Verzeichnisstruktur phpBB-3.0.0
- Veröffentlicht
- 12.12.2007
So funktioniert es
|
|
Auf das letzte Element klicken. Dies geht jeweils ein Schritt zurück |
Auf das Icon klicken, dies öffnet das Verzeichnis. Nochmal klicken schließt das Verzeichnis. |
|
|
(Beispiel Datei-Icons)
|
Auf das Icon klicken um den Quellcode anzuzeigen |
utf_tools.php
0001 <?php
0002 /**
0003 *
0004 * @package utf
0005 * @version $Id$
0006 * @copyright (c) 2006 phpBB Group
0007 * @license http://opensource.org/licenses/gpl-license.php GNU Public License
0008 *
0009 */
0010
0011 /**
0012 */
0013 if (!defined('IN_PHPBB'))
0014 {
0015 exit;
0016 }
0017
0018 // Enforce ASCII only string handling
0019 setlocale(LC_CTYPE, 'C');
0020
0021 /**
0022 * UTF-8 tools
0023 *
0024 * Whenever possible, these functions will try to use PHP's built-in functions or
0025 * extensions, otherwise they will default to custom routines.
0026 *
0027 * @package utf
0028 */
0029
0030 if (!extension_loaded('xml'))
0031 {
0032 /**
0033 * Implementation of PHP's native utf8_encode for people without XML support
0034 * This function exploits some nice things that ISO-8859-1 and UTF-8 have in common
0035 *
0036 * @param string $str ISO-8859-1 encoded data
0037 * @return string UTF-8 encoded data
0038 */
0039 function utf8_encode($str)
0040 {
0041 $out = '';
0042 for ($i = 0, $len = strlen($str); $i < $len; $i++)
0043 {
0044 $letter = $str[$i];
0045 $num = ord($letter);
0046 if ($num < 0x80)
0047 {
0048 $out .= $letter;
0049 }
0050 else if ($num < 0xC0)
0051 {
0052 $out .= "\xC2" . $letter;
0053 }
0054 else
0055 {
0056 $out .= "\xC3" . chr($num - 64);
0057 }
0058 }
0059 return $out;
0060 }
0061
0062 /**
0063 * Implementation of PHP's native utf8_decode for people without XML support
0064 *
0065 * @param string $str UTF-8 encoded data
0066 * @return string ISO-8859-1 encoded data
0067 */
0068 function utf8_decode($str)
0069 {
0070 $pos = 0;
0071 $len = strlen($str);
0072 $ret = '';
0073
0074 while ($pos < $len)
0075 {
0076 $ord = ord($str[$pos]) & 0xF0;
0077 if ($ord === 0xC0 || $ord === 0xD0)
0078 {
0079 $charval = ((ord($str[$pos]) & 0x1F) << 6) | (ord($str[$pos + 1]) & 0x3F);
0080 $pos += 2;
0081 $ret .= (($charval < 256) ? chr($charval) : '?');
0082 }
0083 else if ($ord === 0xE0)
0084 {
0085 $ret .= '?';
0086 $pos += 3;
0087 }
0088 else if ($ord === 0xF0)
0089 {
0090 $ret .= '?';
0091 $pos += 4;
0092 }
0093 else
0094 {
0095 $ret .= $str[$pos];
0096 ++$pos;
0097 }
0098 }
0099 return $ret;
0100 }
0101 }
0102
0103 // mbstring is old and has it's functions around for older versions of PHP.
0104 // if mbstring is not loaded, we go into native mode.
0105 if (extension_loaded('mbstring'))
0106 {
0107 mb_internal_encoding('UTF-8');
0108
0109 /**
0110 * UTF-8 aware alternative to strrpos
0111 * Find position of last occurrence of a char in a string
0112 *
0113 * Notes:
0114 * - offset for mb_strrpos was added in 5.2.0, we emulate if it is lower
0115 */
0116 if (version_compare(PHP_VERSION, '5.2.0', '>='))
0117 {
0118 /**
0119 * UTF-8 aware alternative to strrpos
0120 * @ignore
0121 */
0122 function utf8_strrpos($str, $needle, $offset = null)
0123 {
0124 // Emulate behaviour of strrpos rather than raising warning
0125 if (empty($str))
0126 {
0127 return false;
0128 }
0129
0130 if (is_null($offset))
0131 {
0132 return mb_strrpos($str, $needle);
0133 }
0134 else
0135 {
0136 return mb_strrpos($str, $needle, $offset);
0137 }
0138 }
0139 }
0140 else
0141 {
0142 /**
0143 * UTF-8 aware alternative to strrpos
0144 * @ignore
0145 */
0146 function utf8_strrpos($str, $needle, $offset = null)
0147 {
0148 // offset for mb_strrpos was added in 5.2.0
0149 if (is_null($offset))
0150 {
0151 // Emulate behaviour of strrpos rather than raising warning
0152 if (empty($str))
0153 {
0154 return false;
0155 }
0156
0157 return mb_strrpos($str, $needle);
0158 }
0159 else
0160 {
0161 if (!is_int($offset))
0162 {
0163 trigger_error('utf8_strrpos expects parameter 3 to be long', E_USER_ERROR);
0164 return false;
0165 }
0166
0167 $str = mb_substr($str, $offset);
0168
0169 if (false !== ($pos = mb_strrpos($str, $needle)))
0170 {
0171 return $pos + $offset;
0172 }
0173
0174 return false;
0175 }
0176 }
0177 }
0178
0179 /**
0180 * UTF-8 aware alternative to strpos
0181 * @ignore
0182 */
0183 function utf8_strpos($str, $needle, $offset = null)
0184 {
0185 if (is_null($offset))
0186 {
0187 return mb_strpos($str, $needle);
0188 }
0189 else
0190 {
0191 return mb_strpos($str, $needle, $offset);
0192 }
0193 }
0194
0195 /**
0196 * UTF-8 aware alternative to strtolower
0197 * @ignore
0198 */
0199 function utf8_strtolower($str)
0200 {
0201 return mb_strtolower($str);
0202 }
0203
0204 /**
0205 * UTF-8 aware alternative to strtoupper
0206 * @ignore
0207 */
0208 function utf8_strtoupper($str)
0209 {
0210 return mb_strtoupper($str);
0211 }
0212
0213 /**
0214 * UTF-8 aware alternative to substr
0215 * @ignore
0216 */
0217 function utf8_substr($str, $offset, $length = null)
0218 {
0219 if (is_null($length))
0220 {
0221 return mb_substr($str, $offset);
0222 }
0223 else
0224 {
0225 return mb_substr($str, $offset, $length);
0226 }
0227 }
0228
0229 /**
0230 * Return the length (in characters) of a UTF-8 string
0231 * @ignore
0232 */
0233 function utf8_strlen($text)
0234 {
0235 return mb_strlen($text, 'utf-8');
0236 }
0237 }
0238 else
0239 {
0240 /**
0241 * UTF-8 aware alternative to strrpos
0242 * Find position of last occurrence of a char in a string
0243 *
0244 * @author Harry Fuecks
0245 * @param string $str haystack
0246 * @param string $needle needle
0247 * @param integer $offset (optional) offset (from left)
0248 * @return mixed integer position or FALSE on failure
0249 */
0250 function utf8_strrpos($str, $needle, $offset = null)
0251 {
0252 if (is_null($offset))
0253 {
0254 $ar = explode($needle, $str);
0255
0256 if (sizeof($ar) > 1)
0257 {
0258 // Pop off the end of the string where the last match was made
0259 array_pop($ar);
0260 $str = join($needle, $ar);
0261
0262 return utf8_strlen($str);
0263 }
0264 return false;
0265 }
0266 else
0267 {
0268 if (!is_int($offset))
0269 {
0270 trigger_error('utf8_strrpos expects parameter 3 to be long', E_USER_ERROR);
0271 return false;
0272 }
0273
0274 $str = utf8_substr($str, $offset);
0275
0276 if (false !== ($pos = utf8_strrpos($str, $needle)))
0277 {
0278 return $pos + $offset;
0279 }
0280
0281 return false;
0282 }
0283 }
0284
0285 /**
0286 * UTF-8 aware alternative to strpos
0287 * Find position of first occurrence of a string
0288 *
0289 * @author Harry Fuecks
0290 * @param string $str haystack
0291 * @param string $needle needle
0292 * @param integer $offset offset in characters (from left)
0293 * @return mixed integer position or FALSE on failure
0294 */
0295 function utf8_strpos($str, $needle, $offset = null)
0296 {
0297 if (is_null($offset))
0298 {
0299 $ar = explode($needle, $str);
0300 if (sizeof($ar) > 1)
0301 {
0302 return utf8_strlen($ar[0]);
0303 }
0304 return false;
0305 }
0306 else
0307 {
0308 if (!is_int($offset))
0309 {
0310 trigger_error('utf8_strpos: Offset must be an integer', E_USER_ERROR);
0311 return false;
0312 }
0313
0314 $str = utf8_substr($str, $offset);
0315
0316 if (false !== ($pos = utf8_strpos($str, $needle)))
0317 {
0318 return $pos + $offset;
0319 }
0320
0321 return false;
0322 }
0323 }
0324
0325 /**
0326 * UTF-8 aware alternative to strtolower
0327 * Make a string lowercase
0328 * Note: The concept of a characters "case" only exists is some alphabets
0329 * such as Latin, Greek, Cyrillic, Armenian and archaic Georgian - it does
0330 * not exist in the Chinese alphabet, for example. See Unicode Standard
0331 * Annex #21: Case Mappings
0332 *
0333 * @param string
0334 * @return string string in lowercase
0335 */
0336 function utf8_strtolower($string)
0337 {
0338 static $utf8_upper_to_lower = array(
0339 "\xC3\x80" => "\xC3\xA0", "\xC3\x81" => "\xC3\xA1",
0340 "\xC3\x82" => "\xC3\xA2", "\xC3\x83" => "\xC3\xA3", "\xC3\x84" => "\xC3\xA4", "\xC3\x85" => "\xC3\xA5",
0341 "\xC3\x86" => "\xC3\xA6", "\xC3\x87" => "\xC3\xA7", "\xC3\x88" => "\xC3\xA8", "\xC3\x89" => "\xC3\xA9",
0342 "\xC3\x8A" => "\xC3\xAA", "\xC3\x8B" => "\xC3\xAB", "\xC3\x8C" => "\xC3\xAC", "\xC3\x8D" => "\xC3\xAD",
0343 "\xC3\x8E" => "\xC3\xAE", "\xC3\x8F" => "\xC3\xAF", "\xC3\x90" => "\xC3\xB0", "\xC3\x91" => "\xC3\xB1",
0344 "\xC3\x92" => "\xC3\xB2", "\xC3\x93" => "\xC3\xB3", "\xC3\x94" => "\xC3\xB4", "\xC3\x95" => "\xC3\xB5",
