PageRenderTime 80ms CodeModel.GetById 11ms app.highlight 52ms RepoModel.GetById 1ms app.codeStats 1ms

/utf8.php

https://github.com/npisenti/folders2web
PHP | 1618 lines | 984 code | 149 blank | 485 comment | 168 complexity | 0c5597f8ecec2f4b3ada4d0743facfa0 MD5 | raw file

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/**
  10 * check for mb_string support
  11 */
  12if(!defined('UTF8_MBSTRING')){
  13    if(function_exists('mb_substr') && !defined('UTF8_NOMBSTRING')){
  14        define('UTF8_MBSTRING',1);
  15    }else{
  16        define('UTF8_MBSTRING',0);
  17    }
  18}
  19
  20if(UTF8_MBSTRING){ mb_internal_encoding('UTF-8'); }
  21
  22if(!function_exists('utf8_isASCII')){
  23    /**
  24     * Checks if a string contains 7bit ASCII only
  25     *
  26     * @author Andreas Haerter <andreas.haerter@dev.mail-node.com>
  27     */
  28    function utf8_isASCII($str){
  29        return (preg_match('/(?:[^\x00-\x7F])/', $str) !== 1);
  30    }
  31}
  32
  33if(!function_exists('utf8_strip')){
  34    /**
  35     * Strips all highbyte chars
  36     *
  37     * Returns a pure ASCII7 string
  38     *
  39     * @author Andreas Gohr <andi@splitbrain.org>
  40     */
  41    function utf8_strip($str){
  42        $ascii = '';
  43        $len = strlen($str);
  44        for($i=0; $i<$len; $i++){
  45            if(ord($str{$i}) <128){
  46                $ascii .= $str{$i};
  47            }
  48        }
  49        return $ascii;
  50    }
  51}
  52
  53if(!function_exists('utf8_check')){
  54    /**
  55     * Tries to detect if a string is in Unicode encoding
  56     *
  57     * @author <bmorel@ssi.fr>
  58     * @link   http://www.php.net/manual/en/function.utf8-encode.php
  59     */
  60    function utf8_check($Str) {
  61        $len = strlen($Str);
  62        for ($i=0; $i<$len; $i++) {
  63            $b = ord($Str[$i]);
  64            if ($b < 0x80) continue; # 0bbbbbbb
  65            elseif (($b & 0xE0) == 0xC0) $n=1; # 110bbbbb
  66            elseif (($b & 0xF0) == 0xE0) $n=2; # 1110bbbb
  67            elseif (($b & 0xF8) == 0xF0) $n=3; # 11110bbb
  68            elseif (($b & 0xFC) == 0xF8) $n=4; # 111110bb
  69            elseif (($b & 0xFE) == 0xFC) $n=5; # 1111110b
  70            else return false; # Does not match any model
  71
  72            for ($j=0; $j<$n; $j++) { # n bytes matching 10bbbbbb follow ?
  73                if ((++$i == $len) || ((ord($Str[$i]) & 0xC0) != 0x80))
  74                    return false;
  75            }
  76        }
  77        return true;
  78    }
  79}
  80
  81if(!function_exists('utf8_strlen')){
  82    /**
  83     * Unicode aware replacement for strlen()
  84     *
  85     * utf8_decode() converts characters that are not in ISO-8859-1
  86     * to '?', which, for the purpose of counting, is alright - It's
  87     * even faster than mb_strlen.
  88     *
  89     * @author <chernyshevsky at hotmail dot com>
  90     * @see    strlen()
  91     * @see    utf8_decode()
  92     */
  93    function utf8_strlen($string){
  94        return strlen(utf8_decode($string));
  95    }
  96}
  97
  98if(!function_exists('utf8_substr')){
  99    /**
 100     * UTF-8 aware alternative to substr
 101     *
 102     * Return part of a string given character offset (and optionally length)
 103     *
 104     * @author Harry Fuecks <hfuecks@gmail.com>
 105     * @author Chris Smith <chris@jalakai.co.uk>
 106     * @param string
 107     * @param integer number of UTF-8 characters offset (from left)
 108     * @param integer (optional) length in UTF-8 characters from offset
 109     * @return mixed string or false if failure
 110     */
 111    function utf8_substr($str, $offset, $length = null) {
 112        if(UTF8_MBSTRING){
 113            if( $length === null ){
 114                return mb_substr($str, $offset);
 115            }else{
 116                return mb_substr($str, $offset, $length);
 117            }
 118        }
 119
 120        /*
 121         * Notes:
 122         *
 123         * no mb string support, so we'll use pcre regex's with 'u' flag
 124         * pcre only supports repetitions of less than 65536, in order to accept up to MAXINT values for
 125         * offset and length, we'll repeat a group of 65535 characters when needed (ok, up to MAXINT-65536)
 126         *
 127         * substr documentation states false can be returned in some cases (e.g. offset > string length)
 128         * mb_substr never returns false, it will return an empty string instead.
 129         *
 130         * calculating the number of characters in the string is a relatively expensive operation, so
 131         * we only carry it out when necessary. It isn't necessary for +ve offsets and no specified length
 132         */
 133
 134        // cast parameters to appropriate types to avoid multiple notices/warnings
 135        $str = (string)$str;                          // generates E_NOTICE for PHP4 objects, but not PHP5 objects
 136        $offset = (int)$offset;
 137        if (!is_null($length)) $length = (int)$length;
 138
 139        // handle trivial cases
 140        if ($length === 0) return '';
 141        if ($offset < 0 && $length < 0 && $length < $offset) return '';
 142
 143        $offset_pattern = '';
 144        $length_pattern = '';
 145
 146        // normalise -ve offsets (we could use a tail anchored pattern, but they are horribly slow!)
 147        if ($offset < 0) {
 148            $strlen = strlen(utf8_decode($str));        // see notes
 149            $offset = $strlen + $offset;
 150            if ($offset < 0) $offset = 0;
 151        }
 152
 153        // establish a pattern for offset, a non-captured group equal in length to offset
 154        if ($offset > 0) {
 155            $Ox = (int)($offset/65535);
 156            $Oy = $offset%65535;
 157
 158            if ($Ox) $offset_pattern = '(?:.{65535}){'.$Ox.'}';
 159            $offset_pattern = '^(?:'.$offset_pattern.'.{'.$Oy.'})';
 160        } else {
 161            $offset_pattern = '^';                      // offset == 0; just anchor the pattern
 162        }
 163
 164        // establish a pattern for length
 165        if (is_null($length)) {
 166            $length_pattern = '(.*)$';                  // the rest of the string
 167        } else {
 168
 169            if (!isset($strlen)) $strlen = strlen(utf8_decode($str));    // see notes
 170            if ($offset > $strlen) return '';           // another trivial case
 171
 172            if ($length > 0) {
 173
 174                $length = min($strlen-$offset, $length);  // reduce any length that would go passed the end of the string
 175
 176                $Lx = (int)($length/65535);
 177                $Ly = $length%65535;
 178
 179                // +ve length requires ... a captured group of length characters
 180                if ($Lx) $length_pattern = '(?:.{65535}){'.$Lx.'}';
 181                    $length_pattern = '('.$length_pattern.'.{'.$Ly.'})';
 182
 183            } else if ($length < 0) {
 184
 185                if ($length < ($offset - $strlen)) return '';
 186
 187                $Lx = (int)((-$length)/65535);
 188                $Ly = (-$length)%65535;
 189
 190                // -ve length requires ... capture everything except a group of -length characters
 191                //                         anchored at the tail-end of the string
 192                if ($Lx) $length_pattern = '(?:.{65535}){'.$Lx.'}';
 193                $length_pattern = '(.*)(?:'.$length_pattern.'.{'.$Ly.'})$';
 194            }
 195        }
 196
 197        if (!preg_match('#'.$offset_pattern.$length_pattern.'#us',$str,$match)) return '';
 198        return $match[1];
 199    }
 200}
 201
 202if(!function_exists('utf8_substr_replace')){
 203    /**
 204     * Unicode aware replacement for substr_replace()
 205     *
 206     * @author Andreas Gohr <andi@splitbrain.org>
 207     * @see    substr_replace()
 208     */
 209    function utf8_substr_replace($string, $replacement, $start , $length=0 ){
 210        $ret = '';
 211        if($start>0) $ret .= utf8_substr($string, 0, $start);
 212        $ret .= $replacement;
 213        $ret .= utf8_substr($string, $start+$length);
 214        return $ret;
 215    }
 216}
 217
 218if(!function_exists('utf8_ltrim')){
 219    /**
 220     * Unicode aware replacement for ltrim()
 221     *
 222     * @author Andreas Gohr <andi@splitbrain.org>
 223     * @see    ltrim()
 224     * @return string
 225     */
 226    function utf8_ltrim($str,$charlist=''){
 227        if($charlist == '') return ltrim($str);
 228
