>"Ã","â"=>"Â",
  919. "á"=>"Á","à"=>"À","µ"=>"Μ","z"=>"Z","y"=>"Y","x"=>"X","w"=>"W","v"=>"V","u"=>"U","t"=>"T",
  920. "s"=>"S","r"=>"R","q"=>"Q","p"=>"P","o"=>"O","n"=>"N","m"=>"M","l"=>"L","k"=>"K","j"=>"J",
  921. "i"=>"I","h"=>"H","g"=>"G","f"=>"F","e"=>"E","d"=>"D","c"=>"C","b"=>"B","a"=>"A"
  922. );
  923. /**
  924. * UTF-8 Case lookup table
  925. *
  926. * This lookuptable defines the lower case letters to their correspponding
  927. * upper case letter in UTF-8
  928. *
  929. * @author Andreas Gohr <andi@splitbrain.org>
  930. */
  931. global $UTF8_UPPER_TO_LOWER;
  932. if(empty($UTF8_UPPER_TO_LOWER)) $UTF8_UPPER_TO_LOWER = array (
  933. "Z"=>"z","Y"=>"y","X"=>"x","W"=>"w","V"=>"v","U"=>"u","T"=>"t","S"=>"s","R"=>"r","Q"=>"q",
  934. "P"=>"p","O"=>"o","N"=>"n","M"=>"m","L"=>"l","K"=>"k","J"=>"j","I"=>"i","H"=>"h","G"=>"g",
  935. "F"=>"f","E"=>"e","D"=>"d","C"=>"c","B"=>"b","A"=>"a","ῼ"=>"ῳ","Ῥ"=>"ῥ","Ῡ"=>"ῡ","Ῑ"=>"ῑ",
  936. "Ῐ"=>"ῐ","ῌ"=>"ῃ","Ι"=>"ι","ᾼ"=>"ᾳ","Ᾱ"=>"ᾱ","Ᾰ"=>"ᾰ","ᾯ"=>"ᾧ","ᾮ"=>"ᾦ","ᾭ"=>"ᾥ","ᾬ"=>"ᾤ",
  937. "ᾫ"=>"ᾣ","ᾪ"=>"ᾢ","ᾩ"=>"ᾡ","ᾟ"=>"ᾗ","ᾞ"=>"ᾖ","ᾝ"=>"ᾕ","ᾜ"=>"ᾔ","ᾛ"=>"ᾓ","ᾚ"=>"ᾒ","ᾙ"=>"ᾑ",
  938. "ᾘ"=>"ᾐ","ᾏ"=>"ᾇ","ᾎ"=>"ᾆ","ᾍ"=>"ᾅ","ᾌ"=>"ᾄ","ᾋ"=>"ᾃ","ᾊ"=>"ᾂ","ᾉ"=>"ᾁ","ᾈ"=>"ᾀ","Ώ"=>"ώ",
  939. "Ὼ"=>"ὼ","Ύ"=>"ύ","Ὺ"=>"ὺ","Ό"=>"ό","Ὸ"=>"ὸ","Ί"=>"ί","Ὶ"=>"ὶ","Ή"=>"ή","Ὴ"=>"ὴ","Έ"=>"έ",
  940. "Ὲ"=>"ὲ","ឝ"=>"ό","ឺ"=>"ὰ","Ὧ"=>"ὧ","὎"=>"Ὄ","Ὥ"=>"ὼ","὏"=>"ὤ","Ὅ"=>"ὣ","Ὢ"=>"ὢ","Ὂ"=>"ὥ",
  941. "Ὗ"=>"ὗ","Ὕ"=>"ὕ","Ὓ"=>"ὓ","Ὑ"=>"ὑ","Ὅ"=>"ὅ","Ὄ"=>"ὄ","Ὃ"=>"ὃ","Ὂ"=>"ὂ","Ὁ"=>"ὁ","Ὀ"=>"ὀ",
  942. "៿"=>"១","៞"=>"ៜ","៽"=>"៾","៟"=>"៴","៝"=>"៳","៺"=>"៲","៚"=>"៹","៸"=>"៰","៯"=>"៧","៎"=>"៌",
  943. "Ἥ"=>"ἥ","Ἤ"=>"ἤ","Ἣ"=>"ἣ","Ἢ"=>"ἢ","Ἡ"=>"ἡ","Ἕ"=>"ἕ","Ἔ"=>"ἔ","Ἓ"=>"ἓ","Ἒ"=>"ἒ","Ἑ"=>"ἑ",