0345 "\xC3\x96" => "\xC3\xB6", "\xC3\x98" => "\xC3\xB8", "\xC3\x99" => "\xC3\xB9", "\xC3\x9A" => "\xC3\xBA",
0346 "\xC3\x9B" => "\xC3\xBB", "\xC3\x9C" => "\xC3\xBC", "\xC3\x9D" => "\xC3\xBD", "\xC3\x9E" => "\xC3\xBE",
0347 "\xC4\x80" => "\xC4\x81", "\xC4\x82" => "\xC4\x83", "\xC4\x84" => "\xC4\x85", "\xC4\x86" => "\xC4\x87",
0348 "\xC4\x88" => "\xC4\x89", "\xC4\x8A" => "\xC4\x8B", "\xC4\x8C" => "\xC4\x8D", "\xC4\x8E" => "\xC4\x8F",
0349 "\xC4\x90" => "\xC4\x91", "\xC4\x92" => "\xC4\x93", "\xC4\x96" => "\xC4\x97", "\xC4\x98" => "\xC4\x99",
0350 "\xC4\x9A" => "\xC4\x9B", "\xC4\x9C" => "\xC4\x9D", "\xC4\x9E" => "\xC4\x9F", "\xC4\xA0" => "\xC4\xA1",
0351 "\xC4\xA2" => "\xC4\xA3", "\xC4\xA4" => "\xC4\xA5", "\xC4\xA6" => "\xC4\xA7", "\xC4\xA8" => "\xC4\xA9",
0352 "\xC4\xAA" => "\xC4\xAB", "\xC4\xAE" => "\xC4\xAF", "\xC4\xB4" => "\xC4\xB5", "\xC4\xB6" => "\xC4\xB7",
0353 "\xC4\xB9" => "\xC4\xBA", "\xC4\xBB" => "\xC4\xBC", "\xC4\xBD" => "\xC4\xBE", "\xC5\x81" => "\xC5\x82",
0354 "\xC5\x83" => "\xC5\x84", "\xC5\x85" => "\xC5\x86", "\xC5\x87" => "\xC5\x88", "\xC5\x8A" => "\xC5\x8B",
0355 "\xC5\x8C" => "\xC5\x8D", "\xC5\x90" => "\xC5\x91", "\xC5\x94" => "\xC5\x95", "\xC5\x96" => "\xC5\x97",
0356 "\xC5\x98" => "\xC5\x99", "\xC5\x9A" => "\xC5\x9B", "\xC5\x9C" => "\xC5\x9D", "\xC5\x9E" => "\xC5\x9F",
0357 "\xC5\xA0" => "\xC5\xA1", "\xC5\xA2" => "\xC5\xA3", "\xC5\xA4" => "\xC5\xA5", "\xC5\xA6" => "\xC5\xA7",
0358 "\xC5\xA8" => "\xC5\xA9", "\xC5\xAA" => "\xC5\xAB", "\xC5\xAC" => "\xC5\xAD", "\xC5\xAE" => "\xC5\xAF",
0359 "\xC5\xB0" => "\xC5\xB1", "\xC5\xB2" => "\xC5\xB3", "\xC5\xB4" => "\xC5\xB5", "\xC5\xB6" => "\xC5\xB7",
0360 "\xC5\xB8" => "\xC3\xBF", "\xC5\xB9" => "\xC5\xBA", "\xC5\xBB" => "\xC5\xBC", "\xC5\xBD" => "\xC5\xBE",
0361 "\xC6\xA0" => "\xC6\xA1", "\xC6\xAF" => "\xC6\xB0", "\xC8\x98" => "\xC8\x99", "\xC8\x9A" => "\xC8\x9B",
0362 "\xCE\x86" => "\xCE\xAC", "\xCE\x88" => "\xCE\xAD", "\xCE\x89" => "\xCE\xAE", "\xCE\x8A" => "\xCE\xAF",
0363 "\xCE\x8C" => "\xCF\x8C", "\xCE\x8E" => "\xCF\x8D", "\xCE\x8F" => "\xCF\x8E", "\xCE\x91" => "\xCE\xB1",
0364 "\xCE\x92" => "\xCE\xB2", "\xCE\x93" => "\xCE\xB3", "\xCE\x94" => "\xCE\xB4", "\xCE\x95" => "\xCE\xB5",
0365 "\xCE\x96" => "\xCE\xB6", "\xCE\x97" => "\xCE\xB7", "\xCE\x98" => "\xCE\xB8", "\xCE\x99" => "\xCE\xB9",
0366 "\xCE\x9A" => "\xCE\xBA", "\xCE\x9B" => "\xCE\xBB", "\xCE\x9C" => "\xCE\xBC", "\xCE\x9D" => "\xCE\xBD",
0367 "\xCE\x9E" => "\xCE\xBE", "\xCE\x9F" => "\xCE\xBF", "\xCE\xA0" => "\xCF\x80", "\xCE\xA1" => "\xCF\x81",
0368 "\xCE\xA3" => "\xCF\x83", "\xCE\xA4" => "\xCF\x84", "\xCE\xA5" => "\xCF\x85", "\xCE\xA6" => "\xCF\x86",
0369 "\xCE\xA7" => "\xCF\x87", "\xCE\xA8" => "\xCF\x88", "\xCE\xA9" => "\xCF\x89", "\xCE\xAA" => "\xCF\x8A",
0370 "\xCE\xAB" => "\xCF\x8B", "\xD0\x81" => "\xD1\x91", "\xD0\x82" => "\xD1\x92", "\xD0\x83" => "\xD1\x93",
0371 "\xD0\x84" => "\xD1\x94", "\xD0\x85" => "\xD1\x95", "\xD0\x86" => "\xD1\x96", "\xD0\x87" => "\xD1\x97",
0372 "\xD0\x88" => "\xD1\x98", "\xD0\x89" => "\xD1\x99", "\xD0\x8A" => "\xD1\x9A", "\xD0\x8B" => "\xD1\x9B",
0373 "\xD0\x8C" => "\xD1\x9C", "\xD0\x8E" => "\xD1\x9E", "\xD0\x8F" => "\xD1\x9F", "\xD0\x90" => "\xD0\xB0",
0374 "\xD0\x91" => "\xD0\xB1", "\xD0\x92" => "\xD0\xB2", "\xD0\x93" => "\xD0\xB3", "\xD0\x94" => "\xD0\xB4",
0375 "\xD0\x95" => "\xD0\xB5", "\xD0\x96" => "\xD0\xB6", "\xD0\x97" => "\xD0\xB7", "\xD0\x98" => "\xD0\xB8",
0376 "\xD0\x99" => "\xD0\xB9", "\xD0\x9A" => "\xD0\xBA", "\xD0\x9B" => "\xD0\xBB", "\xD0\x9C" => "\xD0\xBC",
0377 "\xD0\x9D" => "\xD0\xBD", "\xD0\x9E" => "\xD0\xBE", "\xD0\x9F" => "\xD0\xBF", "\xD0\xA0" => "\xD1\x80",
0378 "\xD0\xA1" => "\xD1\x81", "\xD0\xA2" => "\xD1\x82", "\xD0\xA3" => "\xD1\x83", "\xD0\xA4" => "\xD1\x84",
0379 "\xD0\xA5" => "\xD1\x85", "\xD0\xA6" => "\xD1\x86", "\xD0\xA7" => "\xD1\x87", "\xD0\xA8" => "\xD1\x88",
0380 "\xD0\xA9" => "\xD1\x89", "\xD0\xAA" => "\xD1\x8A", "\xD0\xAB" => "\xD1\x8B", "\xD0\xAC" => "\xD1\x8C",
0381 "\xD0\xAD" => "\xD1\x8D", "\xD0\xAE" => "\xD1\x8E", "\xD0\xAF" => "\xD1\x8F", "\xD2\x90" => "\xD2\x91",
0382 "\xE1\xB8\x82" => "\xE1\xB8\x83", "\xE1\xB8\x8A" => "\xE1\xB8\x8B", "\xE1\xB8\x9E" => "\xE1\xB8\x9F", "\xE1\xB9\x80" => "\xE1\xB9\x81",
0383 "\xE1\xB9\x96" => "\xE1\xB9\x97", "\xE1\xB9\xA0" => "\xE1\xB9\xA1", "\xE1\xB9\xAA" => "\xE1\xB9\xAB", "\xE1\xBA\x80" => "\xE1\xBA\x81",
0384 "\xE1\xBA\x82" => "\xE1\xBA\x83", "\xE1\xBA\x84" => "\xE1\xBA\x85", "\xE1\xBB\xB2" => "\xE1\xBB\xB3"
0385 );
0386
0387 return strtr(strtolower($string), $utf8_upper_to_lower);
0388 }
0389
0390 /**
0391 * UTF-8 aware alternative to strtoupper
0392 * Make a string uppercase
0393 * Note: The concept of a characters "case" only exists is some alphabets
0394 * such as Latin, Greek, Cyrillic, Armenian and archaic Georgian - it does
0395 * not exist in the Chinese alphabet, for example. See Unicode Standard
0396 * Annex #21: Case Mappings
0397 *
0398 * @param string
0399 * @return string string in uppercase
0400 */
0401 function utf8_strtoupper($string)
0402 {
0403 static $utf8_lower_to_upper = array(
0404 "\xC3\xA0" => "\xC3\x80", "\xC3\xA1" => "\xC3\x81",
0405 "\xC3\xA2" => "\xC3\x82", "\xC3\xA3" => "\xC3\x83", "\xC3\xA4" => "\xC3\x84", "\xC3\xA5" => "\xC3\x85",
0406 "\xC3\xA6" => "\xC3\x86", "\xC3\xA7" => "\xC3\x87", "\xC3\xA8" => "\xC3\x88", "\xC3\xA9" => "\xC3\x89",
0407 "\xC3\xAA" => "\xC3\x8A", "\xC3\xAB" => "\xC3\x8B", "\xC3\xAC" => "\xC3\x8C", "\xC3\xAD" => "\xC3\x8D",
0408 "\xC3\xAE" => "\xC3\x8E", "\xC3\xAF" => "\xC3\x8F", "\xC3\xB0" => "\xC3\x90", "\xC3\xB1" => "\xC3\x91",
0409 "\xC3\xB2" => "\xC3\x92", "\xC3\xB3" => "\xC3\x93", "\xC3\xB4" => "\xC3\x94", "\xC3\xB5" => "\xC3\x95",
0410 "\xC3\xB6" => "\xC3\x96", "\xC3\xB8" => "\xC3\x98", "\xC3\xB9" => "\xC3\x99", "\xC3\xBA" => "\xC3\x9A",
0411 "\xC3\xBB" => "\xC3\x9B", "\xC3\xBC" => "\xC3\x9C", "\xC3\xBD" => "\xC3\x9D", "\xC3\xBE" => "\xC3\x9E",
0412 "\xC3\xBF" => "\xC5\xB8", "\xC4\x81" => "\xC4\x80", "\xC4\x83" => "\xC4\x82", "\xC4\x85" => "\xC4\x84",
0413 "\xC4\x87" => "\xC4\x86", "\xC4\x89" => "\xC4\x88", "\xC4\x8B" => "\xC4\x8A", "\xC4\x8D" => "\xC4\x8C",
0414 "\xC4\x8F" => "\xC4\x8E", "\xC4\x91" => "\xC4\x90", "\xC4\x93" => "\xC4\x92", "\xC4\x97" => "\xC4\x96",
0415 "\xC4\x99" => "\xC4\x98", "\xC4\x9B" => "\xC4\x9A", "\xC4\x9D" => "\xC4\x9C", "\xC4\x9F" => "\xC4\x9E",
0416 "\xC4\xA1" => "\xC4\xA0", "\xC4\xA3" => "\xC4\xA2", "\xC4\xA5" => "\xC4\xA4", "\xC4\xA7" => "\xC4\xA6",
0417 "\xC4\xA9" => "\xC4\xA8", "\xC4\xAB" => "\xC4\xAA", "\xC4\xAF" => "\xC4\xAE", "\xC4\xB5" => "\xC4\xB4",
0418 "\xC4\xB7" => "\xC4\xB6", "\xC4\xBA" => "\xC4\xB9", "\xC4\xBC" => "\xC4\xBB", "\xC4\xBE" => "\xC4\xBD",
0419 "\xC5\x82" => "\xC5\x81", "\xC5\x84" => "\xC5\x83", "\xC5\x86" => "\xC5\x85", "\xC5\x88" => "\xC5\x87",
0420 "\xC5\x8B" => "\xC5\x8A", "\xC5\x8D" => "\xC5\x8C", "\xC5\x91" => "\xC5\x90", "\xC5\x95" => "\xC5\x94",
0421 "\xC5\x97" => "\xC5\x96", "\xC5\x99" => "\xC5\x98", "\xC5\x9B" => "\xC5\x9A", "\xC5\x9D" => "\xC5\x9C",
0422 "\xC5\x9F" => "\xC5\x9E", "\xC5\xA1" => "\xC5\xA0", "\xC5\xA3" => "\xC5\xA2", "\xC5\xA5" => "\xC5\xA4",
0423 "\xC5\xA7" => "\xC5\xA6", "\xC5\xA9" => "\xC5\xA8", "\xC5\xAB" => "\xC5\xAA", "\xC5\xAD" => "\xC5\xAC",
0424 "\xC5\xAF" => "\xC5\xAE", "\xC5\xB1" => "\xC5\xB0", "\xC5\xB3" => "\xC5\xB2", "\xC5\xB5" => "\xC5\xB4",
0425 "\xC5\xB7" => "\xC5\xB6", "\xC5\xBA" => "\xC5\xB9", "\xC5\xBC" => "\xC5\xBB", "\xC5\xBE" => "\xC5\xBD",
0426 "\xC6\xA1" => "\xC6\xA0", "\xC6\xB0" => "\xC6\xAF", "\xC8\x99" => "\xC8\x98", "\xC8\x9B" => "\xC8\x9A",
0427 "\xCE\xAC" => "\xCE\x86", "\xCE\xAD" => "\xCE\x88", "\xCE\xAE" => "\xCE\x89", "\xCE\xAF" => "\xCE\x8A",
0428 "\xCE\xB1" => "\xCE\x91", "\xCE\xB2" => "\xCE\x92", "\xCE\xB3" => "\xCE\x93", "\xCE\xB4" => "\xCE\x94",
0429 "\xCE\xB5" => "\xCE\x95", "\xCE\xB6" => "\xCE\x96", "\xCE\xB7" => "\xCE\x97", "\xCE\xB8" => "\xCE\x98",
0430 "\xCE\xB9" => "\xCE\x99", "\xCE\xBA" => "\xCE\x9A", "\xCE\xBB" => "\xCE\x9B", "\xCE\xBC" => "\xCE\x9C",
0431 "\xCE\xBD" => "\xCE\x9D", "\xCE\xBE" => "\xCE\x9E", "\xCE\xBF" => "\xCE\x9F", "\xCF\x80" => "\xCE\xA0",
0432 "\xCF\x81" => "\xCE\xA1", "\xCF\x83" => "\xCE\xA3", "\xCF\x84" => "\xCE\xA4", "\xCF\x85" => "\xCE\xA5",
0433 "\xCF\x86" => "\xCE\xA6", "\xCF\x87" => "\xCE\xA7", "\xCF\x88" => "\xCE\xA8", "\xCF\x89" => "\xCE\xA9",
0434 "\xCF\x8A" => "\xCE\xAA", "\xCF\x8B" => "\xCE\xAB", "\xCF\x8C" => "\xCE\x8C", "\xCF\x8D" => "\xCE\x8E",