 229        //quote charlist for use in a characterclass
 230        $charlist = preg_replace('!([\\\\\\-\\]\\[/])!','\\\${1}',$charlist);
 231
 232        return preg_replace('/^['.$charlist.']+/u','',$str);
 233    }
 234}
 235
 236if(!function_exists('utf8_rtrim')){
 237    /**
 238     * Unicode aware replacement for rtrim()
 239     *
 240     * @author Andreas Gohr <andi@splitbrain.org>
 241     * @see    rtrim()
 242     * @return string
 243     */
 244    function  utf8_rtrim($str,$charlist=''){
 245        if($charlist == '') return rtrim($str);
 246
 247        //quote charlist for use in a characterclass
 248        $charlist = preg_replace('!([\\\\\\-\\]\\[/])!','\\\${1}',$charlist);
 249
 250        return preg_replace('/['.$charlist.']+$/u','',$str);
 251    }
 252}
 253
 254if(!function_exists('utf8_trim')){
 255    /**
 256     * Unicode aware replacement for trim()
 257     *
 258     * @author Andreas Gohr <andi@splitbrain.org>
 259     * @see    trim()
 260     * @return string
 261     */
 262    function  utf8_trim($str,$charlist='') {
 263        if($charlist == '') return trim($str);
 264
 265        return utf8_ltrim(utf8_rtrim($str,$charlist),$charlist);
 266    }
 267}
 268
 269if(!function_exists('utf8_strtolower')){
 270    /**
 271     * This is a unicode aware replacement for strtolower()
 272     *
 273     * Uses mb_string extension if available
 274     *
 275     * @author Leo Feyer <leo@typolight.org>
 276     * @see    strtolower()
 277     * @see    utf8_strtoupper()
 278     */
 279    function utf8_strtolower($string){
 280        if(UTF8_MBSTRING) return mb_strtolower($string,'utf-8');
 281
 282        global $UTF8_UPPER_TO_LOWER;
 283        return strtr($string,$UTF8_UPPER_TO_LOWER);
 284    }
 285}
 286
 287if(!function_exists('utf8_strtoupper')){
 288    /**
 289     * This is a unicode aware replacement for strtoupper()
 290     *
 291     * Uses mb_string extension if available
 292     *
 293     * @author Leo Feyer <leo@typolight.org>
 294     * @see    strtoupper()
 295     * @see    utf8_strtoupper()
 296     */
 297    function utf8_strtoupper($string){
 298        if(UTF8_MBSTRING) return mb_strtoupper($string,'utf-8');
 299
 300        global $UTF8_LOWER_TO_UPPER;
 301        return strtr($string,$UTF8_LOWER_TO_UPPER);
 302    }
 303}
 304
 305if(!function_exists('utf8_ucfirst')){
 306    /**
 307     * UTF-8 aware alternative to ucfirst
 308     * Make a string's first character uppercase
 309     *
 310     * @author Harry Fuecks
 311     * @param string
 312     * @return string with first character as upper case (if applicable)
 313     */
 314    function utf8_ucfirst($str){
 315        switch ( utf8_strlen($str) ) {
 316            case 0:
 317                return '';
 318            case 1:
 319                return utf8_strtoupper($str);
 320            default:
 321                preg_match('/^(.{1})(.*)$/us', $str, $matches);
 322                return utf8_strtoupper($matches[1]).$matches[2];
 323        }
 324    }
 325}
 326
 327if(!function_exists('utf8_ucwords')){
 328    /**
 329     * UTF-8 aware alternative to ucwords
 330     * Uppercase the first character of each word in a string
 331     *
 332     * @author Harry Fuecks
 333     * @param string
 334     * @return string with first char of each word uppercase
 335     * @see http://www.php.net/ucwords
 336     */
 337    function utf8_ucwords($str) {
 338        // Note: [\x0c\x09\x0b\x0a\x0d\x20] matches;
 339        // form feeds, horizontal tabs, vertical tabs, linefeeds and carriage returns
 340        // This corresponds to the definition of a "word" defined at http://www.php.net/ucwords
 341        $pattern = '/(^|([\x0c\x09\x0b\x0a\x0d\x20]+))([^\x0c\x09\x0b\x0a\x0d\x20]{1})[^\x0c\x09\x0b\x0a\x0d\x20]*/u';
 342
 343        return preg_replace_callback($pattern, 'utf8_ucwords_callback',$str);
 344    }
 345
 346    /**
 347     * Callback function for preg_replace_callback call in utf8_ucwords
 348     * You don't need to call this yourself
 349     *
 350     * @author Harry Fuecks
 351     * @param array of matches corresponding to a single word
 352     * @return string with first char of the word in uppercase
 353     * @see utf8_ucwords
 354     * @see utf8_strtoupper
 355     */
 356    function utf8_ucwords_callback($matches) {
 357        $leadingws = $matches[2];
 358        $ucfirst = utf8_strtoupper($matches[3]);
 359        $ucword = utf8_substr_replace(ltrim($matches[0]),$ucfirst,0,1);
 360        return $leadingws . $ucword;
 361    }
 362}
 363
 364if(!function_exists('utf8_deaccent')){
 365    /**
 366     * Replace accented UTF-8 characters by unaccented ASCII-7 equivalents
 367     *
 368     * Use the optional parameter to just deaccent lower ($case = -1) or upper ($case = 1)
 369     * letters. Default is to deaccent both cases ($case = 0)
 370     *
 371     * @author Andreas Gohr <andi@splitbrain.org>
 372     */
 373    function utf8_deaccent($string,$case=0){
 374        if($case <= 0){
 375            global $UTF8_LOWER_ACCENTS;
 376            $string = strtr($string,$UTF8_LOWER_ACCENTS);
 377        }
 378        if($case >= 0){
 379            global $UTF8_UPPER_ACCENTS;
 380            $string = strtr($string,$UTF8_UPPER_ACCENTS);
 381        }
 382        return $string;
 383    }
 384}
 385
 386if(!function_exists('utf8_romanize')){
 387    /**
 388     * Romanize a non-latin string
 389     *
 390     * @author Andreas Gohr <andi@splitbrain.org>
 391     */
 392    function utf8_romanize($string){
 393        if(utf8_isASCII($string)) return $string; //nothing to do
 394
 395        global $UTF8_ROMANIZATION;
 396        return strtr($string,$UTF8_ROMANIZATION);
 397    }
 398}
 399
 400if(!function_exists('utf8_stripspecials')){
 401    /**
 402     * Removes special characters (nonalphanumeric) from a UTF-8 string
 403     *
 404     * This function adds the controlchars 0x00 to 0x19 to the array of
 405     * stripped chars (they are not included in $UTF8_SPECIAL_CHARS)
 406     *
 407     * @author Andreas Gohr <andi@splitbrain.org>
 408     * @param  string $string     The UTF8 string to strip of special chars
 409     * @param  string $repl       Replace special with this string
 410     * @param  string $additional Additional chars to strip (used in regexp char class)
 411     */
 412    function utf8_stripspecials($string,$repl='',$additional=''){
 413        global $UTF8_SPECIAL_CHARS;
 414        global $UTF8_SPECIAL_CHARS2;
 415
 416        static $specials = null;
 417        if(is_null($specials)){
 418            #$specials = preg_quote(unicode_to_utf8($UTF8_SPECIAL_CHARS), '/');
 419            $specials = preg_quote($UTF8_SPECIAL_CHARS2, '/');
 420        }
 421
 422        return preg_replace('/['.$additional.'\x00-\x19'.$specials.']/u',$repl,$string);
 423    }
 424}
 425
 426if(!function_exists('utf8_strpos')){
 427    /**
 428     * This is an Unicode aware replacement for strpos
 429     *
 430     * @author Leo Feyer <leo@typolight.org>
 431     * @see    strpos()
 432     * @param  string
 433     * @param  string
 434     * @param  integer
 435     * @return integer
 436     */
 437    function utf8_strpos($haystack, $needle, $offset=0){
 438        $comp = 0;
 439        $length = null;
 440
 441        while (is_null($length) || $length < $offset) {
 442            $pos = strpos($haystack, $needle, $offset + $comp);
 443
 444            if ($pos === false)
 445                return false;
 446
 447            $length = utf8_strlen(substr($haystack, 0, $pos));
 448
 449            if ($length < $offset)
 450                $comp = $pos - $length;
 451        }
 452
 453        return $length;
 454    }
 455}
 456
 457if(!function_exists('utf8_tohtml')){
 458    /**
 459     * Encodes UTF-8 characters to HTML entities
 460     *
 461     * @author Tom N Harris <tnharris@whoopdedo.org>
 462     * @author <vpribish at shopping dot com>
 463     * @link   http://www.php.net/manual/en/function.utf8-decode.php
 464     */
 465    function utf8_tohtml ($str) {
 466        $ret = '';
 467        foreach (utf8_to_unicode($str) as $cp) {
 468            if ($cp < 0x80)
 469                $ret .= chr($cp);
 470            elseif ($cp < 0x100)
 471                $ret .= "&#$cp;";
 472            else
 473                $ret .= '&#x'.dechex($cp).';';
 474        }
 475        return $ret;
 476    }
 477}
 478
 479if(!function_exists('utf8_unhtml')){
 480    /**
 481     * Decodes HTML entities to UTF-8 characters
 482     *
 483     * Convert any &#..; entity to a codepoint,
 484     * The entities flag defaults to only decoding numeric entities.
