PageRenderTime 5ms CodeModel.GetById 72ms app.highlight 28ms RepoModel.GetById 1ms app.codeStats 1ms

/application/libraries/3rd_party/phputf8/native/core.php

http://github.com/tcm-project/tangocms
PHP | 437 lines | 221 code | 68 blank | 148 comment | 40 complexity | 7ce187155505db682820a4b8e44f5b25 MD5 | raw file
  1<?php
  2/**
  3* @package utf8
  4* @subpackage strings
  5*/
  6
  7/**
  8* Define UTF8_CORE as required
  9*/
 10if ( !defined('UTF8_CORE') ) {
 11    define('UTF8_CORE',TRUE);
 12}
 13
 14//--------------------------------------------------------------------
 15/**
 16* Unicode aware replacement for strlen(). Returns the number
 17* of characters in the string (not the number of bytes), replacing
 18* multibyte characters with a single byte equivalent
 19* utf8_decode() converts characters that are not in ISO-8859-1
 20* to '?', which, for the purpose of counting, is alright - It's
 21* much faster than iconv_strlen
 22* Note: this function does not count bad UTF-8 bytes in the string
 23* - these are simply ignored
 24* @author <chernyshevsky at hotmail dot com>
 25* @link   http://www.php.net/manual/en/function.strlen.php
 26* @link   http://www.php.net/manual/en/function.utf8-decode.php
 27* @param string UTF-8 string
 28* @return int number of UTF-8 characters in string
 29* @package utf8
 30* @subpackage strings
 31*/
 32function utf8_strlen($str){
 33    return strlen(utf8_decode($str));
 34}
 35
 36
 37//--------------------------------------------------------------------
 38/**
 39* UTF-8 aware alternative to strpos
 40* Find position of first occurrence of a string
 41* Note: This will get alot slower if offset is used
 42* Note: requires utf8_strlen amd utf8_substr to be loaded
 43* @param string haystack
 44* @param string needle (you should validate this with utf8_is_valid)
 45* @param int offset in characters (from left)
 46* @return mixed integer position or FALSE on failure
 47* @see http://www.php.net/strpos
 48* @see utf8_strlen
 49* @see utf8_substr
 50* @package utf8
 51* @subpackage strings
 52*/
 53function utf8_strpos($str, $needle, $offset = NULL) {
 54
 55    if ( is_null($offset) ) {
 56
 57        $ar = explode($needle, $str, 2);
 58        if ( count($ar) > 1 ) {
 59            return utf8_strlen($ar[0]);
 60        }
 61        return FALSE;
 62
 63    } else {
 64
 65        if ( !is_int($offset) ) {
 66            trigger_error('utf8_strpos: Offset must be an integer',E_USER_ERROR);
 67            return FALSE;
 68        }
 69
 70        $str = utf8_substr($str, $offset);
 71
 72        if ( FALSE !== ( $pos = utf8_strpos($str, $needle) ) ) {
 73            return $pos + $offset;
 74        }
 75
 76        return FALSE;
 77    }
 78
 79}
 80
 81//--------------------------------------------------------------------
 82/**
 83* UTF-8 aware alternative to strrpos
 84* Find position of last occurrence of a char in a string
 85* Note: This will get alot slower if offset is used
 86* Note: requires utf8_substr and utf8_strlen to be loaded
 87* @param string haystack
 88* @param string needle (you should validate this with utf8_is_valid)
 89* @param int (optional) offset (from left)
 90* @return mixed integer position or FALSE on failure
 91* @see http://www.php.net/strrpos
 92* @see utf8_substr
 93* @see utf8_strlen
 94* @package utf8
 95* @subpackage strings
 96*/
 97function utf8_strrpos($str, $needle, $offset = NULL) {
 98
 99    if ( is_null($offset) ) {
100
101        $ar = explode($needle, $str);
102
103        if ( count($ar) > 1 ) {
104            // Pop off the end of the string where the last match was made
105            array_pop($ar);
106            $str = join($needle,$ar);
107            return utf8_strlen($str);
108        }
109        return FALSE;
110
111    } else {
112
113        if ( !is_int($offset) ) {
114            trigger_error('utf8_strrpos expects parameter 3 to be long',E_USER_WARNING);
115            return FALSE;
116        }
117
118        $str = utf8_substr($str, $offset);
119
120        if ( FALSE !== ( $pos = utf8_strrpos($str, $needle) ) ) {
121            return $pos + $offset;
122        }
123
124        return FALSE;
125    }
126
127}
128
129//--------------------------------------------------------------------
130/**
131* UTF-8 aware alternative to substr
132* Return part of a string given character offset (and optionally length)
133*
134* Note arguments: comparied to substr - if offset or length are
135* not integers, this version will not complain but rather massages them
136* into an integer.