0435 "\xCF\x8E" => "\xCE\x8F", "\xD0\xB0" => "\xD0\x90", "\xD0\xB1" => "\xD0\x91", "\xD0\xB2" => "\xD0\x92",
0436 "\xD0\xB3" => "\xD0\x93", "\xD0\xB4" => "\xD0\x94", "\xD0\xB5" => "\xD0\x95", "\xD0\xB6" => "\xD0\x96",
0437 "\xD0\xB7" => "\xD0\x97", "\xD0\xB8" => "\xD0\x98", "\xD0\xB9" => "\xD0\x99", "\xD0\xBA" => "\xD0\x9A",
0438 "\xD0\xBB" => "\xD0\x9B", "\xD0\xBC" => "\xD0\x9C", "\xD0\xBD" => "\xD0\x9D", "\xD0\xBE" => "\xD0\x9E",
0439 "\xD0\xBF" => "\xD0\x9F", "\xD1\x80" => "\xD0\xA0", "\xD1\x81" => "\xD0\xA1", "\xD1\x82" => "\xD0\xA2",
0440 "\xD1\x83" => "\xD0\xA3", "\xD1\x84" => "\xD0\xA4", "\xD1\x85" => "\xD0\xA5", "\xD1\x86" => "\xD0\xA6",
0441 "\xD1\x87" => "\xD0\xA7", "\xD1\x88" => "\xD0\xA8", "\xD1\x89" => "\xD0\xA9", "\xD1\x8A" => "\xD0\xAA",
0442 "\xD1\x8B" => "\xD0\xAB", "\xD1\x8C" => "\xD0\xAC", "\xD1\x8D" => "\xD0\xAD", "\xD1\x8E" => "\xD0\xAE",
0443 "\xD1\x8F" => "\xD0\xAF", "\xD1\x91" => "\xD0\x81", "\xD1\x92" => "\xD0\x82", "\xD1\x93" => "\xD0\x83",
0444 "\xD1\x94" => "\xD0\x84", "\xD1\x95" => "\xD0\x85", "\xD1\x96" => "\xD0\x86", "\xD1\x97" => "\xD0\x87",
0445 "\xD1\x98" => "\xD0\x88", "\xD1\x99" => "\xD0\x89", "\xD1\x9A" => "\xD0\x8A", "\xD1\x9B" => "\xD0\x8B",
0446 "\xD1\x9C" => "\xD0\x8C", "\xD1\x9E" => "\xD0\x8E", "\xD1\x9F" => "\xD0\x8F", "\xD2\x91" => "\xD2\x90",
0447 "\xE1\xB8\x83" => "\xE1\xB8\x82", "\xE1\xB8\x8B" => "\xE1\xB8\x8A", "\xE1\xB8\x9F" => "\xE1\xB8\x9E", "\xE1\xB9\x81" => "\xE1\xB9\x80",
0448 "\xE1\xB9\x97" => "\xE1\xB9\x96", "\xE1\xB9\xA1" => "\xE1\xB9\xA0", "\xE1\xB9\xAB" => "\xE1\xB9\xAA", "\xE1\xBA\x81" => "\xE1\xBA\x80",
0449 "\xE1\xBA\x83" => "\xE1\xBA\x82", "\xE1\xBA\x85" => "\xE1\xBA\x84", "\xE1\xBB\xB3" => "\xE1\xBB\xB2"
0450 );
0451
0452 return strtr(strtoupper($string), $utf8_lower_to_upper);
0453 }
0454
0455 /**
0456 * UTF-8 aware alternative to substr
0457 * Return part of a string given character offset (and optionally length)
0458 *
0459 * Note arguments: comparied to substr - if offset or length are
0460 * not integers, this version will not complain but rather massages them
0461 * into an integer.
0462 *
0463 * Note on returned values: substr documentation states false can be
0464 * returned in some cases (e.g. offset > string length)
0465 * mb_substr never returns false, it will return an empty string instead.
0466 * This adopts the mb_substr approach
0467 *
0468 * Note on implementation: PCRE only supports repetitions of less than
0469 * 65536, in order to accept up to MAXINT values for offset and length,
0470 * we'll repeat a group of 65535 characters when needed.
0471 *
0472 * Note on implementation: calculating the number of characters in the
0473 * string is a relatively expensive operation, so we only carry it out when
0474 * necessary. It isn't necessary for +ve offsets and no specified length
0475 *
0476 * @author Chris Smith<chris@jalakai.co.uk>
0477 * @param string $str
0478 * @param integer $offset number of UTF-8 characters offset (from left)
0479 * @param integer $length (optional) length in UTF-8 characters from offset
0480 * @return mixed string or FALSE if failure
0481 */
0482 function utf8_substr($str, $offset, $length = NULL)
0483 {
0484 // generates E_NOTICE
0485 // for PHP4 objects, but not PHP5 objects
0486 $str = (string) $str;
0487 $offset = (int) $offset;
0488 if (!is_null($length))
0489 {
0490 $length = (int) $length;
0491 }
0492
0493 // handle trivial cases
0494 if ($length === 0 || ($offset < 0 && $length < 0 && $length < $offset))
0495 {
0496 return '';
0497 }
0498
0499 // normalise negative offsets (we could use a tail
0500 // anchored pattern, but they are horribly slow!)
0501 if ($offset < 0)
0502 {
0503 // see notes
0504 $strlen = utf8_strlen($str);
0505 $offset = $strlen + $offset;
0506 if ($offset < 0)
0507 {
0508 $offset = 0;
0509 }
0510 }
0511
0512 $op = '';
0513 $lp = '';
0514
0515 // establish a pattern for offset, a
0516 // non-captured group equal in length to offset
0517 if ($offset > 0)
0518 {
0519 $ox = (int) ($offset / 65535);
0520 $oy = $offset % 65535;
0521
0522 if ($ox)
0523 {
0524 $op = '(?:.{65535}){' . $ox . '}';
0525 }
0526
0527 $op = '^(?:' . $op . '.{' . $oy . '})';
0528 }
0529 else
0530 {
0531 // offset == 0; just anchor the pattern
0532 $op = '^';
0533 }
0534
0535 // establish a pattern for length
0536 if (is_null($length))
0537 {
0538 // the rest of the string
0539 $lp = '(.*)$';
0540 }
0541 else
0542 {
0543 if (!isset($strlen))
0544 {
0545 // see notes
0546 $strlen = utf8_strlen($str);
0547 }
0548
0549 // another trivial case
0550 if ($offset > $strlen)
0551 {
0552 return '';
0553 }
0554
0555 if ($length > 0)
0556 {
0557 // reduce any length that would
0558 // go passed the end of the string
0559 $length = min($strlen - $offset, $length);
0560
0561 $lx = (int) ($length / 65535);
0562 $ly = $length % 65535;
0563
0564 // negative length requires a captured group
0565 // of length characters
0566 if ($lx)
0567 {
0568 $lp = '(?:.{65535}){' . $lx . '}';
0569 }
0570 $lp = '(' . $lp . '.{'. $ly . '})';
0571 }
0572 else if ($length < 0)
0573 {
0574 if ($length < ($offset - $strlen))
0575 {
0576 return '';
0577 }
0578
0579 $lx = (int)((-$length) / 65535);
0580 $ly = (-$length) % 65535;
0581
0582 // negative length requires ... capture everything
0583 // except a group of -length characters
0584 // anchored at the tail-end of the string
0585 if ($lx)
0586 {
0587 $lp = '(?:.{65535}){' . $lx . '}';
0588 }
0589 $lp = '(.*)(?:' . $lp . '.{' . $ly . '})$';
0590 }
0591 }
0592
0593 if (!preg_match('#' . $op . $lp . '#us', $str, $match))
0594 {
0595 return '';
0596 }
0597
0598 return $match[1];
0599 }
0600
0601 /**
0602 * Return the length (in characters) of a UTF-8 string
0603 *
0604 * @param string $text UTF-8 string
0605 * @return integer Length (in chars) of given string
0606 */
0607 function utf8_strlen($text)
0608 {
0609 // Since utf8_decode is replacing multibyte characters to ? strlen works fine
0610 return strlen(utf8_decode($text));
0611 }
0612 }
0613
0614 /**
0615 * UTF-8 aware alternative to str_split
0616 * Convert a string to an array
0617 *
0618 * @author Harry Fuecks
0619 * @param string $str UTF-8 encoded
0620 * @param int $split_len number to characters to split string by
0621 * @return array characters in string reverses
0622 */
0623 function utf8_str_split($str, $split_len = 1)
0624 {
0625 if (!is_int($split_len) || $split_len < 1)
0626 {
0627 return false;
0628 }
0629
0630 $len = utf8_strlen($str);
0631 if ($len <= $split_len)
0632 {
0633 return array($str);
0634 }
0635
0636 preg_match_all('/.{' . $split_len . '}|[^\x00]{1,' . $split_len . '}$/us', $str, $ar);
0637 return $ar[0];
0638 }
0639
0640 /**
0641 * UTF-8 aware alternative to strspn
0642 * Find length of initial segment matching the mask
0643 *
0644 * @author Harry Fuecks
0645 */
0646 function utf8_strspn($str, $mask, $start = null, $length = null)
0647 {
0648 if ($start !== null || $length !== null)
0649 {
0650 $str = utf8_substr($str, $start, $length);
0651 }
0652
0653 preg_match('/^[' . $mask . ']+/u', $str, $matches);
0654
0655 if (isset($matches[0]))
0656 {
0657 return utf8_strlen($matches[0]);
0658 }
0659
0660 return 0;
0661 }
0662
0663 /**
0664 * UTF-8 aware alternative to ucfirst
0665 * Make a string's first character uppercase
0666 *
0667 * @author Harry Fuecks
0668 * @param string
0669 * @return string with first character as upper case (if applicable)
0670 */
0671 function utf8_ucfirst($str)
0672 {
0673 switch (utf8_strlen($str))
0674 {
0675 case 0:
0676 return '';
0677 break;
0678
0679 case 1:
0680 return utf8_strtoupper($str);
0681 break;
0682
0683 default:
0684 preg_match('/^(.{1})(.*)$/us', $str, $matches);
0685 return utf8_strtoupper($matches[1]) . $matches[2];
0686 break;
0687 }
0688 }
0689
0690 /**
0691 * Recode a string to UTF-8
0692 *
0693 * If the encoding is not supported, the string is returned as-is
0694 *
0695 * @param string $string Original string
0696 * @param string $encoding Original encoding (lowered)
0697 * @return string The string, encoded in UTF-8
0698 */
0699 function utf8_recode($string, $encoding)
0700 {
0701 $encoding = strtolower($encoding);
0702
0703 if ($encoding == 'utf-8' || !is_string($string) || empty($string))
0704 {
0705 return $string;
0706 }
0707
0708 // we force iso-8859-1 to be cp1252
0709 if ($encoding == 'iso-8859-1')
0710 {
0711 $encoding = 'cp1252';
0712 }