 485     * Pass HTML_ENTITIES and named entities, including &amp; &lt; etc.
 486     * are handled as well. Avoids the problem that would occur if you
 487     * had to decode "&amp;#38;&#38;amp;#38;"
 488     *
 489     * unhtmlspecialchars(utf8_unhtml($s)) -> "&#38;&#38;"
 490     * utf8_unhtml(unhtmlspecialchars($s)) -> "&&amp#38;"
 491     * what it should be                   -> "&#38;&amp#38;"
 492     *
 493     * @author Tom N Harris <tnharris@whoopdedo.org>
 494     * @param  string  $str      UTF-8 encoded string
 495     * @param  boolean $entities Flag controlling decoding of named entities.
 496     * @return UTF-8 encoded string with numeric (and named) entities replaced.
 497     */
 498    function utf8_unhtml($str, $entities=null) {
 499        static $decoder = null;
 500        if (is_null($decoder))
 501            $decoder = new utf8_entity_decoder();
 502        if (is_null($entities))
 503            return preg_replace_callback('/(&#([Xx])?([0-9A-Za-z]+);)/m',
 504                                         'utf8_decode_numeric', $str);
 505        else
 506            return preg_replace_callback('/&(#)?([Xx])?([0-9A-Za-z]+);/m',
 507                                         array(&$decoder, 'decode'), $str);
 508    }
 509}
 510
 511if(!function_exists('utf8_decode_numeric')){
 512    function utf8_decode_numeric($ent) {
 513        switch ($ent[2]) {
 514            case 'X':
 515            case 'x':
 516                $cp = hexdec($ent[3]);
 517                break;
 518            default:
 519                $cp = intval($ent[3]);
 520                break;
 521        }
 522        return unicode_to_utf8(array($cp));
 523    }
 524}
 525
 526if(!class_exists('utf8_entity_decoder')){
 527    class utf8_entity_decoder {
 528        var $table;
 529        function utf8_entity_decoder() {
 530            $table = get_html_translation_table(HTML_ENTITIES);
 531            $table = array_flip($table);
 532            $this->table = array_map(array(&$this,'makeutf8'), $table);
 533        }
 534        function makeutf8($c) {
 535            return unicode_to_utf8(array(ord($c)));
 536        }
 537        function decode($ent) {
 538            if ($ent[1] == '#') {
 539                return utf8_decode_numeric($ent);
 540            } elseif (array_key_exists($ent[0],$this->table)) {
 541                return $this->table[$ent[0]];
 542            } else {
 543                return $ent[0];
 544            }
 545        }
 546    }
 547}
 548
 549if(!function_exists('utf8_to_unicode')){
 550    /**
 551     * Takes an UTF-8 string and returns an array of ints representing the
 552     * Unicode characters. Astral planes are supported ie. the ints in the
 553     * output can be > 0xFFFF. Occurrances of the BOM are ignored. Surrogates
 554     * are not allowed.
 555     *
 556     * If $strict is set to true the function returns false if the input
 557     * string isn't a valid UTF-8 octet sequence and raises a PHP error at
 558     * level E_USER_WARNING
 559     *
 560     * Note: this function has been modified slightly in this library to
 561     * trigger errors on encountering bad bytes
 562     *
 563     * @author <hsivonen@iki.fi>
 564     * @author Harry Fuecks <hfuecks@gmail.com>
 565     * @param  string  UTF-8 encoded string
 566     * @param  boolean Check for invalid sequences?
 567     * @return mixed array of unicode code points or false if UTF-8 invalid
 568     * @see    unicode_to_utf8
 569     * @link   http://hsivonen.iki.fi/php-utf8/
 570     * @link   http://sourceforge.net/projects/phputf8/
 571     */
 572    function utf8_to_unicode($str,$strict=false) {
 573        $mState = 0;     // cached expected number of octets after the current octet
 574                         // until the beginning of the next UTF8 character sequence
 575        $mUcs4  = 0;     // cached Unicode character
 576        $mBytes = 1;     // cached expected number of octets in the current sequence
 577
 578        $out = array();
 579
 580        $len = strlen($str);
 581
 582        for($i = 0; $i < $len; $i++) {
 583
 584            $in = ord($str{$i});
 585
 586            if ( $mState == 0) {
 587
 588                // When mState is zero we expect either a US-ASCII character or a
 589                // multi-octet sequence.
 590                if (0 == (0x80 & ($in))) {
 591                    // US-ASCII, pass straight through.
 592                    $out[] = $in;
 593                    $mBytes = 1;
 594
 595                } else if (0xC0 == (0xE0 & ($in))) {
 596                    // First octet of 2 octet sequence
 597                    $mUcs4 = ($in);
 598                    $mUcs4 = ($mUcs4 & 0x1F) << 6;
 599                    $mState = 1;
 600                    $mBytes = 2;
 601
 602                } else if (0xE0 == (0xF0 & ($in))) {
 603                    // First octet of 3 octet sequence
 604                    $mUcs4 = ($in);
 605                    $mUcs4 = ($mUcs4 & 0x0F) << 12;
 606                    $mState = 2;
 607                    $mBytes = 3;
 608
 609                } else if (0xF0 == (0xF8 & ($in))) {
 610                    // First octet of 4 octet sequence
 611                    $mUcs4 = ($in);
 612                    $mUcs4 = ($mUcs4 & 0x07) << 18;
 613                    $mState = 3;
 614                    $mBytes = 4;
 615
 616                } else if (0xF8 == (0xFC & ($in))) {
 617                    /* First octet of 5 octet sequence.
 618                     *
 619                     * This is illegal because the encoded codepoint must be either
 620                     * (a) not the shortest form or
 621                     * (b) outside the Unicode range of 0-0x10FFFF.
 622                     * Rather than trying to resynchronize, we will carry on until the end
 623                     * of the sequence and let the later error handling code catch it.
 624                     */
 625                    $mUcs4 = ($in);
 626                    $mUcs4 = ($mUcs4 & 0x03) << 24;
 627                    $mState = 4;
 628                    $mBytes = 5;
 629
 630                } else if (0xFC == (0xFE & ($in))) {
 631                    // First octet of 6 octet sequence, see comments for 5 octet sequence.
 632                    $mUcs4 = ($in);
 633                    $mUcs4 = ($mUcs4 & 1) << 30;
 634                    $mState = 5;
 635                    $mBytes = 6;
 636
 637                } elseif($strict) {
 638                    /* Current octet is neither in the US-ASCII range nor a legal first
 639                     * octet of a multi-octet sequence.
 640                     */
 641                    trigger_error(
 642                            'utf8_to_unicode: Illegal sequence identifier '.
 643                                'in UTF-8 at byte '.$i,
 644                            E_USER_WARNING
 645                        );
 646                    return false;
 647
 648                }
 649
 650            } else {
 651
 652                // When mState is non-zero, we expect a continuation of the multi-octet
 653                // sequence
 654                if (0x80 == (0xC0 & ($in))) {
 655
 656                    // Legal continuation.
 657                    $shift = ($mState - 1) * 6;
 658                    $tmp = $in;
 659                    $tmp = ($tmp & 0x0000003F) << $shift;
 660                    $mUcs4 |= $tmp;
 661
 662                    /**
 663                     * End of the multi-octet sequence. mUcs4 now contains the final
 664                     * Unicode codepoint to be output
 665                     */
 666                    if (0 == --$mState) {
 667
 668                        /*
 669                         * Check for illegal sequences and codepoints.