137*
138* Note on returned values: substr documentation states false can be
139* returned in some cases (e.g. offset > string length)
140* mb_substr never returns false, it will return an empty string instead.
141* This adopts the mb_substr approach
142*
143* Note on implementation: PCRE only supports repetitions of less than
144* 65536, in order to accept up to MAXINT values for offset and length,
145* we'll repeat a group of 65535 characters when needed.
146*
147* Note on implementation: calculating the number of characters in the
148* string is a relatively expensive operation, so we only carry it out when
149* necessary. It isn't necessary for +ve offsets and no specified length
150*
151* @author Chris Smith<chris@jalakai.co.uk>
152* @param string
153* @param int number of UTF-8 characters offset (from left)
154* @param int (optional) length in UTF-8 characters from offset
155* @return mixed string or FALSE if failure
156* @package utf8
157* @subpackage strings
158*/
159function utf8_substr($str, $offset, $length = NULL) {
160
161    // generates E_NOTICE
162    // for PHP4 objects, but not PHP5 objects
163    $str = (string)$str;
164    $offset = (int)$offset;
165    if (!is_null($length)) $length = (int)$length;
166
167    // handle trivial cases
168    if ($length === 0) return '';
169    if ($offset < 0 && $length < 0 && $length < $offset)
170        return '';
171
172    // normalise negative offsets (we could use a tail
173    // anchored pattern, but they are horribly slow!)
174    if ($offset < 0) {
175
176        // see notes
177        $strlen = strlen(utf8_decode($str));
178        $offset = $strlen + $offset;
179        if ($offset < 0) $offset = 0;
180
181    }
182
183    $Op = '';
184    $Lp = '';
185
186    // establish a pattern for offset, a
187    // non-captured group equal in length to offset
188    if ($offset > 0) {
189
190        $Ox = (int)($offset/65535);
191        $Oy = $offset%65535;
192
193        if ($Ox) {
194            $Op = '(?:.{65535}){'.$Ox.'}';
195        }
196
197        $Op = '^(?:'.$Op.'.{'.$Oy.'})';
198
199    } else {
200
201        // offset == 0; just anchor the pattern
202        $Op = '^';
203
204    }
205
206    // establish a pattern for length
207    if (is_null($length)) {
208
209        // the rest of the string
210        $Lp = '(.*)$';
211
212    } else {
213
214        if (!isset($strlen)) {
215            // see notes
216            $strlen = strlen(utf8_decode($str));
217        }
218
219        // another trivial case
220        if ($offset > $strlen) return '';
221
222        if ($length > 0) {
223
224            // reduce any length that would
225            // go passed the end of the string
226            $length = min($strlen-$offset, $length);
227
228            $Lx = (int)( $length / 65535 );