0713 // convert iso-8859-8-i to iso-8859-8
0714 else if ($encoding == 'iso-8859-8-i')
0715 {
0716 $encoding = 'iso-8859-8';
0717 $string = hebrev($string);
0718 }
0719
0720 // First, try iconv()
0721 if (function_exists('iconv'))
0722 {
0723 $ret = @iconv($encoding, 'utf-8', $string);
0724
0725 if (!empty($ret))
0726 {
0727 return $ret;
0728 }
0729 }
0730
0731 // Try the mb_string extension
0732 if (function_exists('mb_convert_encoding'))
0733 {
0734 // mbstring is nasty on PHP4, we must make *sure* that we send a good encoding
0735 switch ($encoding)
0736 {
0737 case 'iso-8859-1':
0738 case 'iso-8859-2':
0739 case 'iso-8859-4':
0740 case 'iso-8859-7':
0741 case 'iso-8859-9':
0742 case 'iso-8859-15':
0743 case 'windows-1251':
0744 case 'windows-1252':
0745 case 'cp1252':
0746 case 'shift_jis':
0747 case 'euc-kr':
0748 case 'big5':
0749 case 'gb2312':
0750 $ret = @mb_convert_encoding($string, 'utf-8', $encoding);
0751
0752 if (!empty($ret))
0753 {
0754 return $ret;
0755 }
0756 }
0757 }
0758
0759 // Try the recode extension
0760 if (function_exists('recode_string'))
0761 {
0762 $ret = @recode_string($encoding . '..utf-8', $string);
0763
0764 if (!empty($ret))
0765 {
0766 return $ret;
0767 }
0768 }
0769
0770 // If nothing works, check if we have a custom transcoder available
0771 if (!preg_match('#^[a-z0-9_ \\-]+$#', $encoding))
0772 {
0773 // Make sure the encoding name is alphanumeric, we don't want it to be abused into loading arbitrary files
0774 trigger_error('Unknown encoding: ' . $encoding, E_USER_ERROR);
0775 }
0776
0777 global $phpbb_root_path, $phpEx;
0778
0779 // iso-8859-* character encoding
0780 if (preg_match('/iso[_ -]?8859[_ -]?(\\d+)/', $encoding, $array))
0781 {
0782 switch ($array[1])
0783 {
0784 case '1':
0785 case '2':
0786 case '4':
0787 case '7':
0788 case '8':
0789 case '9':
0790 case '15':
0791 if (!function_exists('iso_8859_' . $array[1]))
0792 {
0793 if (!file_exists($phpbb_root_path . 'includes/utf/data/recode_basic.' . $phpEx))
0794 {
0795 trigger_error('Basic reencoder file is missing', E_USER_ERROR);
0796 }
0797 include($phpbb_root_path . 'includes/utf/data/recode_basic.' . $phpEx);
0798 }
0799 return call_user_func('iso_8859_' . $array[1], $string);
0800 break;
0801
0802 default:
0803 trigger_error('Unknown encoding: ' . $encoding, E_USER_ERROR);
0804 break;
0805 }
0806 }
0807
0808 // CP/WIN character encoding
0809 if (preg_match('/(?:cp|windows)[_\- ]?(\\d+)/', $encoding, $array))
0810 {
0811 switch ($array[1])
0812 {
0813 case '932':
0814 break;
0815 case '1250':
0816 case '1251':
0817 case '1252':
0818 case '1254':
0819 case '1255':
0820 case '1256':
0821 case '1257':
0822 case '874':
0823 if (!function_exists('cp' . $array[1]))
0824 {
0825 if (!file_exists($phpbb_root_path . 'includes/utf/data/recode_basic.' . $phpEx))
0826 {
0827 trigger_error('Basic reencoder file is missing', E_USER_ERROR);
0828 }
0829 include($phpbb_root_path . 'includes/utf/data/recode_basic.' . $phpEx);
0830 }
0831 return call_user_func('cp' . $array[1], $string);
0832 break;
0833
0834 default:
0835 trigger_error('Unknown encoding: ' . $encoding, E_USER_ERROR);
0836 break;
0837 }
0838 }
0839
0840 // TIS-620
0841 if (preg_match('/tis[_ -]?620/', $encoding))
0842 {
0843 if (!function_exists('tis_620'))
0844 {
0845 if (!file_exists($phpbb_root_path . 'includes/utf/data/recode_basic.' . $phpEx))
0846 {
0847 trigger_error('Basic reencoder file is missing', E_USER_ERROR);
0848 }
0849 include($phpbb_root_path . 'includes/utf/data/recode_basic.' . $phpEx);
0850 }
0851 return tis_620($string);
0852 }
0853
0854 // SJIS
0855 if (preg_match('/sjis(?:[_ -]?win)?|(?:cp|ibm)[_ -]?932|shift[_ -]?jis/', $encoding))
0856 {
0857 if (!function_exists('sjis'))
0858 {
0859 if (!file_exists($phpbb_root_path . 'includes/utf/data/recode_cjk.' . $phpEx))
0860 {
0861 trigger_error('CJK reencoder file is missing', E_USER_ERROR);
0862 }
0863 include($phpbb_root_path . 'includes/utf/data/recode_cjk.' . $phpEx);
0864 }
0865 return sjis($string);
0866 }
0867
0868 // EUC_KR
0869 if (preg_match('/euc[_ -]?kr/', $encoding))
0870 {
0871 if (!function_exists('euc_kr'))
0872 {
0873 if (!file_exists($phpbb_root_path . 'includes/utf/data/recode_cjk.' . $phpEx))
0874 {
0875 trigger_error('CJK reencoder file is missing', E_USER_ERROR);
0876 }
0877 include($phpbb_root_path . 'includes/utf/data/recode_cjk.' . $phpEx);
0878 }
0879 return euc_kr($string);
0880 }
0881
0882 // BIG-5
0883 if (preg_match('/big[_ -]?5/', $encoding))
0884 {
0885 if (!function_exists('big5'))
0886 {
0887 if (!file_exists($phpbb_root_path . 'includes/utf/data/recode_cjk.' . $phpEx))
0888 {
0889 trigger_error('CJK reencoder file is missing', E_USER_ERROR);
0890 }
0891 include($phpbb_root_path . 'includes/utf/data/recode_cjk.' . $phpEx);
0892 }
0893 return big5($string);
0894 }
0895
0896 // GB2312
0897 if (preg_match('/gb[_ -]?2312/', $encoding))
0898 {
0899 if (!function_exists('gb2312'))
0900 {
0901 if (!file_exists($phpbb_root_path . 'includes/utf/data/recode_cjk.' . $phpEx))
0902 {
0903 trigger_error('CJK reencoder file is missing', E_USER_ERROR);
0904 }
0905 include($phpbb_root_path . 'includes/utf/data/recode_cjk.' . $phpEx);
0906 }
0907 return gb2312($string);
0908 }
0909
0910 // Trigger an error?! Fow now just give bad data :-(
0911 trigger_error('Unknown encoding: ' . $encoding, E_USER_ERROR);
0912 //return $string; // use utf_normalizer::cleanup() ?
0913 }
0914
0915 /**
0916 * Replace all UTF-8 chars that are not in ASCII with their NCR
0917 *
0918 * @param string $text UTF-8 string in NFC
0919 * @return string ASCII string using NCRs for non-ASCII chars
0920 */
0921 function utf8_encode_ncr($text)
0922 {
0923 return preg_replace_callback('#[\\xC2-\\xF4][\\x80-\\xBF]{1,3}#', 'utf8_encode_ncr_callback', $text);
0924 }
0925
0926 /**
0927 * Callback used in encode_ncr()
0928 *
0929 * Takes a UTF-8 char and replaces it with its NCR. Attention, $m is an array
0930 *
0931 * @param array $m 0-based numerically indexed array passed by preg_replace_callback()
0932 * @return string A HTML NCR if the character is valid, or the original string otherwise
0933 */
0934 function utf8_encode_ncr_callback($m)
0935 {
0936 return '&#' . utf8_ord($m[0]) . ';';
0937 }
0938
0939 /**
0940 * Converts a UTF-8 char to an NCR
0941 *
0942 * @param string $chr UTF-8 char
0943 * @return integer UNICODE code point
0944 */
0945 function utf8_ord($chr)
0946 {
0947 switch (strlen($chr))
0948 {
0949 case 1:
0950 return ord($chr);
0951 break;
0952
0953 case 2:
0954 return ((ord($chr[0]) & 0x1F) << 6) | (ord($chr[1]) & 0x3F);
0955 break;
0956
0957 case 3:
0958 return ((ord($chr[0]) & 0x0F) << 12) | ((ord($chr[1]) & 0x3F) << 6) | (ord($chr[2]) & 0x3F);
0959 break;
0960
0961 case 4:
0962 return ((ord($chr[0]) & 0x07) << 18) | ((ord($chr[1]) & 0x3F) << 12) | ((ord($chr[2]) & 0x3F) << 6) | (ord($chr[3]) & 0x3F);
0963 break;
0964
0965 default:
0966 return $chr;
0967 }
0968 }
0969
0970 /**
0971 * Converts an NCR to a UTF-8 char
0972 *
0973 * @param int $cp UNICODE code point
0974 * @return string UTF-8 char
0975 */
0976 function utf8_chr($cp)
0977 {
0978 if ($cp > 0xFFFF)
0979 {
0980 return chr(0xF0 | ($cp >> 18)) . chr(0x80 | (($cp >> 12) & 0x3F)) . chr(0x80 | (($cp >> 6) & 0x3F)) . chr(0x80 | ($cp & 0x3F));
0981 }
0982 else if ($cp > 0x7FF)
0983 {
0984 return chr(0xE0 | ($cp >> 12)) . chr(0x80 | (($cp >> 6) & 0x3F)) . chr(0x80 | ($cp & 0x3F));
0985 }
0986 else if ($cp > 0x7F)
0987 {
0988 return chr(0xC0 | ($cp >> 6)) . chr(0x80 | ($cp & 0x3F));
0989 }
0990 else
0991 {
0992 return chr($cp);
0993 }
0994 }
0995
0996 /**
0997 * Convert Numeric Character References to UTF-8 chars
0998 *
0999 * Notes:
1000 * - we do not convert NCRs recursively, if you pass &#38; it will return &
1001 * - we DO NOT check for the existence of the Unicode characters, therefore an entity may be converted to an inexistent codepoint
1002 *
1003 * @param string $text String to convert, encoded in UTF-8 (no normal form required)
1004 * @return string UTF-8 string where NCRs have been replaced with the actual chars
1005 */
1006 function utf8_decode_ncr($text)
1007 {
1008 return preg_replace_callback('/&#([0-9]{1,6}|x[0-9A-F]{1,5});/i', 'utf8_decode_ncr_callback', $text);
1009 }
1010
1011 /**
1012 * Callback used in decode_ncr()
1013 *
1014 * Takes a NCR (in decimal or hexadecimal) and returns a UTF-8 char. Attention, $m is an array.