 670                         */
 671                        // From Unicode 3.1, non-shortest form is illegal
 672                        if (((2 == $mBytes) && ($mUcs4 < 0x0080)) ||
 673                            ((3 == $mBytes) && ($mUcs4 < 0x0800)) ||
 674                            ((4 == $mBytes) && ($mUcs4 < 0x10000)) ||
 675                            (4 < $mBytes) ||
 676                            // From Unicode 3.2, surrogate characters are illegal
 677                            (($mUcs4 & 0xFFFFF800) == 0xD800) ||
 678                            // Codepoints outside the Unicode range are illegal
 679                            ($mUcs4 > 0x10FFFF)) {
 680
 681                            if($strict){
 682                                trigger_error(
 683                                        'utf8_to_unicode: Illegal sequence or codepoint '.
 684                                            'in UTF-8 at byte '.$i,
 685                                        E_USER_WARNING
 686                                    );
 687
 688                                return false;
 689                            }
 690
 691                        }
 692
 693                        if (0xFEFF != $mUcs4) {
 694                            // BOM is legal but we don't want to output it
 695                            $out[] = $mUcs4;
 696                        }
 697
 698                        //initialize UTF8 cache
 699                        $mState = 0;
 700                        $mUcs4  = 0;
 701                        $mBytes = 1;
 702                    }
 703
 704                } elseif($strict) {
 705                    /**
 706                     *((0xC0 & (*in) != 0x80) && (mState != 0))
 707                     * Incomplete multi-octet sequence.
 708                     */
 709                    trigger_error(
 710                            'utf8_to_unicode: Incomplete multi-octet '.
 711                            '   sequence in UTF-8 at byte '.$i,
 712                            E_USER_WARNING
 713                        );
 714
 715                    return false;
 716                }
 717            }
 718        }
 719        return $out;
 720    }
 721}
 722
 723if(!function_exists('unicode_to_utf8')){
 724    /**
 725     * Takes an array of ints representing the Unicode characters and returns
 726     * a UTF-8 string. Astral planes are supported ie. the ints in the
 727     * input can be > 0xFFFF. Occurrances of the BOM are ignored. Surrogates
 728     * are not allowed.
 729     *
 730     * If $strict is set to true the function returns false if the input
 731     * array contains ints that represent surrogates or are outside the
 732     * Unicode range and raises a PHP error at level E_USER_WARNING
 733     *
 734     * Note: this function has been modified slightly in this library to use
 735     * output buffering to concatenate the UTF-8 string (faster) as well as
 736     * reference the array by it's keys
 737     *
 738     * @param  array of unicode code points representing a string
 739     * @param  boolean Check for invalid sequences?
 740     * @return mixed UTF-8 string or false if array contains invalid code points
 741     * @author <hsivonen@iki.fi>
 742     * @author Harry Fuecks <hfuecks@gmail.com>
 743     * @see    utf8_to_unicode
 744     * @link   http://hsivonen.iki.fi/php-utf8/
 745     * @link   http://sourceforge.net/projects/phputf8/
 746     */
 747    function unicode_to_utf8($arr,$strict=false) {
 748        if (!is_array($arr)) return '';
 749        ob_start();
 750
 751        foreach (array_keys($arr) as $k) {
 752
 753            if ( ($arr[$k] >= 0) && ($arr[$k] <= 0x007f) ) {
 754                # ASCII range (including control chars)
 755
 756                echo chr($arr[$k]);
 757
 758            } else if ($arr[$k] <= 0x07ff) {
 759                # 2 byte sequence
 760
 761                echo chr(0xc0 | ($arr[$k] >> 6));
 762                echo chr(0x80 | ($arr[$k] & 0x003f));
 763
 764            } else if($arr[$k] == 0xFEFF) {
 765                # Byte order mark (skip)
 766
 767                // nop -- zap the BOM
 768
 769            } else if ($arr[$k] >= 0xD800 && $arr[$k] <= 0xDFFF) {
 770                # Test for illegal surrogates
 771
 772                // found a surrogate
 773                if($strict){
 774                    trigger_error(
 775                        'unicode_to_utf8: Illegal surrogate '.
 776                            'at index: '.$k.', value: '.$arr[$k],
 777                        E_USER_WARNING
 778                        );
 779                    return false;
 780                }
 781
 782            } else if ($arr[$k] <= 0xffff) {
 783                # 3 byte sequence
 784
 785                echo chr(0xe0 | ($arr[$k] >> 12));
 786                echo chr(0x80 | (($arr[$k] >> 6) & 0x003f));
 787                echo chr(0x80 | ($arr[$k] & 0x003f));
 788
 789            } else if ($arr[$k] <= 0x10ffff) {
 790                # 4 byte sequence
 791
 792                echo chr(0xf0 | ($arr[$k] >> 18));
 793                echo chr(0x80 | (($arr[$k] >> 12) & 0x3f));
 794                echo chr(0x80 | (($arr[$k] >> 6) & 0x3f));
 795                echo chr(0x80 | ($arr[$k] & 0x3f));
 796
 797            } elseif($strict) {
 798
 799                trigger_error(
 800                    'unicode_to_utf8: Codepoint out of Unicode range '.
 801                        'at index: '.$k.', value: '.$arr[$k],
 802                    E_USER_WARNING
 803                    );
 804
 805                // out of range
 806                return false;
 807            }
 808        }
 809
 810        $result = ob_get_contents();
 811        ob_end_clean();
 812        return $result;
 813    }
 814}
 815
 816if(!function_exists('utf8_to_utf16be')){
 817    /**
 818     * UTF-8 to UTF-16BE conversion.
 819     *
 820     * Maybe really UCS-2 without mb_string due to utf8_to_unicode limits
 821     */
 822    function utf8_to_utf16be(&$str, $bom = false) {
 823        $out = $bom ? "\xFE\xFF" : '';
 824        if(UTF8_MBSTRING) return $out.mb_convert_encoding($str,'UTF-16BE','UTF-8');
 825
 826        $uni = utf8_to_unicode($str);
 827        foreach($uni as $cp){
 828            $out .= pack('n',$cp);
 829        }
 830        return $out;
 831    }
 832}
 833
 834if(!function_exists('utf16be_to_utf8')){
 835    /**
 836     * UTF-8 to UTF-16BE conversion.