229            $Ly = $length % 65535;
230
231            // negative length requires a captured group
232            // of length characters
233            if ($Lx) $Lp = '(?:.{65535}){'.$Lx.'}';
234            $Lp = '('.$Lp.'.{'.$Ly.'})';
235
236        } else if ($length < 0) {
237
238            if ( $length < ($offset - $strlen) ) {
239                return '';
240            }
241
242            $Lx = (int)((-$length)/65535);
243            $Ly = (-$length)%65535;
244
245            // negative length requires ... capture everything
246            // except a group of  -length characters
247            // anchored at the tail-end of the string
248            if ($Lx) $Lp = '(?:.{65535}){'.$Lx.'}';
249            $Lp = '(.*)(?:'.$Lp.'.{'.$Ly.'})$';
250
251        }
252
253    }
254
255    if (!preg_match( '#'.$Op.$Lp.'#us',$str, $match )) {
256        return '';
257    }
258
259    return $match[1];
260
261}
262
263//---------------------------------------------------------------
264/**
265* UTF-8 aware alternative to strtolower
266* Make a string lowercase
267* Note: The concept of a characters "case" only exists is some alphabets
268* such as Latin, Greek, Cyrillic, Armenian and archaic Georgian - it does
269* not exist in the Chinese alphabet, for example. See Unicode Standard
270* Annex #21: Case Mappings
271* Note: requires utf8_to_unicode and utf8_from_unicode
272* @author Andreas Gohr <andi@splitbrain.org>
273* @param string
274* @return mixed either string in lowercase or FALSE is UTF-8 invalid
275* @see http://www.php.net/strtolower
276* @see utf8_to_unicode
277* @see utf8_from_unicode
278* @see http://www.unicode.org/reports/tr21/tr21-5.html
279* @see http://dev.splitbrain.org/view/darcs/dokuwiki/inc/utf8.php
280* @package utf8
281* @subpackage strings
282*/
283function utf8_strtolower($string){
284
285    static $UTF8_UPPER_TO_LOWER = NULL;
286
287    if ( is_null($UTF8_UPPER_TO_LOWER) ) {
288        $UTF8_UPPER_TO_LOWER = array(
289    0x0041=>0x0061, 0x03A6=>0x03C6, 0x0162=>0x0163, 0x00C5=>0x00E5, 0x0042=>0x0062,
290    0x0139=>0x013A, 0x00C1=>0x00E1, 0x0141=>0x0142, 0x038E=>0x03CD, 0x0100=>0x0101,
291    0x0490=>0x0491, 0x0394=>0x03B4, 0x015A=>0x015B, 0x0044=>0x0064, 0x0393=>0x03B3,
292    0x00D4=>0x00F4, 0x042A=>0x044A, 0x0419=>0x0439, 0x0112=>0x0113, 0x041C=>0x043C,
293    0x015E=>0x015F, 0x0143=>0x0144, 0x00CE=>0x00EE, 0x040E=>0x045E, 0x042F=>0x044F,
294    0x039A=>0x03BA, 0x0154=>0x0155, 0x0049=>0x0069, 0x0053=>0x0073, 0x1E1E=>0x1E1F,
295    0x0134=>0x0135, 0x0427=>0x0447, 0x03A0=>0x03C0, 0x0418=>0x0438, 0x00D3=>0x00F3,
296    0x0420=>0x0440, 0x0404=>0x0454, 0x0415=>0x0435, 0x0429=>0x0449, 0x014A=>0x014B,
297    0x0411=>0x0431, 0x0409=>0x0459, 0x1E02=>0x1E03, 0x00D6=>0x00F6, 0x00D9=>0x00F9,