  944. "Ἐ"=>"ἐ","Ἇ"=>"ἇ","Ἆ"=>"ἆ","Ἅ"=>"ἅ","Ἄ"=>"ἄ","Ἃ"=>"ἃ","Ἂ"=>"ἂ","Ἁ"=>"ἁ","Ἀ"=>"ἀ","Ỹ"=>"ỹ",
  945. "᝜"=>"ᝡ","᝴"=>"᝾","ᝲ"=>"ᝳ","ᝰ"=>"᝹","ᝎ"=>"ᝯ","ᝏ"=>"᝭","ᝪ"=>"ᝍ","ᝨ"=>"ᝊ","ᝌ"=>"ᝧ","ᝤ"=>"᝼",
  946. "Ợ"=>"ợ","Ỡ"=>"ỡ","Ở"=>"ở","Ờ"=>"ờ","Ớ"=>"ớ","Ộ"=>"ộ","Ỗ"=>"ỗ","Ổ"=>"ổ","Ồ"=>"ồ","Ố"=>"ố",
  947. "Ỏ"=>"ỏ","Ọ"=>"ọ","Ị"=>"ị","Ỉ"=>"ỉ","Ệ"=>"ệ","Ễ"=>"ễ","Ể"=>"ể","Ề"=>"ề","Ế"=>"ế","Ẽ"=>"ẽ",
  948. "Ẻ"=>"ẝ","Ẹ"=>"ẚ","ẜ"=>"ạ","Ẵ"=>"Ế","Ẳ"=>"ẳ","Ằ"=>"ẹ","Ẏ"=>"ắ","ẏ"=>"ậ","Ẫ"=>"ẍ","Ẩ"=>"Ẋ",
  949. "Ầ"=>"ầ","Ấ"=>"ấ","Ả"=>"ả","Ạ"=>"ạ","Ṡ"=>"ẛ","Ẕ"=>"ẕ","Ẓ"=>"ẓ","Ẑ"=>"ẑ","Ẏ"=>"ẏ","Ẍ"=>"ẍ",
  950. "Ẋ"=>"ẋ","Ẉ"=>"ẉ","Ẇ"=>"ẇ","Ẅ"=>"ẅ","Ẃ"=>"ẃ","Ẁ"=>"ẁ","Ṿ"=>"ṿ","Ṽ"=>"ṽ","Ṻ"=>"ṻ","Ṹ"=>"ṹ",
  951. "᚜"=>"ᚡ","ᚴ"=>"ᚾ","ᚲ"=>"ᚳ","ᚰ"=>"ᚹ","ᚎ"=>"ᚯ","ᚏ"=>"ᚭ","ᚪ"=>"ᚍ","ᚨ"=>"ᚊ","ᚌ"=>"ᚧ","ᚤ"=>"ᚼ",
  952. "Ṣ"=>"ṣ","Ṡ"=>"ṡ","Ṟ"=>"ṟ","Ṝ"=>"ṝ","Ṛ"=>"ṛ","Ṙ"=>"ṙ","Ṗ"=>"ṗ","Ṕ"=>"ṕ","Ṓ"=>"ṓ","Ṑ"=>"ṑ",
  953. "Ṏ"=>"ṏ","Ṍ"=>"ṍ","Ṋ"=>"ṋ","Ṉ"=>"ṉ","Ṇ"=>"ṇ","Ṅ"=>"ṅ","Ṃ"=>"ṃ","Ṁ"=>"ṁ","Ḿ"=>"ḿ","Ḽ"=>"ḽ",
  954. "Ḻ"=>"ḝ","Ḹ"=>"Ḛ","Ḝ"=>"ḡ","Ḵ"=>"Ḿ","Ḳ"=>"ḳ","Ḱ"=>"ḹ","Ḏ"=>"ḯ","ḏ"=>"ḭ","Ḫ"=>"ḍ","Ḩ"=>"Ḋ",
  955. "Ḧ"=>"ḧ","Ḥ"=>"ḥ","Ḣ"=>"ḣ","Ḡ"=>"ḡ","Ḟ"=>"ḟ","Ḝ"=>"ḝ","Ḛ"=>"ḛ","Ḙ"=>"ḙ","Ḗ"=>"ḗ","Ḕ"=>"ḕ",
  956. "Ḓ"=>"ḓ","Ḑ"=>"ḑ","Ḏ"=>"ḏ","Ḍ"=>"ḍ","Ḋ"=>"ḋ","Ḉ"=>"ḉ","Ḇ"=>"ḇ","Ḅ"=>"ḅ","Ḃ"=>"ḃ","Ḁ"=>"ḁ",
  957. "Ֆ"=>"ֆ","Օ"=>"օ","Ք"=>"ք","Փ"=>"փ","Ւ"=>"ւ","Ց"=>"ց","Ր"=>"ր","Տ"=>"տ","Վ"=>"վ","Ս"=>"ս",
  958. "Ռ"=>"ռ","Ջ"=>"ջ","Պ"=>"պ","Չ"=>"չ","Ո"=>"ո","Շ"=>"շ","Ն"=>"ն","Յ"=>"յ","Մ"=>"մ","Ճ"=>"ճ",
  959. "Ղ"=>"ղ","Ձ"=>"ձ","Հ"=>"հ","Կ"=>"կ","Ծ"=>"ծ","Խ"=>"խ","Լ"=>"լ","Ի"=>"ի","Ժ"=>"ժ","Թ"=>"թ",
  960. "Ը"=>"ը","Է"=>"է","Զ"=>"զ","Ե"=>"ե","Դ"=>"դ","Գ"=>"գ","Բ"=>"բ","Ա"=>"ա","Ԏ"=>"ԏ","Ԍ"=

Large files files are truncated, but you can click here to view the full file