 837     *
 838     * Maybe really UCS-2 without mb_string due to utf8_to_unicode limits
 839     */
 840    function utf16be_to_utf8(&$str) {
 841        $uni = unpack('n*',$str);
 842        return unicode_to_utf8($uni);
 843    }
 844}
 845
 846if(!function_exists('utf8_bad_replace')){
 847    /**
 848     * Replace bad bytes with an alternative character
 849     *
 850     * ASCII character is recommended for replacement char
 851     *
 852     * PCRE Pattern to locate bad bytes in a UTF-8 string
 853     * Comes from W3 FAQ: Multilingual Forms
 854     * Note: modified to include full ASCII range including control chars
 855     *
 856     * @author Harry Fuecks <hfuecks@gmail.com>
 857     * @see http://www.w3.org/International/questions/qa-forms-utf-8
 858     * @param string to search
 859     * @param string to replace bad bytes with (defaults to '?') - use ASCII
 860     * @return string
 861     */
 862    function utf8_bad_replace($str, $replace = '') {
 863        $UTF8_BAD =
 864         '([\x00-\x7F]'.                          # ASCII (including control chars)
 865         '|[\xC2-\xDF][\x80-\xBF]'.               # non-overlong 2-byte
 866         '|\xE0[\xA0-\xBF][\x80-\xBF]'.           # excluding overlongs
 867         '|[\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}'.    # straight 3-byte
 868         '|\xED[\x80-\x9F][\x80-\xBF]'.           # excluding surrogates
 869         '|\xF0[\x90-\xBF][\x80-\xBF]{2}'.        # planes 1-3
 870         '|[\xF1-\xF3][\x80-\xBF]{3}'.            # planes 4-15
 871         '|\xF4[\x80-\x8F][\x80-\xBF]{2}'.        # plane 16
 872         '|(.{1}))';                              # invalid byte
 873        ob_start();
 874        while (preg_match('/'.$UTF8_BAD.'/S', $str, $matches)) {
 875            if ( !isset($matches[2])) {
 876                echo $matches[0];
 877            } else {
 878                echo $replace;
 879            }
 880            $str = substr($str,strlen($matches[0]));
 881        }
 882        $result = ob_get_contents();
 883        ob_end_clean();
 884        return $result;
 885    }
 886}
 887
 888if(!function_exists('utf8_correctIdx')){
 889    /**
 890     * adjust a byte index into a utf8 string to a utf8 character boundary
 891     *
 892     * @param $str   string   utf8 character string
 893     * @param $i     int      byte index into $str
 894     * @param $next  bool     direction to search for boundary,
 895     *                           false = up (current character)
 896     *                           true = down (next character)
 897     *
 898     * @return int            byte index into $str now pointing to a utf8 character boundary
 899     *
 900     * @author       chris smith <chris@jalakai.co.uk>
 901     */
 902    function utf8_correctIdx(&$str,$i,$next=false) {
 903
 904        if ($i <= 0) return 0;
 905
 906        $limit = strlen($str);
 907        if ($i>=$limit) return $limit;
 908
 909        if ($next) {
 910            while (($i<$limit) && ((ord($str[$i]) & 0xC0) == 0x80)) $i++;
 911        } else {
 912            while ($i && ((ord($str[$i]) & 0xC0) == 0x80)) $i--;
 913        }
 914
 915        return $i;
 916    }
 917}
 918
 919// only needed if no mb_string available
 920if(!UTF8_MBSTRING){
 921    /**
 922     * UTF-8 Case lookup table
 923     *
 924     * This lookuptable defines the upper case letters to their correspponding
 925     * lower case letter in UTF-8
 926     *
 927     * @author Andreas Gohr <andi@splitbrain.org>
 928     */
 929    global $UTF8_LOWER_TO_UPPER;
 930    if(empty($UTF8_LOWER_TO_UPPER)) $UTF8_LOWER_TO_UPPER = array(
 931            "z"=>"Z","y"=>"Y","x"=>"X","w"=>"W","v"=>"V","u"=>"U","t"=>"T","s"=>"S","r"=>"R","q"=>"Q",