1015 * It will ignore most of invalid NCRs, but not all!
1016 *
1017 * @param array $m 0-based numerically indexed array passed by preg_replace_callback()
1018 * @return string UTF-8 char
1019 */
1020 function utf8_decode_ncr_callback($m)
1021 {
1022 $cp = (strncasecmp($m[1], 'x', 1)) ? $m[1] : hexdec(substr($m[1], 1));
1023
1024 return utf8_chr($cp);
1025 }
1026
1027 /**
1028 * Case folds a unicode string as per Unicode 5.0, section 3.13
1029 *
1030 * @param string $text text to be case folded
1031 * @param string $option determines how we will fold the cases
1032 * @return string case folded text
1033 */
1034 function utf8_case_fold($text, $option = 'full')
1035 {
1036 static $uniarray = array();
1037 global $phpbb_root_path, $phpEx;
1038
1039 // common is always set
1040 if (!isset($uniarray['c']))
1041 {
1042 $uniarray['c'] = include($phpbb_root_path . 'includes/utf/data/case_fold_c.' . $phpEx);
1043 }
1044
1045 // only set full if we need to
1046 if ($option === 'full' && !isset($uniarray['f']))
1047 {
1048 $uniarray['f'] = include($phpbb_root_path . 'includes/utf/data/case_fold_f.' . $phpEx);
1049 }
1050
1051 // only set simple if we need to
1052 if ($option !== 'full' && !isset($uniarray['s']))
1053 {
1054 $uniarray['s'] = include($phpbb_root_path . 'includes/utf/data/case_fold_s.' . $phpEx);
1055 }
1056
1057 // common is always replaced
1058 $text = strtr($text, $uniarray['c']);
1059
1060 if ($option === 'full')
1061 {
1062 // full replaces a character with multiple characters
1063 $text = strtr($text, $uniarray['f']);
1064 }
1065 else
1066 {
1067 // simple replaces a character with another character
1068 $text = strtr($text, $uniarray['s']);
1069 }
1070
1071 return $text;
1072 }
1073
1074 /**
1075 * Takes the input and does a "special" case fold. It does minor normalization
1076 * and returns NFKC compatable text
1077 *
1078 * @param string $text text to be case folded
1079 * @param string $option determines how we will fold the cases
1080 * @return string case folded text
1081 */
1082 function utf8_case_fold_nfkc($text, $option = 'full')
1083 {
1084 static $fc_nfkc_closure = array(
1085 "\xCD\xBA" => "\x20\xCE\xB9",
1086 "\xCF\x92" => "\xCF\x85",
1087 "\xCF\x93" => "\xCF\x8D",
1088 "\xCF\x94" => "\xCF\x8B",
1089 "\xCF\xB2" => "\xCF\x83",
1090 "\xCF\xB9" => "\xCF\x83",
1091 "\xE1\xB4\xAC" => "\x61",
1092 "\xE1\xB4\xAD" => "\xC3\xA6",
1093 "\xE1\xB4\xAE" => "\x62",
1094 "\xE1\xB4\xB0" => "\x64",
1095 "\xE1\xB4\xB1" => "\x65",
1096 "\xE1\xB4\xB2" => "\xC7\x9D",
1097 "\xE1\xB4\xB3" => "\x67",
1098 "\xE1\xB4\xB4" => "\x68",
1099 "\xE1\xB4\xB5" => "\x69",
1100 "\xE1\xB4\xB6" => "\x6A",
1101 "\xE1\xB4\xB7" => "\x6B",
1102 "\xE1\xB4\xB8" => "\x6C",
1103 "\xE1\xB4\xB9" => "\x6D",
1104 "\xE1\xB4\xBA" => "\x6E",
1105 "\xE1\xB4\xBC" => "\x6F",
1106 "\xE1\xB4\xBD" => "\xC8\xA3",
1107 "\xE1\xB4\xBE" => "\x70",
1108 "\xE1\xB4\xBF" => "\x72",
1109 "\xE1\xB5\x80" => "\x74",
1110 "\xE1\xB5\x81" => "\x75",
1111 "\xE1\xB5\x82" => "\x77",
1112 "\xE2\x82\xA8" => "\x72\x73",
1113 "\xE2\x84\x82" => "\x63",
1114 "\xE2\x84\x83" => "\xC2\xB0\x63",
1115 "\xE2\x84\x87" => "\xC9\x9B",
1116 "\xE2\x84\x89" => "\xC2\xB0\x66",
1117 "\xE2\x84\x8B" => "\x68",
1118 "\xE2\x84\x8C" => "\x68",
1119 "\xE2\x84\x8D" => "\x68",
1120 "\xE2\x84\x90" => "\x69",
1121 "\xE2\x84\x91" => "\x69",
1122 "\xE2\x84\x92" => "\x6C",
1123 "\xE2\x84\x95" => "\x6E",
1124 "\xE2\x84\x96" => "\x6E\x6F",
1125 "\xE2\x84\x99" => "\x70",
1126 "\xE2\x84\x9A" => "\x71",
1127 "\xE2\x84\x9B" => "\x72",
1128 "\xE2\x84\x9C" => "\x72",
1129 "\xE2\x84\x9D" => "\x72",
1130 "\xE2\x84\xA0" => "\x73\x6D",
1131 "\xE2\x84\xA1" => "\x74\x65\x6C",
1132 "\xE2\x84\xA2" => "\x74\x6D",
1133 "\xE2\x84\xA4" => "\x7A",
1134 "\xE2\x84\xA8" => "\x7A",
1135 "\xE2\x84\xAC" => "\x62",
1136 "\xE2\x84\xAD" => "\x63",
1137 "\xE2\x84\xB0" => "\x65",
1138 "\xE2\x84\xB1" => "\x66",
1139 "\xE2\x84\xB3" => "\x6D",
1140 "\xE2\x84\xBB" => "\x66\x61\x78",
1141 "\xE2\x84\xBE" => "\xCE\xB3",
1142 "\xE2\x84\xBF" => "\xCF\x80",
1143 "\xE2\x85\x85" => "\x64",
1144 "\xE3\x89\x90" => "\x70\x74\x65",
1145 "\xE3\x8B\x8C" => "\x68\x67",
1146 "\xE3\x8B\x8E" => "\x65\x76",
1147 "\xE3\x8B\x8F" => "\x6C\x74\x64",
1148 "\xE3\x8D\xB1" => "\x68\x70\x61",
1149 "\xE3\x8D\xB3" => "\x61\x75",
1150 "\xE3\x8D\xB5" => "\x6F\x76",
1151 "\xE3\x8D\xBA" => "\x69\x75",
1152 "\xE3\x8E\x80" => "\x70\x61",
1153 "\xE3\x8E\x81" => "\x6E\x61",
1154 "\xE3\x8E\x82" => "\xCE\xBC\x61",
1155 "\xE3\x8E\x83" => "\x6D\x61",
1156 "\xE3\x8E\x84" => "\x6B\x61",
1157 "\xE3\x8E\x85" => "\x6B\x62",
1158 "\xE3\x8E\x86" => "\x6D\x62",
1159 "\xE3\x8E\x87" => "\x67\x62",
1160 "\xE3\x8E\x8A" => "\x70\x66",
1161 "\xE3\x8E\x8B" => "\x6E\x66",
1162 "\xE3\x8E\x8C" => "\xCE\xBC\x66",
1163 "\xE3\x8E\x90" => "\x68\x7A",
1164 "\xE3\x8E\x91" => "\x6B\x68\x7A",
1165 "\xE3\x8E\x92" => "\x6D\x68\x7A",
1166 "\xE3\x8E\x93" => "\x67\x68\x7A",
1167 "\xE3\x8E\x94" => "\x74\x68\x7A",
1168 "\xE3\x8E\xA9" => "\x70\x61",
1169 "\xE3\x8E\xAA" => "\x6B\x70\x61",
1170 "\xE3\x8E\xAB" => "\x6D\x70\x61",
1171 "\xE3\x8E\xAC" => "\x67\x70\x61",
1172 "\xE3\x8E\xB4" => "\x70\x76",
1173 "\xE3\x8E\xB5" => "\x6E\x76",
1174 "\xE3\x8E\xB6" => "\xCE\xBC\x76",
1175 "\xE3\x8E\xB7" => "\x6D\x76",
1176 "\xE3\x8E\xB8" => "\x6B\x76",
1177 "\xE3\x8E\xB9" => "\x6D\x76",
1178 "\xE3\x8E\xBA" => "\x70\x77",
1179 "\xE3\x8E\xBB" => "\x6E\x77",
1180 "\xE3\x8E\xBC" => "\xCE\xBC\x77",
1181 "\xE3\x8E\xBD" => "\x6D\x77",
1182 "\xE3\x8E\xBE" => "\x6B\x77",
1183 "\xE3\x8E\xBF" => "\x6D\x77",
1184 "\xE3\x8F\x80" => "\x6B\xCF\x89",
1185 "\xE3\x8F\x81" => "\x6D\xCF\x89",
1186 "\xE3\x8F\x83" => "\x62\x71",
1187 "\xE3\x8F\x86" => "\x63\xE2\x88\x95\x6B\x67",
1188 "\xE3\x8F\x87" => "\x63\x6F\x2E",
1189 "\xE3\x8F\x88" => "\x64\x62",
1190 "\xE3\x8F\x89" => "\x67\x79",
1191 "\xE3\x8F\x8B" => "\x68\x70",
1192 "\xE3\x8F\x8D" => "\x6B\x6B",
1193 "\xE3\x8F\x8E" => "\x6B\x6D",
1194 "\xE3\x8F\x97" => "\x70\x68",
1195 "\xE3\x8F\x99" => "\x70\x70\x6D",
1196 "\xE3\x8F\x9A" => "\x70\x72",
1197 "\xE3\x8F\x9C" => "\x73\x76",
1198 "\xE3\x8F\x9D" => "\x77\x62",
1199 "\xE3\x8F\x9E" => "\x76\xE2\x88\x95\x6D",
1200 "\xE3\x8F\x9F" => "\x61\xE2\x88\x95\x6D",
1201 "\xF0\x9D\x90\x80" => "\x61",
1202 "\xF0\x9D\x90\x81" => "\x62",
1203 "\xF0\x9D\x90\x82" => "\x63",