298    0x004E=>0x006E, 0x0401=>0x0451, 0x03A4=>0x03C4, 0x0423=>0x0443, 0x015C=>0x015D,
299    0x0403=>0x0453, 0x03A8=>0x03C8, 0x0158=>0x0159, 0x0047=>0x0067, 0x00C4=>0x00E4,
300    0x0386=>0x03AC, 0x0389=>0x03AE, 0x0166=>0x0167, 0x039E=>0x03BE, 0x0164=>0x0165,
301    0x0116=>0x0117, 0x0108=>0x0109, 0x0056=>0x0076, 0x00DE=>0x00FE, 0x0156=>0x0157,
302    0x00DA=>0x00FA, 0x1E60=>0x1E61, 0x1E82=>0x1E83, 0x00C2=>0x00E2, 0x0118=>0x0119,
303    0x0145=>0x0146, 0x0050=>0x0070, 0x0150=>0x0151, 0x042E=>0x044E, 0x0128=>0x0129,
304    0x03A7=>0x03C7, 0x013D=>0x013E, 0x0422=>0x0442, 0x005A=>0x007A, 0x0428=>0x0448,
305    0x03A1=>0x03C1, 0x1E80=>0x1E81, 0x016C=>0x016D, 0x00D5=>0x00F5, 0x0055=>0x0075,
306    0x0176=>0x0177, 0x00DC=>0x00FC, 0x1E56=>0x1E57, 0x03A3=>0x03C3, 0x041A=>0x043A,
307    0x004D=>0x006D, 0x016A=>0x016B, 0x0170=>0x0171, 0x0424=>0x0444, 0x00CC=>0x00EC,
308    0x0168=>0x0169, 0x039F=>0x03BF, 0x004B=>0x006B, 0x00D2=>0x00F2, 0x00C0=>0x00E0,
309    0x0414=>0x0434, 0x03A9=>0x03C9, 0x1E6A=>0x1E6B, 0x00C3=>0x00E3, 0x042D=>0x044D,
310    0x0416=>0x0436, 0x01A0=>0x01A1, 0x010C=>0x010D, 0x011C=>0x011D, 0x00D0=>0x00F0,
311    0x013B=>0x013C, 0x040F=>0x045F, 0x040A=>0x045A, 0x00C8=>0x00E8, 0x03A5=>0x03C5,
312    0x0046=>0x0066, 0x00DD=>0x00FD, 0x0043=>0x0063, 0x021A=>0x021B, 0x00CA=>0x00EA,
313    0x0399=>0x03B9, 0x0179=>0x017A, 0x00CF=>0x00EF, 0x01AF=>0x01B0, 0x0045=>0x0065,
314    0x039B=>0x03BB, 0x0398=>0x03B8, 0x039C=>0x03BC, 0x040C=>0x045C, 0x041F=>0x043F,
315    0x042C=>0x044C, 0x00DE=>0x00FE, 0x00D0=>0x00F0, 0x1EF2=>0x1EF3, 0x0048=>0x0068,
316    0x00CB=>0x00EB, 0x0110=>0x0111, 0x0413=>0x0433, 0x012E=>0x012F, 0x00C6=>0x00E6,
317    0x0058=>0x0078, 0x0160=>0x0161, 0x016E=>0x016F, 0x0391=>0x03B1, 0x0407=>0x0457,
318    0x0172=>0x0173, 0x0178=>0x00FF, 0x004F=>0x006F, 0x041B=>0x043B, 0x0395=>0x03B5,
319    0x0425=>0x0445, 0x0120=>0x0121, 0x017D=>0x017E, 0x017B=>0x017C, 0x0396=>0x03B6,
320    0x0392=>0x03B2, 0x0388=>0x03AD, 0x1E84=>0x1E85, 0x0174=>0x0175, 0x0051=>0x0071,
321    0x0417=>0x0437, 0x1E0A=>0x1E0B, 0x0147=>0x0148, 0x0104=>0x0105, 0x0408=>0x0458,
322    0x014C=>0x014D, 0x00CD=>0x00ED, 0x0059=>0x0079, 0x010A=>0x010B, 0x038F=>0x03CE,
323    0x0052=>0x0072, 0x0410=>0x0430, 0x0405=>0x0455, 0x0402=>0x0452, 0x0126=>0x0127,
324    0x0136=>0x0137, 0x012A=>0x012B, 0x038A=>0x03AF, 0x042B=>0x044B, 0x004C=>0x006C,
325    0x0397=>0x03B7, 0x0124=>0x0125, 0x0218=>0x0219, 0x00DB=>0x00FB, 0x011E=>0x011F,
326    0x041E=>0x043E, 0x1E40=>0x1E41, 0x039D=>0x03BD, 0x0106=>0x0107, 0x03AB=>0x03CB,
327    0x0426=>0x0446, 0x00DE=>0x00FE, 0x00C7=>0x00E7, 0x03AA=>0x03CA, 0x0421=>0x0441,
328    0x0412=>0x0432, 0x010E=>0x010F, 0x00D8=>0x00F8, 0x0057=>0x0077, 0x011A=>0x011B,
329    0x0054=>0x0074, 0x004A=>0x006A, 0x040B=>0x045B, 0x0406=>0x0456, 0x0102=>0x0103,
330    0x039B=>0x03BB, 0x00D1=>0x00F1, 0x041D=>0x043D, 0x038C=>0x03CC, 0x00C9=>0x00E9,