 932            "p"=>"P","o"=>"O","n"=>"N","m"=>"M","l"=>"L","k"=>"K","j"=>"J","i"=>"I","h"=>"H","g"=>"G",
 933            "f"=>"F","e"=>"E","d"=>"D","c"=>"C","b"=>"B","a"=>"A","ῳ"=>"ῼ","ῥ"=>"Ῥ","ῡ"=>"Ῡ","ῑ"=>"Ῑ",
 934            "ῐ"=>"Ῐ","ῃ"=>"ῌ","ι"=>"Ι","ᾳ"=>"ᾼ","ᾱ"=>"Ᾱ","ᾰ"=>"Ᾰ","ᾧ"=>"ᾯ","ᾦ"=>"ᾮ","ᾥ"=>"ᾭ","ᾤ"=>"ᾬ",
 935            "ᾣ"=>"ᾫ","ᾢ"=>"ᾪ","ᾡ"=>"ᾩ","ᾗ"=>"ᾟ","ᾖ"=>"ᾞ","ᾕ"=>"ᾝ","ᾔ"=>"ᾜ","ᾓ"=>"ᾛ","ᾒ"=>"ᾚ","ᾑ"=>"ᾙ",
 936            "ᾐ"=>"ᾘ","ᾇ"=>"ᾏ","ᾆ"=>"ᾎ","ᾅ"=>"ᾍ","ᾄ"=>"ᾌ","ᾃ"=>"ᾋ","ᾂ"=>"ᾊ","ᾁ"=>"ᾉ","ᾀ"=>"ᾈ","ώ"=>"Ώ",
 937            "ὼ"=>"Ὼ","ύ"=>"Ύ","ὺ"=>"Ὺ","ό"=>"Ό","ὸ"=>"Ὸ","ί"=>"Ί","ὶ"=>"Ὶ","ή"=>"Ή","ὴ"=>"Ὴ","έ"=>"Έ",
 938            "ὲ"=>"Ὲ","ό"=>"ឝ","ὰ"=>"ឺ","ὧ"=>"Ὧ","Ὄ"=>"὎","ὼ"=>"Ὥ","ὤ"=>"὏","ὣ"=>"Ὅ","ὢ"=>"Ὢ","ὥ"=>"Ὂ",
 939            "ὗ"=>"Ὗ","ὕ"=>"Ὕ","ὓ"=>"Ὓ","ὑ"=>"Ὑ","ὅ"=>"Ὅ","ὄ"=>"Ὄ","ὃ"=>"Ὃ","ὂ"=>"Ὂ","ὁ"=>"Ὁ","ὀ"=>"Ὀ",
 940            "១"=>"៿","ៜ"=>"៞","៾"=>"៽","៴"=>"៟","៳"=>"៝","៲"=>"៺","៹"=>"៚","៰"=>"៸","៧"=>"៯","៌"=>"៎",
 941            "ἥ"=>"Ἥ","ἤ"=>"Ἤ","ἣ"=>"Ἣ","ἢ"=>"Ἢ","ἡ"=>"Ἡ","ἕ"=>"Ἕ","ἔ"=>"Ἔ","ἓ"=>"Ἓ","ἒ"=>"Ἒ","ἑ"=>"Ἑ",
 942            "ἐ"=>"Ἐ","ἇ"=>"Ἇ","ἆ"=>"Ἆ","ἅ"=>"Ἅ","ἄ"=>"Ἄ","ἃ"=>"Ἃ","ἂ"=>"Ἂ","ἁ"=>"Ἁ","ἀ"=>"Ἀ","ỹ"=>"Ỹ",
 943            "ᝡ"=>"᝜","᝾"=>"᝴","ᝳ"=>"ᝲ","᝹"=>"ᝰ","ᝯ"=>"ᝎ","᝭"=>"ᝏ","ᝍ"=>"ᝪ","ᝊ"=>"ᝨ","ᝧ"=>"ᝌ","᝼"=>"ᝤ",
 944            "ợ"=>"Ợ","ỡ"=>"Ỡ","ở"=>"Ở","ờ"=>"Ờ","ớ"=>"Ớ","ộ"=>"Ộ","ỗ"=>"Ỗ","ổ"=>"Ổ","ồ"=>"Ồ","ố"=>"Ố",
 945            "ỏ"=>"Ỏ","ọ"=>"Ọ","ị"=>"Ị","ỉ"=>"Ỉ","ệ"=>"Ệ","ễ"=>"Ễ","ể"=>"Ể","ề"=>"Ề","ế"=>"Ế","ẽ"=>"Ẽ",
 946            "ẝ"=>"Ẻ","ẚ"=>"Ẹ","ạ"=>"ẜ","Ế"=>"Ẵ","ẳ"=>"Ẳ","ẹ"=>"Ằ","ắ"=>"Ẏ","ậ"=>"ẏ","ẍ"=>"Ẫ","Ẋ"=>"Ẩ",
 947            "ầ"=>"Ầ","ấ"=>"Ấ","ả"=>"Ả","ạ"=>"Ạ","ẛ"=>"Ṡ","ẕ"=>"Ẕ","ẓ"=>"Ẓ","ẑ"=>"Ẑ","ẏ"=>"Ẏ","ẍ"=>"Ẍ",
 948            "ẋ"=>"Ẋ","ẉ"=>"Ẉ","ẇ"=>"Ẇ","ẅ"=>"Ẅ","ẃ"=>"Ẃ","ẁ"=>"Ẁ","ṿ"=>"Ṿ","ṽ"=>"Ṽ","ṻ"=>"Ṻ","ṹ"=>"Ṹ",
 949            "ᚡ"=>"᚜","ᚾ"=>"ᚴ","ᚳ"=>"ᚲ","ᚹ"=>"ᚰ","ᚯ"=>"ᚎ","ᚭ"=>"ᚏ","ᚍ"=>"ᚪ","ᚊ"=>"ᚨ","ᚧ"=>"ᚌ","ᚼ"=>"ᚤ",
 950            "ṣ"=>"Ṣ","ṡ"=>"Ṡ","ṟ"=>"Ṟ","ṝ"=>"Ṝ","ṛ"=>"Ṛ","ṙ"=>"Ṙ","ṗ"=>"Ṗ","ṕ"=>"Ṕ","ṓ"=>"Ṓ","ṑ"=>"Ṑ",
 951            "ṏ"=>"Ṏ","ṍ"=>"Ṍ","ṋ"=>"Ṋ","ṉ"=>"Ṉ","ṇ"=>"Ṇ","ṅ"=>"Ṅ","ṃ"=>"Ṃ","ṁ"=>"Ṁ","ḿ"=>"Ḿ","ḽ"=>"Ḽ",
 952            "ḝ"=>"Ḻ","Ḛ"=>"Ḹ","ḡ"=>"Ḝ","Ḿ"=>"Ḵ","ḳ"=>"Ḳ","ḹ"=>"Ḱ","ḯ"=>"Ḏ","ḭ"=>"ḏ","ḍ"=>"Ḫ","Ḋ"=>"Ḩ",
 953            "ḧ"=>"Ḧ","ḥ"=>"Ḥ","ḣ"=>"Ḣ","ḡ"=>"Ḡ","ḟ"=>"Ḟ","ḝ"=>"Ḝ","ḛ"=>"Ḛ","ḙ"=>"Ḙ","ḗ"=>"Ḗ","ḕ"=>"Ḕ",
 954            "ḓ"=>"Ḓ","ḑ"=>"Ḑ","ḏ"=>"Ḏ","ḍ"=>"Ḍ","ḋ"=>"Ḋ","ḉ"=>"Ḉ","ḇ"=>"Ḇ","ḅ"=>"Ḅ","ḃ"=>"Ḃ","ḁ"=>"Ḁ",
 955            "ֆ"=>"Ֆ","օ"=>"Օ","ք"=>"Ք","փ"=>"Փ","ւ"=>"Ւ","ց"=>"Ց","ր"=>"Ր","տ"=>"Տ","վ"=>"Վ","ս"=>"Ս",
 956            "ռ"=>"Ռ","ջ"=>"Ջ","պ"=>"Պ","չ"=>"Չ","ո"=>"Ո","շ"=>"Շ","ն"=>"Ն","յ"=>"Յ","մ"=>"Մ","ճ"=>"Ճ",
 957            "ղ"=>"Ղ","ձ"=>"Ձ","հ"=>"Հ","կ"=>"Կ","ծ"=>"Ծ","խ"=>"Խ","լ"=>"Լ","ի"=>"Ի","ժ"=>"Ժ","թ"=>"Թ",
 958            "ը"=>"Ը","է"=>"Է","զ"=>"Զ","ե"=>"Ե","դ"=>"Դ","գ"=>"Գ","բ"=>"Բ","ա"=>"Ա","ԏ"=>"Ԏ","ԍ"=>"Ԍ",
 959            "ԋ"=>"Ԋ","ԉ"=>"Ԉ","ԇ"=>"Ԇ","ԅ"=>"Ԅ","ԃ"=>"Ԃ","ԁ"=>"Ԁ","ӹ"=>"Ӹ","ӵ"=>"Ӵ","ӳ"=>"Ӳ","ӱ"=>"Ӱ",
 960            "ӯ"=>"Ӯ","ӭ"=>"Ӭ","ӫ"=>"Ӫ","ө"=>"Ө","ӧ"=>"Ӧ","ӥ"=>"Ӥ","ӣ"=>"Ӣ","ӡ"=>"Ӡ","ӟ"=>"Ӟ","ӝ"=>"Ӝ",
 961            "ӛ"=>"Ӛ","ә"=>"Ә","ӗ"=>"Ӗ","ӕ"=>"Ӕ","ӓ"=>"Ӓ","ӑ"=>"Ӑ","ӎ"=>"Ӎ","ӌ"=>"Ӌ","ӊ"=>"Ӊ","ӈ"=>"Ӈ",
 962            "ӆ"=>"Ӆ","ӄ"=>"Ӄ","ӂ"=>"Ӂ","ҿ"=>"Ҿ","ҽ"=>"Ҽ","һ"=>"Һ","ҹ"=>"Ҹ","ҷ"=>"Ҷ","ҵ"=>"Ҵ","ҳ"=>"Ҳ",
 963            "ұ"=>"Ұ","ү"=>"Ү","ҭ"=>"Ҭ","ҫ"=>"Ҫ","ҩ"=>"Ҩ","ҧ"=>"Ҧ","ҥ"=>"Ҥ","ң"=>"Ң","ҡ"=>"Ҡ","ҟ"=>"Ҟ",
 964            "ҝ"=>"Ҝ","қ"=>"Қ","ҙ"=>"Ҙ","җ"=>"Җ","ҕ"=>"Ҕ","ғ"=>"Ғ","ґ"=>"Ґ","ҏ"=>"Ҏ","ҍ"=>"Ҍ","ҋ"=>"Ҋ",
 965            "ҁ"=>"Ҁ","ѿ"=>"Ѿ","ѽ"=>"Ѽ","ѻ"=>"Ѻ","ѹ"=>"Ѹ","ѷ"=>"Ѷ","ѵ"=>"Ѵ","ѳ"=>"Ѳ","ѱ"=>"Ѱ","ѯ"=>"Ѯ",
 966            "ѭ"=>"Ѭ","ѫ"=>"Ѫ","ѩ"=>"Ѩ","ѧ"=>"Ѧ","ѥ"=>"Ѥ","ѣ"=>"Ѣ","ѡ"=>"Ѡ","џ"=>"Џ","ў"=>"Ў","ѝ"=>"Ѝ",
