PageRenderTime 62ms CodeModel.GetById 21ms RepoModel.GetById 0ms app.codeStats 1ms

/wiki/inc/utf8.php

https://github.com/gbby/folders2web
PHP | 1624 lines | 988 code | 149 blank | 487 comment | 168 complexity | f8750418e7437f331696aff455367f49 MD5 | raw file
Possible License(s): GPL-2.0

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

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