331    0x00D0=>0x00F0, 0x0407=>0x0457, 0x0122=>0x0123,
332            );
333    }
334
335    $uni = utf8_to_unicode($string);
336
337    if ( !$uni ) {
338        return FALSE;
339    }
340
341    $cnt = count($uni);
342    for ($i=0; $i < $cnt; $i++){
343        if ( isset($UTF8_UPPER_TO_LOWER[$uni[$i]]) ) {
344            $uni[$i] = $UTF8_UPPER_TO_LOWER[$uni[$i]];
345        }
346    }
347
348    return utf8_from_unicode($uni);
349}
350
351//---------------------------------------------------------------
352/**
353* UTF-8 aware alternative to strtoupper
354* Make a string uppercase
355* Note: The concept of a characters "case" only exists is some alphabets
356* such as Latin, Greek, Cyrillic, Armenian and archaic Georgian - it does
357* not exist in the Chinese alphabet, for example. See Unicode Standard
358* Annex #21: Case Mappings
359* Note: requires utf8_to_unicode and utf8_from_unicode
360* @author Andreas Gohr <andi@splitbrain.org>
361* @param string
362* @return mixed either string in lowercase or FALSE is UTF-8 invalid
363* @see http://www.php.net/strtoupper
364* @see utf8_to_unicode
365* @see utf8_from_unicode
366* @see http://www.unicode.org/reports/tr21/tr21-5.html
367* @see http://dev.splitbrain.org/view/darcs/dokuwiki/inc/utf8.php
368* @package utf8
369* @subpackage strings
370*/
371function utf8_strtoupper($string){
372
373    static $UTF8_LOWER_TO_UPPER = NULL;
374
375    if ( is_null($UTF8_LOWER_TO_UPPER) ) {
376        $UTF8_LOWER_TO_UPPER = array(
377    0x0061=>0x0041, 0x03C6=>0x03A6, 0x0163=>0x0162, 0x00E5=>0x00C5, 0x0062=>0x0042,
378    0x013A=>0x0139, 0x00E1=>0x00C1, 0x0142=>0x0141, 0x03CD=>0x038E, 0x0101=>0x0100,
379    0x0491=>0x0490, 0x03B4=>0x0394, 0x015B=>0x015A, 0x0064=>0x0044, 0x03B3=>0x0393,
380    0x00F4=>0x00D4, 0x044A=>0x042A, 0x0439=>0x0419, 0x0113=>0x0112, 0x043C=>0x041C,
381    0x015F=>0x015E, 0x0144=>0x0143, 0x00EE=>0x00CE, 0x045E=>0x040E, 0x044F=>0x042F,
382    0x03BA=>0x039A, 0x0155=>0x0154, 0x0069=>0x0049, 0x0073=>0x0053, 0x1E1F=>0x1E1E,
383    0x0135=>0x0134, 0x0447=>0x0427, 0x03C0=>0x03A0, 0x0438=>0x0418, 0x00F3=>0x00D3,
384    0x0440=>0x0420, 0x0454=>0x0404, 0x0435=>0x0415, 0x0449=>0x0429, 0x014B=>0x014A,
385    0x0431=>0x0411, 0x0459=>0x0409, 0x1E03=>0x1E02, 0x00F6=>0x00D6, 0x00F9=>0x00D9,
386    0x006E=>0x004E, 0x0451=>0x0401, 0x03C4=>0x03A4, 0x0443=>0x0423, 0x015D=>0x015C,
387    0x0453=>0x0403, 0x03C8=>0x03A8, 0x0159=>0x0158, 0x0067=>0x0047, 0x00E4=>0x00C4,
388    0x03AC=>0x0386, 0x03AE=>0x0389, 0x0167=>0x0166, 0x03BE=>0x039E, 0x0165=>0x0164,
389    0x0117=>0x0116, 0x0109=>0x0108, 0x0076=>0x0056, 0x00FE=>0x00DE, 0x0157=>0x0156,
390    0x00FA=>0x00DA, 0x1E61=>0x1E60, 0x1E83=>0x1E82, 0x00E2=>0x00C2, 0x0119=>0x0118,
391    0x0146=>0x0145, 0x0070=>0x0050, 0x0151=>0x0150, 0x044E=>0x042E, 0x0129=>0x0128,