1204 "\xF0\x9D\x90\x83" => "\x64",
1205 "\xF0\x9D\x90\x84" => "\x65",
1206 "\xF0\x9D\x90\x85" => "\x66",
1207 "\xF0\x9D\x90\x86" => "\x67",
1208 "\xF0\x9D\x90\x87" => "\x68",
1209 "\xF0\x9D\x90\x88" => "\x69",
1210 "\xF0\x9D\x90\x89" => "\x6A",
1211 "\xF0\x9D\x90\x8A" => "\x6B",
1212 "\xF0\x9D\x90\x8B" => "\x6C",
1213 "\xF0\x9D\x90\x8C" => "\x6D",
1214 "\xF0\x9D\x90\x8D" => "\x6E",
1215 "\xF0\x9D\x90\x8E" => "\x6F",
1216 "\xF0\x9D\x90\x8F" => "\x70",
1217 "\xF0\x9D\x90\x90" => "\x71",
1218 "\xF0\x9D\x90\x91" => "\x72",
1219 "\xF0\x9D\x90\x92" => "\x73",
1220 "\xF0\x9D\x90\x93" => "\x74",
1221 "\xF0\x9D\x90\x94" => "\x75",
1222 "\xF0\x9D\x90\x95" => "\x76",
1223 "\xF0\x9D\x90\x96" => "\x77",
1224 "\xF0\x9D\x90\x97" => "\x78",
1225 "\xF0\x9D\x90\x98" => "\x79",
1226 "\xF0\x9D\x90\x99" => "\x7A",
1227 "\xF0\x9D\x90\xB4" => "\x61",
1228 "\xF0\x9D\x90\xB5" => "\x62",
1229 "\xF0\x9D\x90\xB6" => "\x63",
1230 "\xF0\x9D\x90\xB7" => "\x64",
1231 "\xF0\x9D\x90\xB8" => "\x65",
1232 "\xF0\x9D\x90\xB9" => "\x66",
1233 "\xF0\x9D\x90\xBA" => "\x67",
1234 "\xF0\x9D\x90\xBB" => "\x68",
1235 "\xF0\x9D\x90\xBC" => "\x69",
1236 "\xF0\x9D\x90\xBD" => "\x6A",
1237 "\xF0\x9D\x90\xBE" => "\x6B",
1238 "\xF0\x9D\x90\xBF" => "\x6C",
1239 "\xF0\x9D\x91\x80" => "\x6D",
1240 "\xF0\x9D\x91\x81" => "\x6E",
1241 "\xF0\x9D\x91\x82" => "\x6F",
1242 "\xF0\x9D\x91\x83" => "\x70",
1243 "\xF0\x9D\x91\x84" => "\x71",
1244 "\xF0\x9D\x91\x85" => "\x72",
1245 "\xF0\x9D\x91\x86" => "\x73",
1246 "\xF0\x9D\x91\x87" => "\x74",
1247 "\xF0\x9D\x91\x88" => "\x75",
1248 "\xF0\x9D\x91\x89" => "\x76",
1249 "\xF0\x9D\x91\x8A" => "\x77",
1250 "\xF0\x9D\x91\x8B" => "\x78",
1251 "\xF0\x9D\x91\x8C" => "\x79",
1252 "\xF0\x9D\x91\x8D" => "\x7A",
1253 "\xF0\x9D\x91\xA8" => "\x61",
1254 "\xF0\x9D\x91\xA9" => "\x62",
1255 "\xF0\x9D\x91\xAA" => "\x63",
1256 "\xF0\x9D\x91\xAB" => "\x64",
1257 "\xF0\x9D\x91\xAC" => "\x65",
1258 "\xF0\x9D\x91\xAD" => "\x66",
1259 "\xF0\x9D\x91\xAE" => "\x67",
1260 "\xF0\x9D\x91\xAF" => "\x68",
1261 "\xF0\x9D\x91\xB0" => "\x69",
1262 "\xF0\x9D\x91\xB1" => "\x6A",
1263 "\xF0\x9D\x91\xB2" => "\x6B",
1264 "\xF0\x9D\x91\xB3" => "\x6C",
1265 "\xF0\x9D\x91\xB4" => "\x6D",
1266 "\xF0\x9D\x91\xB5" => "\x6E",
1267 "\xF0\x9D\x91\xB6" => "\x6F",
1268 "\xF0\x9D\x91\xB7" => "\x70",
1269 "\xF0\x9D\x91\xB8" => "\x71",
1270 "\xF0\x9D\x91\xB9" => "\x72",
1271 "\xF0\x9D\x91\xBA" => "\x73",
1272 "\xF0\x9D\x91\xBB" => "\x74",
1273 "\xF0\x9D\x91\xBC" => "\x75",
1274 "\xF0\x9D\x91\xBD" => "\x76",
1275 "\xF0\x9D\x91\xBE" => "\x77",
1276 "\xF0\x9D\x91\xBF" => "\x78",
1277 "\xF0\x9D\x92\x80" => "\x79",
1278 "\xF0\x9D\x92\x81" => "\x7A",
1279 "\xF0\x9D\x92\x9C" => "\x61",
1280 "\xF0\x9D\x92\x9E" => "\x63",
1281 "\xF0\x9D\x92\x9F" => "\x64",
1282 "\xF0\x9D\x92\xA2" => "\x67",
1283 "\xF0\x9D\x92\xA5" => "\x6A",
1284 "\xF0\x9D\x92\xA6" => "\x6B",
1285 "\xF0\x9D\x92\xA9" => "\x6E",
1286 "\xF0\x9D\x92\xAA" => "\x6F",
1287 "\xF0\x9D\x92\xAB" => "\x70",
1288 "\xF0\x9D\x92\xAC" => "\x71",
1289 "\xF0\x9D\x92\xAE" => "\x73",
1290 "\xF0\x9D\x92\xAF" => "\x74",
1291 "\xF0\x9D\x92\xB0" => "\x75",
1292 "\xF0\x9D\x92\xB1" => "\x76",
1293 "\xF0\x9D\x92\xB2" => "\x77",
1294 "\xF0\x9D\x92\xB3" => "\x78",
1295 "\xF0\x9D\x92\xB4" => "\x79",
1296 "\xF0\x9D\x92\xB5" => "\x7A",
1297 "\xF0\x9D\x93\x90" => "\x61",
1298 "\xF0\x9D\x93\x91" => "\x62",
1299 "\xF0\x9D\x93\x92" => "\x63",
1300 "\xF0\x9D\x93\x93" => "\x64",
1301 "\xF0\x9D\x93\x94" => "\x65",
1302 "\xF0\x9D\x93\x95" => "\x66",
1303 "\xF0\x9D\x93\x96" => "\x67",
1304 "\xF0\x9D\x93\x97" => "\x68",
1305 "\xF0\x9D\x93\x98" => "\x69",
1306 "\xF0\x9D\x93\x99" => "\x6A",
1307 "\xF0\x9D\x93\x9A" => "\x6B",
1308 "\xF0\x9D\x93\x9B" => "\x6C",
1309 "\xF0\x9D\x93\x9C" => "\x6D",
1310 "\xF0\x9D\x93\x9D" => "\x6E",
1311 "\xF0\x9D\x93\x9E" => "\x6F",
1312 "\xF0\x9D\x93\x9F" => "\x70",
1313 "\xF0\x9D\x93\xA0" => "\x71",
1314 "\xF0\x9D\x93\xA1" => "\x72",
1315 "\xF0\x9D\x93\xA2" => "\x73",
1316 "\xF0\x9D\x93\xA3" => "\x74",
1317 "\xF0\x9D\x93\xA4" => "\x75",
1318 "\xF0\x9D\x93\xA5" => "\x76",
1319 "\xF0\x9D\x93\xA6" => "\x77",
1320 "\xF0\x9D\x93\xA7" => "\x78",
1321 "\xF0\x9D\x93\xA8" => "\x79",
1322 "\xF0\x9D\x93\xA9" => "\x7A",
1323 "\xF0\x9D\x94\x84" => "\x61",
1324 "\xF0\x9D\x94\x85" => "\x62",
1325 "\xF0\x9D\x94\x87" => "\x64",
1326 "\xF0\x9D\x94\x88" => "\x65",
1327 "\xF0\x9D\x94\x89" => "\x66",
1328 "\xF0\x9D\x94\x8A" => "\x67",
1329 "\xF0\x9D\x94\x8D" => "\x6A",
1330 "\xF0\x9D\x94\x8E" => "\x6B",
1331 "\xF0\x9D\x94\x8F" => "\x6C",
1332 "\xF0\x9D\x94\x90" => "\x6D",
1333 "\xF0\x9D\x94\x91" => "\x6E",
1334 "\xF0\x9D\x94\x92" => "\x6F",
1335 "\xF0\x9D\x94\x93" => "\x70",
1336 "\xF0\x9D\x94\x94" => "\x71",
1337 "\xF0\x9D\x94\x96" => "\x73",
1338 "\xF0\x9D\x94\x97" => "\x74",
1339 "\xF0\x9D\x94\x98" => "\x75",
1340 "\xF0\x9D\x94\x99" => "\x76",
1341 "\xF0\x9D\x94\x9A" => "\x77",
1342 "\xF0\x9D\x94\x9B" => "\x78",
1343 "\xF0\x9D\x94\x9C" => "\x79",
1344 "\xF0\x9D\x94\xB8" => "\x61",
1345 "\xF0\x9D\x94\xB9" => "\x62",
1346 "\xF0\x9D\x94\xBB" => "\x64",
1347 "\xF0\x9D\x94\xBC" => "\x65",
1348 "\xF0\x9D\x94\xBD" => "\x66",
1349 "\xF0\x9D\x94\xBE" => "\x67",
1350 "\xF0\x9D\x95\x80" => "\x69",
1351 "\xF0\x9D\x95\x81" => "\x6A",
1352 "\xF0\x9D\x95\x82" => "\x6B",
1353 "\xF0\x9D\x95\x83" => "\x6C",
1354 "\xF0\x9D\x95\x84" => "\x6D",
1355 "\xF0\x9D\x95\x86" => "\x6F",
1356 "\xF0\x9D\x95\x8A" => "\x73",
1357 "\xF0\x9D\x95\x8B" => "\x74",
1358 "\xF0\x9D\x95\x8C" => "\x75",
1359 "\xF0\x9D\x95\x8D" => "\x76",
1360 "\xF0\x9D\x95\x8E" => "\x77",
1361 "\xF0\x9D\x95\x8F" => "\x78",
1362 "\xF0\x9D\x95\x90" => "\x79",
1363 "\xF0\x9D\x95\xAC" => "\x61",
1364 "\xF0\x9D\x95\xAD" => "\x62",
1365 "\xF0\x9D\x95\xAE" => "\x63",
1366 "\xF0\x9D\x95\xAF" => "\x64",
1367 "\xF0\x9D\x95\xB0" => "\x65",
1368 "\xF0\x9D\x95\xB1" => "\x66",
1369 "\xF0\x9D\x95\xB2" => "\x67",
1370 "\xF0\x9D\x95\xB3" => "\x68",
1371 "\xF0\x9D\x95\xB4" => "\x69",
1372 "\xF0\x9D\x95\xB5" => "\x6A",
1373 "\xF0\x9D\x95\xB6" => "\x6B",
1374 "\xF0\x9D\x95\xB7" => "\x6C",
1375 "\xF0\x9D\x95\xB8" => "\x6D",
1376 "\xF0\x9D\x95\xB9" => "\x6E",
1377 "\xF0\x9D\x95\xBA" => "\x6F",
1378 "\xF0\x9D\x95\xBB" => "\x70",
1379 "\xF0\x9D\x95\xBC" => "\x71",
1380 "\xF0\x9D\x95\xBD" => "\x72",
1381 "\xF0\x9D\x95\xBE" => "\x73",
1382 "\xF0\x9D\x95\xBF" => "\x74",