 967            "ќ"=>"Ќ","ћ"=>"Ћ","њ"=>"Њ","љ"=>"Љ","ј"=>"Ј","ї"=>"Ї","і"=>"І","ѕ"=>"Ѕ","є"=>"Є","ѓ"=>"Ѓ",
 968            "ђ"=>"Ђ","ё"=>"Ё","ѐ"=>"Ѐ","я"=>"Я","ю"=>"Ю","э"=>"Э","ь"=>"Ь","ы"=>"Ы","ъ"=>"Ъ","щ"=>"Щ",
 969            "ш"=>"Ш","ч"=>"Ч","ц"=>"Ц","х"=>"Х","ф"=>"Ф","у"=>"У","т"=>"Т","с"=>"С","р"=>"Р","п"=>"П",
 970            "о"=>"О","н"=>"Н","м"=>"М","л"=>"Л","к"=>"К","й"=>"Й","и"=>"И","з"=>"З","ж"=>"Ж","е"=>"Е",
 971            "д"=>"Д","г"=>"Г","в"=>"В","б"=>"Б","а"=>"А","ϵ"=>"Ε","ϲ"=>"Σ","ϱ"=>"Ρ","ϰ"=>"Κ","ϯ"=>"Ϯ",
 972            "ϭ"=>"Ϭ","ϫ"=>"Ϫ","ϩ"=>"Ϩ","ϧ"=>"Ϧ","ϥ"=>"Ϥ","ϣ"=>"Ϣ","ϡ"=>"Ϡ","ϟ"=>"Ϟ","ϝ"=>"Ϝ","ϛ"=>"Ϛ",
 973            "ϙ"=>"Ϙ","ϖ"=>"Π","ϕ"=>"Φ","ϑ"=>"Θ","ϐ"=>"Β","ώ"=>"Ώ","ύ"=>"Ύ","ό"=>"Ό","ϋ"=>"Ϋ","ϊ"=>"Ϊ",
 974            "ω"=>"Ω","ψ"=>"Ψ","χ"=>"Χ","φ"=>"Φ","υ"=>"Υ","τ"=>"Τ","σ"=>"Σ","ς"=>"Σ","ρ"=>"Ρ","π"=>"Π",
 975            "ο"=>"Ο","ξ"=>"Ξ","ν"=>"Ν","μ"=>"Μ","λ"=>"Λ","κ"=>"Κ","ι"=>"Ι","θ"=>"Θ","η"=>"Η","ζ"=>"Ζ",
 976            "ε"=>"Ε","δ"=>"Δ","γ"=>"Γ","β"=>"Β","α"=>"Α","ί"=>"Ί","ή"=>"Ή","έ"=>"Έ","ά"=>"Ά","ʒ"=>"Ʒ",
 977            "ʋ"=>"Ʋ","ʊ"=>"Ʊ","ʈ"=>"Ʈ","ʃ"=>"Ʃ","ʀ"=>"Ʀ","ɵ"=>"Ɵ","ɲ"=>"Ɲ","ɯ"=>"Ɯ","ɩ"=>"Ɩ","ɨ"=>"Ɨ",
 978            "ɣ"=>"Ɣ","ɛ"=>"Ɛ","ə"=>"Ə","ɗ"=>"Ɗ","ɖ"=>"Ɖ","ɔ"=>"Ɔ","ɓ"=>"Ɓ","ȳ"=>"Ȳ","ȱ"=>"Ȱ","ȯ"=>"Ȯ",
 979            "ȭ"=>"Ȭ","ȫ"=>"Ȫ","ȩ"=>"Ȩ","ȧ"=>"Ȧ","ȥ"=>"Ȥ","ȣ"=>"Ȣ","ȟ"=>"Ȟ","ȝ"=>"Ȝ","ț"=>"Ț","ș"=>"Ș",
 980            "ȗ"=>"Ȗ","ȕ"=>"Ȕ","ȓ"=>"Ȓ","ȑ"=>"Ȑ","ȏ"=>"Ȏ","ȍ"=>"Ȍ","ȋ"=>"Ȋ","ȉ"=>"Ȉ","ȇ"=>"Ȇ","ȅ"=>"Ȅ",
 981            "ȃ"=>"Ȃ","ȁ"=>"Ȁ","ǿ"=>"Ǿ","ǽ"=>"Ǽ","ǻ"=>"Ǻ","ǹ"=>"Ǹ","ǵ"=>"Ǵ","dz"=>"Dz","ǯ"=>"Ǯ","ǭ"=>"Ǭ",
 982            "ǫ"=>"Ǫ","ǩ"=>"Ǩ","ǧ"=>"Ǧ","ǥ"=>"Ǥ","ǣ"=>"Ǣ","ǡ"=>"Ǡ","ǟ"=>"Ǟ","ǝ"=>"Ǝ","ǜ"=>"Ǜ","ǚ"=>"Ǚ",
 983            "ǘ"=>"Ǘ","ǖ"=>"Ǖ","ǔ"=>"Ǔ","ǒ"=>"Ǒ","ǐ"=>"Ǐ","ǎ"=>"Ǎ","nj"=>"Nj","lj"=>"Lj","dž"=>"Dž","ƿ"=>"Ƿ",
 984            "ƽ"=>"Ɵ","ƚ"=>"Ƹ","Ɯ"=>"ƾ","ƴ"=>"Ƴ","ư"=>"Ư","ƭ"=>"Ə","ƨ"=>"Ƨ","Ƽ"=>"Ƥ","ƣ"=>"Ƣ","ƥ"=>"Ơ",
 985            "ƞ"=>"Ƞ","ƙ"=>"Ƙ","ƕ"=>"Ƕ","ƒ"=>"Ƒ","ƌ"=>"Ƌ","ƈ"=>"Ƈ","ƅ"=>"Ƅ","ƃ"=>"Ƃ","ſ"=>"S","ž"=>"Ž",
 986            "ş"=>"ŝ","ź"=>"Ś","š"=>"Ŝ","ž"=>"Ŵ","ų"=>"Ų","Ź"=>"Ű","ů"=>"Ŏ","ŭ"=>"ŏ","ō"=>"Ū","Ŋ"=>"Ũ",
 987            "ŧ"=>"Ŧ","ť"=>"Ť","ţ"=>"Ţ","š"=>"Š","ş"=>"Ş","ŝ"=>"Ŝ","ś"=>"Ś","ř"=>"Ř","ŗ"=>"Ŗ","ŕ"=>"Ŕ",
 988            "œ"=>"Œ","ő"=>"Ő","ŏ"=>"Ŏ","ō"=>"Ō","ŋ"=>"Ŋ","ň"=>"Ň","ņ"=>"Ņ","ń"=>"Ń","ł"=>"Ł","ŀ"=>"Ŀ",
 989            "Ğ"=>"Ľ","ğ"=>"ĝ","ĺ"=>"Ě","ġ"=>"Ĝ","ľ"=>"Ĵ","ij"=>"IJ","Ĺ"=>"I","į"=>"Ď","ĭ"=>"ď","č"=>"Ī",
 990            "ĩ"=>"Ĩ","ħ"=>"Ħ","ĥ"=>"Ĥ","ģ"=>"Ģ","ġ"=>"Ġ","ğ"=>"Ğ","ĝ"=>"Ĝ","ě"=>"Ě","ę"=>"Ę","ė"=>"Ė",
 991            "ĕ"=>"Ĕ","ē"=>"Ē","đ"=>"Đ","ď"=>"Ď","č"=>"Č","ċ"=>"Ċ","ĉ"=>"Ĉ","ć"=>"Ć","ą"=>"Ą","ă"=>"Ă",
 992            "ā"=>"Ā","ÿ"=>"Ÿ","þ"=>"Þ","ý"=>"Ý","ü"=>"Ü","û"=>"Û","ú"=>"Ú","ù"=>"Ù","ø"=>"Ø","ö"=>"Ö",
 993            "õ"=>"Õ","ô"=>"Ô","ó"=>"Ó","ò"=>"Ò","ñ"=>"Ñ","ð"=>"Ð","ï"=>"Ï","î"=>"Î","í"=>"Í","ì"=>"Ì",
 994            "ë"=>"Ë","ê"=>"Ê","é"=>"É","è"=>"È","ç"=>"Ç","æ"=>"Æ","å"=>"Å","ä"=>"Ä","ã"=>"Ã","â"=>"Â",
 995            "á"=>"Á","à"=>"À","µ"=>"Μ","z"=>"Z","y"=>"Y","x"=>"X","w"=>"W","v"=>"V","u"=>"U","t"=>"T",
 996            "s"=>"S","r"=>"R","q"=>"Q","p"=>"P","o"=>"O","n"=>"N","m"=>"M","l"=>"L","k"=>"K","j"=>"J",
 997            "i"=>"I","h"=>"H","g"=>"G","f"=>"F","e"=>"E","d"=>"D","c"=>"C","b"=>"B","a"=>"A"
 998                );
 999
1000    /**
1001     * UTF-8 Case lookup table
1002     *
1003     * This lookuptable defines the lower case letters to their correspponding
1004     * upper case letter in UTF-8
1005     *
1006     * @author Andreas Gohr <andi@splitbrain.org>
1007     */
1008    global $UTF8_UPPER_TO_LOWER;
1009    if(empty($UTF8_UPPER_TO_LOWER)) $UTF8_UPPER_TO_LOWER = array (