392    0x03C7=>0x03A7, 0x013E=>0x013D, 0x0442=>0x0422, 0x007A=>0x005A, 0x0448=>0x0428,
393    0x03C1=>0x03A1, 0x1E81=>0x1E80, 0x016D=>0x016C, 0x00F5=>0x00D5, 0x0075=>0x0055,
394    0x0177=>0x0176, 0x00FC=>0x00DC, 0x1E57=>0x1E56, 0x03C3=>0x03A3, 0x043A=>0x041A,
395    0x006D=>0x004D, 0x016B=>0x016A, 0x0171=>0x0170, 0x0444=>0x0424, 0x00EC=>0x00CC,
396    0x0169=>0x0168, 0x03BF=>0x039F, 0x006B=>0x004B, 0x00F2=>0x00D2, 0x00E0=>0x00C0,
397    0x0434=>0x0414, 0x03C9=>0x03A9, 0x1E6B=>0x1E6A, 0x00E3=>0x00C3, 0x044D=>0x042D,
398    0x0436=>0x0416, 0x01A1=>0x01A0, 0x010D=>0x010C, 0x011D=>0x011C, 0x00F0=>0x00D0,
399    0x013C=>0x013B, 0x045F=>0x040F, 0x045A=>0x040A, 0x00E8=>0x00C8, 0x03C5=>0x03A5,
400    0x0066=>0x0046, 0x00FD=>0x00DD, 0x0063=>0x0043, 0x021B=>0x021A, 0x00EA=>0x00CA,
401    0x03B9=>0x0399, 0x017A=>0x0179, 0x00EF=>0x00CF, 0x01B0=>0x01AF, 0x0065=>0x0045,
402    0x03BB=>0x039B, 0x03B8=>0x0398, 0x03BC=>0x039C, 0x045C=>0x040C, 0x043F=>0x041F,
403    0x044C=>0x042C, 0x00FE=>0x00DE, 0x00F0=>0x00D0, 0x1EF3=>0x1EF2, 0x0068=>0x0048,
404    0x00EB=>0x00CB, 0x0111=>0x0110, 0x0433=>0x0413, 0x012F=>0x012E, 0x00E6=>0x00C6,
405    0x0078=>0x0058, 0x0161=>0x0160, 0x016F=>0x016E, 0x03B1=>0x0391, 0x0457=>0x0407,
406    0x0173=>0x0172, 0x00FF=>0x0178, 0x006F=>0x004F, 0x043B=>0x041B, 0x03B5=>0x0395,
407    0x0445=>0x0425, 0x0121=>0x0120, 0x017E=>0x017D, 0x017C=>0x017B, 0x03B6=>0x0396,
408    0x03B2=>0x0392, 0x03AD=>0x0388, 0x1E85=>0x1E84, 0x0175=>0x0174, 0x0071=>0x0051,
409    0x0437=>0x0417, 0x1E0B=>0x1E0A, 0x0148=>0x0147, 0x0105=>0x0104, 0x0458=>0x0408,
410    0x014D=>0x014C, 0x00ED=>0x00CD, 0x0079=>0x0059, 0x010B=>0x010A, 0x03CE=>0x038F,
411    0x0072=>0x0052, 0x0430=>0x0410, 0x0455=>0x0405, 0x0452=>0x0402, 0x0127=>0x0126,
412    0x0137=>0x0136, 0x012B=>0x012A, 0x03AF=>0x038A, 0x044B=>0x042B, 0x006C=>0x004C,
413    0x03B7=>0x0397, 0x0125=>0x0124, 0x0219=>0x0218, 0x00FB=>0x00DB, 0x011F=>0x011E,
414    0x043E=>0x041E, 0x1E41=>0x1E40, 0x03BD=>0x039D, 0x0107=>0x0106, 0x03CB=>0x03AB,
415    0x0446=>0x0426, 0x00FE=>0x00DE, 0x00E7=>0x00C7, 0x03CA=>0x03AA, 0x0441=>0x0421,
416    0x0432=>0x0412, 0x010F=>0x010E, 0x00F8=>0x00D8, 0x0077=>0x0057, 0x011B=>0x011A,
417    0x0074=>0x0054, 0x006A=>0x004A, 0x045B=>0x040B, 0x0456=>0x0406, 0x0103=>0x0102,
418    0x03BB=>0x039B, 0x00F1=>0x00D1, 0x043D=>0x041D, 0x03CC=>0x038C, 0x00E9=>0x00C9,
419    0x00F0=>0x00D0, 0x0457=>0x0407, 0x0123=>0x0122,
420            );
421    }
422
423    $uni = utf8_to_unicode($string);
424
425    if ( !$uni ) {
426        return FALSE;
427    }
428
429    $cnt = count($uni);
430    for ($i=0; $i < $cnt; $i++){
431        if( isset($UTF8_LOWER_TO_UPPER[$uni[$i]]) ) {
432            $uni[$i] = $UTF8_LOWER_TO_UPPER[$uni[$i]];
433        }
434    }
435
436    return utf8_from_unicode($uni);
437}