1383 "\xF0\x9D\x96\x80" => "\x75",
1384 "\xF0\x9D\x96\x81" => "\x76",
1385 "\xF0\x9D\x96\x82" => "\x77",
1386 "\xF0\x9D\x96\x83" => "\x78",
1387 "\xF0\x9D\x96\x84" => "\x79",
1388 "\xF0\x9D\x96\x85" => "\x7A",
1389 "\xF0\x9D\x96\xA0" => "\x61",
1390 "\xF0\x9D\x96\xA1" => "\x62",
1391 "\xF0\x9D\x96\xA2" => "\x63",
1392 "\xF0\x9D\x96\xA3" => "\x64",
1393 "\xF0\x9D\x96\xA4" => "\x65",
1394 "\xF0\x9D\x96\xA5" => "\x66",
1395 "\xF0\x9D\x96\xA6" => "\x67",
1396 "\xF0\x9D\x96\xA7" => "\x68",
1397 "\xF0\x9D\x96\xA8" => "\x69",
1398 "\xF0\x9D\x96\xA9" => "\x6A",
1399 "\xF0\x9D\x96\xAA" => "\x6B",
1400 "\xF0\x9D\x96\xAB" => "\x6C",
1401 "\xF0\x9D\x96\xAC" => "\x6D",
1402 "\xF0\x9D\x96\xAD" => "\x6E",
1403 "\xF0\x9D\x96\xAE" => "\x6F",
1404 "\xF0\x9D\x96\xAF" => "\x70",
1405 "\xF0\x9D\x96\xB0" => "\x71",
1406 "\xF0\x9D\x96\xB1" => "\x72",
1407 "\xF0\x9D\x96\xB2" => "\x73",
1408 "\xF0\x9D\x96\xB3" => "\x74",
1409 "\xF0\x9D\x96\xB4" => "\x75",
1410 "\xF0\x9D\x96\xB5" => "\x76",
1411 "\xF0\x9D\x96\xB6" => "\x77",
1412 "\xF0\x9D\x96\xB7" => "\x78",
1413 "\xF0\x9D\x96\xB8" => "\x79",
1414 "\xF0\x9D\x96\xB9" => "\x7A",
1415 "\xF0\x9D\x97\x94" => "\x61",
1416 "\xF0\x9D\x97\x95" => "\x62",
1417 "\xF0\x9D\x97\x96" => "\x63",
1418 "\xF0\x9D\x97\x97" => "\x64",
1419 "\xF0\x9D\x97\x98" => "\x65",
1420 "\xF0\x9D\x97\x99" => "\x66",
1421 "\xF0\x9D\x97\x9A" => "\x67",
1422 "\xF0\x9D\x97\x9B" => "\x68",
1423 "\xF0\x9D\x97\x9C" => "\x69",
1424 "\xF0\x9D\x97\x9D" => "\x6A",
1425 "\xF0\x9D\x97\x9E" => "\x6B",
1426 "\xF0\x9D\x97\x9F" => "\x6C",
1427 "\xF0\x9D\x97\xA0" => "\x6D",
1428 "\xF0\x9D\x97\xA1" => "\x6E",
1429 "\xF0\x9D\x97\xA2" => "\x6F",
1430 "\xF0\x9D\x97\xA3" => "\x70",
1431 "\xF0\x9D\x97\xA4" => "\x71",
1432 "\xF0\x9D\x97\xA5" => "\x72",
1433 "\xF0\x9D\x97\xA6" => "\x73",
1434 "\xF0\x9D\x97\xA7" => "\x74",
1435 "\xF0\x9D\x97\xA8" => "\x75",
1436 "\xF0\x9D\x97\xA9" => "\x76",
1437 "\xF0\x9D\x97\xAA" => "\x77",
1438 "\xF0\x9D\x97\xAB" => "\x78",
1439 "\xF0\x9D\x97\xAC" => "\x79",
1440 "\xF0\x9D\x97\xAD" => "\x7A",
1441 "\xF0\x9D\x98\x88" => "\x61",
1442 "\xF0\x9D\x98\x89" => "\x62",
1443 "\xF0\x9D\x98\x8A" => "\x63",
1444 "\xF0\x9D\x98\x8B" => "\x64",
1445 "\xF0\x9D\x98\x8C" => "\x65",
1446 "\xF0\x9D\x98\x8D" => "\x66",
1447 "\xF0\x9D\x98\x8E" => "\x67",
1448 "\xF0\x9D\x98\x8F" => "\x68",
1449 "\xF0\x9D\x98\x90" => "\x69",
1450 "\xF0\x9D\x98\x91" => "\x6A",
1451 "\xF0\x9D\x98\x92" => "\x6B",
1452 "\xF0\x9D\x98\x93" => "\x6C",
1453 "\xF0\x9D\x98\x94" => "\x6D",
1454 "\xF0\x9D\x98\x95" => "\x6E",
1455 "\xF0\x9D\x98\x96" => "\x6F",
1456 "\xF0\x9D\x98\x97" => "\x70",
1457 "\xF0\x9D\x98\x98" => "\x71",
1458 "\xF0\x9D\x98\x99" => "\x72",
1459 "\xF0\x9D\x98\x9A" => "\x73",
1460 "\xF0\x9D\x98\x9B" => "\x74",
1461 "\xF0\x9D\x98\x9C" => "\x75",
1462 "\xF0\x9D\x98\x9D" => "\x76",
1463 "\xF0\x9D\x98\x9E" => "\x77",
1464 "\xF0\x9D\x98\x9F" => "\x78",
1465 "\xF0\x9D\x98\xA0" => "\x79",
1466 "\xF0\x9D\x98\xA1" => "\x7A",
1467 "\xF0\x9D\x98\xBC" => "\x61",
1468 "\xF0\x9D\x98\xBD" => "\x62",
1469 "\xF0\x9D\x98\xBE" => "\x63",
1470 "\xF0\x9D\x98\xBF" => "\x64",
1471 "\xF0\x9D\x99\x80" => "\x65",
1472 "\xF0\x9D\x99\x81" => "\x66",
1473 "\xF0\x9D\x99\x82" => "\x67",
1474 "\xF0\x9D\x99\x83" => "\x68",
1475 "\xF0\x9D\x99\x84" => "\x69",
1476 "\xF0\x9D\x99\x85" => "\x6A",
1477 "\xF0\x9D\x99\x86" => "\x6B",
1478 "\xF0\x9D\x99\x87" => "\x6C",
1479 "\xF0\x9D\x99\x88" => "\x6D",
1480 "\xF0\x9D\x99\x89" => "\x6E",
1481 "\xF0\x9D\x99\x8A" => "\x6F",
1482 "\xF0\x9D\x99\x8B" => "\x70",
1483 "\xF0\x9D\x99\x8C" => "\x71",
1484 "\xF0\x9D\x99\x8D" => "\x72",
1485 "\xF0\x9D\x99\x8E" => "\x73",
1486 "\xF0\x9D\x99\x8F" => "\x74",
1487 "\xF0\x9D\x99\x90" => "\x75",
1488 "\xF0\x9D\x99\x91" => "\x76",
1489 "\xF0\x9D\x99\x92" => "\x77",
1490 "\xF0\x9D\x99\x93" => "\x78",
1491 "\xF0\x9D\x99\x94" => "\x79",
1492 "\xF0\x9D\x99\x95" => "\x7A",
1493 "\xF0\x9D\x99\xB0" => "\x61",
1494 "\xF0\x9D\x99\xB1" => "\x62",
1495 "\xF0\x9D\x99\xB2" => "\x63",
1496 "\xF0\x9D\x99\xB3" => "\x64",
1497 "\xF0\x9D\x99\xB4" => "\x65",
1498 "\xF0\x9D\x99\xB5" => "\x66",
1499 "\xF0\x9D\x99\xB6" => "\x67",
1500 "\xF0\x9D\x99\xB7" => "\x68",
1501 "\xF0\x9D\x99\xB8" => "\x69",
1502 "\xF0\x9D\x99\xB9" => "\x6A",
1503 "\xF0\x9D\x99\xBA" => "\x6B",
1504 "\xF0\x9D\x99\xBB" => "\x6C",
1505 "\xF0\x9D\x99\xBC" => "\x6D",
1506 "\xF0\x9D\x99\xBD" => "\x6E",
1507 "\xF0\x9D\x99\xBE" => "\x6F",
1508 "\xF0\x9D\x99\xBF" => "\x70",
1509 "\xF0\x9D\x9A\x80" => "\x71",
1510 "\xF0\x9D\x9A\x81" => "\x72",
1511 "\xF0\x9D\x9A\x82" => "\x73",
1512 "\xF0\x9D\x9A\x83" => "\x74",
1513 "\xF0\x9D\x9A\x84" => "\x75",
1514 "\xF0\x9D\x9A\x85" => "\x76",
1515 "\xF0\x9D\x9A\x86" => "\x77",
1516 "\xF0\x9D\x9A\x87" => "\x78",
1517 "\xF0\x9D\x9A\x88" => "\x79",
1518 "\xF0\x9D\x9A\x89" => "\x7A",
1519 "\xF0\x9D\x9A\xA8" => "\xCE\xB1",
1520 "\xF0\x9D\x9A\xA9" => "\xCE\xB2",
1521 "\xF0\x9D\x9A\xAA" => "\xCE\xB3",
1522 "\xF0\x9D\x9A\xAB" => "\xCE\xB4",
1523 "\xF0\x9D\x9A\xAC" => "\xCE\xB5",
1524 "\xF0\x9D\x9A\xAD" => "\xCE\xB6",
1525 "\xF0\x9D\x9A\xAE" => "\xCE\xB7",
1526 "\xF0\x9D\x9A\xAF" => "\xCE\xB8",
1527 "\xF0\x9D\x9A\xB0" => "\xCE\xB9",
1528 "\xF0\x9D\x9A\xB1" => "\xCE\xBA",
1529 "\xF0\x9D\x9A\xB2" => "\xCE\xBB",
1530 "\xF0\x9D\x9A\xB3" => "\xCE\xBC",
1531 "\xF0\x9D\x9A\xB4" => "\xCE\xBD",
1532 "\xF0\x9D\x9A\xB5" => "\xCE\xBE",
1533 "\xF0\x9D\x9A\xB6" => "\xCE\xBF",
1534 "\xF0\x9D\x9A\xB7" => "\xCF\x80",
1535 "\xF0\x9D\x9A\xB8" => "\xCF\x81",
1536 "\xF0\x9D\x9A\xB9" => "\xCE\xB8",
1537 "\xF0\x9D\x9A\xBA" => "\xCF\x83",
1538 "\xF0\x9D\x9A\xBB" => "\xCF\x84",
1539 "\xF0\x9D\x9A\xBC" => "\xCF\x85",
1540 "\xF0\x9D\x9A\xBD" => "\xCF\x86",
1541 "\xF0\x9D\x9A\xBE" => "\xCF\x87",
1542 "\xF0\x9D\x9A\xBF" => "\xCF\x88",
1543 "\xF0\x9D\x9B\x80" => "\xCF\x89",
1544 "\xF0\x9D\x9B\x93" => "\xCF\x83",
1545 "\xF0\x9D\x9B\xA2" => "\xCE\xB1",
1546 "\xF0\x9D\x9B\xA3" => "\xCE\xB2",
1547 "\xF0\x9D\x9B\xA4" => "\xCE\xB3",
1548 "\xF0\x9D\x9B\xA5" => "\xCE\xB4",
1549 "\xF0\x9D\x9B\xA6" => "\xCE\xB5",
1550 "\xF0\x9D\x9B\xA7" => "\xCE\xB6",
1551 "\xF0\x9D\x9B\xA8" => "\xCE\xB7",
1552 "\xF0\x9D\x9B\xA9" => "\xCE\xB8",
1553 "\xF0\x9D\x9B\xAA" => "\xCE\xB9",
1554 "\xF0\x9D\x9B\xAB" => "\xCE\xBA",
1555 "\xF0\x9D\x9B\xAC" => "\xCE\xBB",
1556 "\xF0\x9D\x9B\xAD" => "\xCE\xBC",