1010            "Z"=>"z","Y"=>"y","X"=>"x","W"=>"w","V"=>"v","U"=>"u","T"=>"t","S"=>"s","R"=>"r","Q"=>"q",
1011            "P"=>"p","O"=>"o","N"=>"n","M"=>"m","L"=>"l","K"=>"k","J"=>"j","I"=>"i","H"=>"h","G"=>"g",
1012            "F"=>"f","E"=>"e","D"=>"d","C"=>"c","B"=>"b","A"=>"a","ῼ"=>"ῳ","Ῥ"=>"ῥ","Ῡ"=>"ῡ","Ῑ"=>"ῑ",
1013            "Ῐ"=>"ῐ","ῌ"=>"ῃ","Ι"=>"ι","ᾼ"=>"ᾳ","Ᾱ"=>"ᾱ","Ᾰ"=>"ᾰ","ᾯ"=>"ᾧ","ᾮ"=>"ᾦ","ᾭ"=>"ᾥ","ᾬ"=>"ᾤ",
1014            "ᾫ"=>"ᾣ","ᾪ"=>"ᾢ","ᾩ"=>"ᾡ","ᾟ"=>"ᾗ","ᾞ"=>"ᾖ","ᾝ"=>"ᾕ","ᾜ"=>"ᾔ","ᾛ"=>"ᾓ","ᾚ"=>"ᾒ","ᾙ"=>"ᾑ",
1015            "ᾘ"=>"ᾐ","ᾏ"=>"ᾇ","ᾎ"=>"ᾆ","ᾍ"=>"ᾅ","ᾌ"=>"ᾄ","ᾋ"=>"ᾃ","ᾊ"=>"ᾂ","ᾉ"=>"ᾁ","ᾈ"=>"ᾀ","Ώ"=>"ώ",
1016            "Ὼ"=>"ὼ","Ύ"=>"ύ","Ὺ"=>"ὺ","Ό"=>"ό","Ὸ"=>"ὸ","Ί"=>"ί","Ὶ"=>"ὶ","Ή"=>"ή","Ὴ"=>"ὴ","Έ"=>"έ",
1017            "Ὲ"=>"ὲ","ឝ"=>"ό","ឺ"=>"ὰ","Ὧ"=>"ὧ","὎"=>"Ὄ","Ὥ"=>"ὼ","὏"=>"ὤ","Ὅ"=>"ὣ","Ὢ"=>"ὢ","Ὂ"=>"ὥ",
1018            "Ὗ"=>"ὗ","Ὕ"=>"ὕ","Ὓ"=>"ὓ","Ὑ"=>"ὑ","Ὅ"=>"ὅ","Ὄ"=>"ὄ","Ὃ"=>"ὃ","Ὂ"=>"ὂ","Ὁ"=>"ὁ","Ὀ"=>"ὀ",
1019            "៿"=>"១","៞"=>"ៜ","៽"=>"៾","៟"=>"៴","៝"=>"៳","៺"=>"៲","៚"=>"៹","៸"=>"៰","៯"=>"៧","៎"=>"៌",
1020            "Ἥ"=>"ἥ","Ἤ"=>"ἤ","Ἣ"=>"ἣ","Ἢ"=>"ἢ","Ἡ"=>"ἡ","Ἕ"=>"ἕ","Ἔ"=>"ἔ","Ἓ"=>"ἓ","Ἒ"=>"ἒ","Ἑ"=>"ἑ",
1021            "Ἐ"=>"ἐ","Ἇ"=>"ἇ","Ἆ"=>"ἆ","Ἅ"=>"ἅ","Ἄ"=>"ἄ","Ἃ"=>"ἃ","Ἂ"=>"ἂ","Ἁ"=>"ἁ","Ἀ"=>"ἀ","Ỹ"=>"ỹ",
1022            "᝜"=>"ᝡ","᝴"=>"᝾","ᝲ"=>"ᝳ","ᝰ"=>"᝹","ᝎ"=>"ᝯ","ᝏ"=>"᝭","ᝪ"=>"ᝍ","ᝨ"=>"ᝊ","ᝌ"=>"ᝧ","ᝤ"=>"᝼",
1023            "Ợ"=>"ợ","Ỡ"=>"ỡ","Ở"=>"ở","Ờ"=>"ờ","Ớ"=>"ớ","Ộ"=>"ộ","Ỗ"=>"ỗ","Ổ"=>"ổ","Ồ"=>"ồ","Ố"=>"ố",
1024            "Ỏ"=>"ỏ","Ọ"=>"ọ","Ị"=>"ị","Ỉ"=>"ỉ","Ệ"=>"ệ","Ễ"=>"ễ","Ể"=>"ể","Ề"=>"ề","Ế"=>"ế","Ẽ"=>"ẽ",
1025            "Ẻ"=>"ẝ","Ẹ"=>"ẚ","ẜ"=>"ạ","Ẵ"=>"Ế","Ẳ"=>"ẳ","Ằ"=>"ẹ","Ẏ"=>"ắ","ẏ"=>"ậ","Ẫ"=>"ẍ","Ẩ"=>"Ẋ",
1026            "Ầ"=>"ầ","Ấ"=>"ấ","Ả"=>"ả","Ạ"=>"ạ","Ṡ"=>"ẛ","Ẕ"=>"ẕ","Ẓ"=>"ẓ","Ẑ"=>"ẑ","Ẏ"=>"ẏ","Ẍ"=>"ẍ",
1027            "Ẋ"=>"ẋ","Ẉ"=>"ẉ","Ẇ"=>"ẇ","Ẅ"=>"ẅ","Ẃ"=>"ẃ","Ẁ"=>"ẁ","Ṿ"=>"ṿ","Ṽ"=>"ṽ","Ṻ"=>"ṻ","Ṹ"=>"ṹ",
1028            "᚜"=>"ᚡ","ᚴ"=>"ᚾ","ᚲ"=>"ᚳ","ᚰ"=>"ᚹ","ᚎ"=>"ᚯ","ᚏ"=>"ᚭ","ᚪ"=>"ᚍ","ᚨ"=>"ᚊ","ᚌ"=>"ᚧ","ᚤ"=>"ᚼ",
1029            "Ṣ"=>"ṣ","Ṡ"=>"ṡ","Ṟ"=>"ṟ","Ṝ"=>"ṝ","Ṛ"=>"ṛ","Ṙ"=>"ṙ","Ṗ"=>"ṗ","Ṕ"=>"ṕ","Ṓ"=>"ṓ","Ṑ"=>"ṑ",
1030            "Ṏ"=>"ṏ","Ṍ"=>"ṍ","Ṋ"=>"ṋ","Ṉ"=>"ṉ","Ṇ"=>"ṇ","Ṅ"=>"ṅ","Ṃ"=>"ṃ","Ṁ"=>"ṁ","Ḿ"=>"ḿ","Ḽ"=>"ḽ",
1031            "Ḻ"=>"ḝ","Ḹ"=>"Ḛ","Ḝ"=>"ḡ","Ḵ"=>"Ḿ","Ḳ"=>"ḳ","Ḱ"=>"ḹ","Ḏ"=>"ḯ","ḏ"=>"ḭ","Ḫ"=>"ḍ","Ḩ"=>"Ḋ",
1032            "Ḧ"=>"ḧ","Ḥ"=>"ḥ","Ḣ"=>"ḣ","Ḡ"=>"ḡ","Ḟ"=>"ḟ","Ḝ"=>"ḝ","Ḛ"=>"ḛ","Ḙ"=>"ḙ","Ḗ"=>"ḗ","Ḕ"=>"ḕ",
1033            "Ḓ"=>"ḓ","Ḑ"=>"ḑ","Ḏ"=>"ḏ","Ḍ"=>"ḍ","Ḋ"=>"ḋ","Ḉ"=>"ḉ","Ḇ"=>"ḇ","Ḅ"=>"ḅ","Ḃ"=>"ḃ","Ḁ"=>"ḁ",
1034            "Ֆ"=>"ֆ","Օ"=>"օ","Ք"=>"ք","Փ"=>"փ","Ւ"=>"ւ","Ց"=>"ց","Ր"=>"ր","Տ"=>"տ","Վ"=>"վ","Ս"=>"ս",
1035            "Ռ"=>"ռ","Ջ"=>"ջ","Պ"=>"պ","Չ"=>"չ","Ո"=>"ո","Շ"=>"շ","Ն"=>"ն","Յ"=>"յ","Մ"=>"մ","Ճ"=>"ճ",
1036            "Ղ"=>"ղ","Ձ"=>"ձ","Հ"=>"հ","Կ"=>"կ","Ծ"=>"ծ","Խ"=>"խ","Լ"=>"լ","Ի"=>"ի","Ժ"=>"ժ","Թ"=>"թ",
1037            "Ը"=>"ը","Է"=>"է","Զ"=>"զ","Ե"=>"ե","Դ"=>"դ","Գ"=>"գ","Բ"=>"բ","Ա"=>"ա","Ԏ"=>"ԏ","Ԍ"=>"ԍ",
1038            "Ԋ"=>"ԋ","Ԉ"=>"ԉ","Ԇ"=>"ԇ","Ԅ"=>"ԅ","Ԃ"=>"ԃ","Ԁ"=>"ԁ","Ӹ"=>"ӹ","Ӵ"=>"ӵ","Ӳ"=>"ӳ","Ӱ"=>"ӱ",
1039            "Ӯ"=>"ӯ","Ӭ"=>"ӭ","Ӫ"=>"ӫ","Ө"=>"ө","Ӧ"=>"ӧ","Ӥ"=>"ӥ","Ӣ"=>"ӣ","Ӡ"=>"ӡ","Ӟ"=>"ӟ","Ӝ"=>"ӝ",
1040            "Ӛ"=>"ӛ","Ә"=>"ә","Ӗ"=>"ӗ","Ӕ"=>"ӕ","Ӓ"=>"ӓ","Ӑ"=>"ӑ","Ӎ"=>"ӎ","Ӌ"=>"ӌ","Ӊ"=>"ӊ","Ӈ"=>"ӈ",
1041            "Ӆ"=>"ӆ","Ӄ"=>"ӄ","Ӂ"=>"ӂ","Ҿ"=>"ҿ","Ҽ"=>"ҽ","Һ"=>"һ","Ҹ"=>"ҹ","Ҷ"=>"ҷ","Ҵ"=>"ҵ","Ҳ"=>"ҳ",
1042            "Ұ"=>"ұ","Ү"=>"ү","Ҭ"=>"ҭ","Ҫ"=>"ҫ","Ҩ"=>"ҩ","Ҧ"=>"ҧ","Ҥ"=>"ҥ","Ң"=>"ң","Ҡ"=>"ҡ","Ҟ"=>"ҟ",
1043            "Ҝ"=>"ҝ","Қ"=>"қ","Ҙ"=>"ҙ","Җ"=>"җ","Ҕ"=>"ҕ","Ғ"=>"ғ","Ґ"=>"ґ","Ҏ"=>"ҏ","Ҍ"=>"ҍ","Ҋ"=>"ҋ",
1044            "Ҁ"=>"ҁ","Ѿ"=>"

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