1557 "\xF0\x9D\x9B\xAE" => "\xCE\xBD",
1558 "\xF0\x9D\x9B\xAF" => "\xCE\xBE",
1559 "\xF0\x9D\x9B\xB0" => "\xCE\xBF",
1560 "\xF0\x9D\x9B\xB1" => "\xCF\x80",
1561 "\xF0\x9D\x9B\xB2" => "\xCF\x81",
1562 "\xF0\x9D\x9B\xB3" => "\xCE\xB8",
1563 "\xF0\x9D\x9B\xB4" => "\xCF\x83",
1564 "\xF0\x9D\x9B\xB5" => "\xCF\x84",
1565 "\xF0\x9D\x9B\xB6" => "\xCF\x85",
1566 "\xF0\x9D\x9B\xB7" => "\xCF\x86",
1567 "\xF0\x9D\x9B\xB8" => "\xCF\x87",
1568 "\xF0\x9D\x9B\xB9" => "\xCF\x88",
1569 "\xF0\x9D\x9B\xBA" => "\xCF\x89",
1570 "\xF0\x9D\x9C\x8D" => "\xCF\x83",
1571 "\xF0\x9D\x9C\x9C" => "\xCE\xB1",
1572 "\xF0\x9D\x9C\x9D" => "\xCE\xB2",
1573 "\xF0\x9D\x9C\x9E" => "\xCE\xB3",
1574 "\xF0\x9D\x9C\x9F" => "\xCE\xB4",
1575 "\xF0\x9D\x9C\xA0" => "\xCE\xB5",
1576 "\xF0\x9D\x9C\xA1" => "\xCE\xB6",
1577 "\xF0\x9D\x9C\xA2" => "\xCE\xB7",
1578 "\xF0\x9D\x9C\xA3" => "\xCE\xB8",
1579 "\xF0\x9D\x9C\xA4" => "\xCE\xB9",
1580 "\xF0\x9D\x9C\xA5" => "\xCE\xBA",
1581 "\xF0\x9D\x9C\xA6" => "\xCE\xBB",
1582 "\xF0\x9D\x9C\xA7" => "\xCE\xBC",
1583 "\xF0\x9D\x9C\xA8" => "\xCE\xBD",
1584 "\xF0\x9D\x9C\xA9" => "\xCE\xBE",
1585 "\xF0\x9D\x9C\xAA" => "\xCE\xBF",
1586 "\xF0\x9D\x9C\xAB" => "\xCF\x80",
1587 "\xF0\x9D\x9C\xAC" => "\xCF\x81",
1588 "\xF0\x9D\x9C\xAD" => "\xCE\xB8",
1589 "\xF0\x9D\x9C\xAE" => "\xCF\x83",
1590 "\xF0\x9D\x9C\xAF" => "\xCF\x84",
1591 "\xF0\x9D\x9C\xB0" => "\xCF\x85",
1592 "\xF0\x9D\x9C\xB1" => "\xCF\x86",
1593 "\xF0\x9D\x9C\xB2" => "\xCF\x87",
1594 "\xF0\x9D\x9C\xB3" => "\xCF\x88",
1595 "\xF0\x9D\x9C\xB4" => "\xCF\x89",
1596 "\xF0\x9D\x9D\x87" => "\xCF\x83",
1597 "\xF0\x9D\x9D\x96" => "\xCE\xB1",
1598 "\xF0\x9D\x9D\x97" => "\xCE\xB2",
1599 "\xF0\x9D\x9D\x98" => "\xCE\xB3",
1600 "\xF0\x9D\x9D\x99" => "\xCE\xB4",
1601 "\xF0\x9D\x9D\x9A" => "\xCE\xB5",
1602 "\xF0\x9D\x9D\x9B" => "\xCE\xB6",
1603 "\xF0\x9D\x9D\x9C" => "\xCE\xB7",
1604 "\xF0\x9D\x9D\x9D" => "\xCE\xB8",
1605 "\xF0\x9D\x9D\x9E" => "\xCE\xB9",
1606 "\xF0\x9D\x9D\x9F" => "\xCE\xBA",
1607 "\xF0\x9D\x9D\xA0" => "\xCE\xBB",
1608 "\xF0\x9D\x9D\xA1" => "\xCE\xBC",
1609 "\xF0\x9D\x9D\xA2" => "\xCE\xBD",
1610 "\xF0\x9D\x9D\xA3" => "\xCE\xBE",
1611 "\xF0\x9D\x9D\xA4" => "\xCE\xBF",
1612 "\xF0\x9D\x9D\xA5" => "\xCF\x80",
1613 "\xF0\x9D\x9D\xA6" => "\xCF\x81",
1614 "\xF0\x9D\x9D\xA7" => "\xCE\xB8",
1615 "\xF0\x9D\x9D\xA8" => "\xCF\x83",
1616 "\xF0\x9D\x9D\xA9" => "\xCF\x84",
1617 "\xF0\x9D\x9D\xAA" => "\xCF\x85",
1618 "\xF0\x9D\x9D\xAB" => "\xCF\x86",
1619 "\xF0\x9D\x9D\xAC" => "\xCF\x87",
1620 "\xF0\x9D\x9D\xAD" => "\xCF\x88",
1621 "\xF0\x9D\x9D\xAE" => "\xCF\x89",
1622 "\xF0\x9D\x9E\x81" => "\xCF\x83",
1623 "\xF0\x9D\x9E\x90" => "\xCE\xB1",
1624 "\xF0\x9D\x9E\x91" => "\xCE\xB2",
1625 "\xF0\x9D\x9E\x92" => "\xCE\xB3",
1626 "\xF0\x9D\x9E\x93" => "\xCE\xB4",
1627 "\xF0\x9D\x9E\x94" => "\xCE\xB5",
1628 "\xF0\x9D\x9E\x95" => "\xCE\xB6",
1629 "\xF0\x9D\x9E\x96" => "\xCE\xB7",
1630 "\xF0\x9D\x9E\x97" => "\xCE\xB8",
1631 "\xF0\x9D\x9E\x98" => "\xCE\xB9",
1632 "\xF0\x9D\x9E\x99" => "\xCE\xBA",
1633 "\xF0\x9D\x9E\x9A" => "\xCE\xBB",
1634 "\xF0\x9D\x9E\x9B" => "\xCE\xBC",
1635 "\xF0\x9D\x9E\x9C" => "\xCE\xBD",
1636 "\xF0\x9D\x9E\x9D" => "\xCE\xBE",
1637 "\xF0\x9D\x9E\x9E" => "\xCE\xBF",
1638 "\xF0\x9D\x9E\x9F" => "\xCF\x80",
1639 "\xF0\x9D\x9E\xA0" => "\xCF\x81",
1640 "\xF0\x9D\x9E\xA1" => "\xCE\xB8",
1641 "\xF0\x9D\x9E\xA2" => "\xCF\x83",
1642 "\xF0\x9D\x9E\xA3" => "\xCF\x84",
1643 "\xF0\x9D\x9E\xA4" => "\xCF\x85",
1644 "\xF0\x9D\x9E\xA5" => "\xCF\x86",
1645 "\xF0\x9D\x9E\xA6" => "\xCF\x87",
1646 "\xF0\x9D\x9E\xA7" => "\xCF\x88",
1647 "\xF0\x9D\x9E\xA8" => "\xCF\x89",
1648 "\xF0\x9D\x9E\xBB" => "\xCF\x83",
1649 "\xF0\x9D\x9F\x8A" => "\xCF\x9D",
1650 );
1651 global $phpbb_root_path, $phpEx;
1652
1653 // do the case fold
1654 $text = utf8_case_fold($text, $option);
1655
1656 if (!class_exists('utf_normalizer'))
1657 {
1658 global $phpbb_root_path, $phpEx;
1659 include($phpbb_root_path . 'includes/utf/utf_normalizer.' . $phpEx);
1660 }
1661
1662 // convert to NFKC
1663 utf_normalizer::nfkc($text);
1664
1665 // FC_NFKC_Closure, http://www.unicode.org/Public/5.0.0/ucd/DerivedNormalizationProps.txt
1666 $text = strtr($text, $fc_nfkc_closure);
1667
1668 return $text;
1669 }
1670
1671 /**
1672 * Assume the input is NFC:
1673 * Takes the input and does a "special" case fold. It does minor normalization as well.
1674 *
1675 * @param string $text text to be case folded
1676 * @param string $option determines how we will fold the cases
1677 * @return string case folded text
1678 */
1679 function utf8_case_fold_nfc($text, $option = 'full')
1680 {
1681 static $uniarray = array();
1682 static $ypogegrammeni = array(
1683 "\xCD\xBA" => "\x20\xCD\x85",
1684 "\xE1\xBE\x80" => "\xE1\xBC\x80\xCD\x85",
1685 "\xE1\xBE\x81" => "\xE1\xBC\x81\xCD\x85",
1686 "\xE1\xBE\x82" => "\xE1\xBC\x82\xCD\x85",
1687 "\xE1\xBE\x83" => "\xE1\xBC\x83\xCD\x85",
1688 "\xE1\xBE\x84" => "\xE1\xBC\x84\xCD\x85",
1689 "\xE1\xBE\x85" => "\xE1\xBC\x85\xCD\x85",
1690 "\xE1\xBE\x86" => "\xE1\xBC\x86\xCD\x85",
1691 "\xE1\xBE\x87" => "\xE1\xBC\x87\xCD\x85",
1692 "\xE1\xBE\x88" => "\xE1\xBC\x88\xCD\x85",
1693 "\xE1\xBE\x89" => "\xE1\xBC\x89\xCD\x85",
1694 "\xE1\xBE\x8A" => "\xE1\xBC\x8A\xCD\x85",
1695 "\xE1\xBE\x8B" => "\xE1\xBC\x8B\xCD\x85",
1696 "\xE1\xBE\x8C" => "\xE1\xBC\x8C\xCD\x85",
1697 "\xE1\xBE\x8D" => "\xE1\xBC\x8D\xCD\x85",
1698 "\xE1\xBE\x8E" => "\xE1\xBC\x8E\xCD\x85",
1699 "\xE1\xBE\x8F" => "\xE1\xBC\x8F\xCD\x85",
1700 "\xE1\xBE\x90" => "\xE1\xBC\xA0\xCD\x85",
1701 "\xE1\xBE\x91" => "\xE1\xBC\xA1\xCD\x85",
1702 "\xE1\xBE\x92" => "\xE1\xBC\xA2\xCD\x85",
1703 "\xE1\xBE\x93" => "\xE1\xBC\xA3\xCD\x85",
1704 "\xE1\xBE\x94" => "\xE1\xBC\xA4\xCD\x85",
1705 "\xE1\xBE\x95" => "\xE1\xBC\xA5\xCD\x85",
1706 "\xE1\xBE\x96" => "\xE1\xBC\xA6\xCD\x85",
1707 "\xE1\xBE\x97" => "\xE1\xBC\xA7\xCD\x85",
1708 "\xE1\xBE\x98" => "\xE1\xBC\xA8\xCD\x85",
1709 "\xE1\xBE\x99" => "\xE1\xBC\xA9\xCD\x85",
1710 "\xE1\xBE\x9A" => "\xE1\xBC\xAA\xCD\x85",
1711 "\xE1\xBE\x9B" => "\xE1\xBC\xAB\xCD\x85",
1712 "\xE1\xBE\x9C" => "\xE1\xBC\xAC\xCD\x85",
1713 "\xE1\xBE\x9D" => "\xE1\xBC\xAD\xCD\x85",
1714 "\xE1\xBE\x9E" => "\xE1\xBC\xAE\xCD\x85",
1715 "\xE1\xBE\x9F" => "\xE1\xBC\xAF\xCD\x85",
1716 "\xE1\xBE\xA0" => "\xE1\xBD\xA0\xCD\x85",
1717 "\xE1\xBE\xA1" => "\xE1\xBD\xA1\xCD\x85",
1718 "\xE1\xBE\xA2" => "\xE1\xBD\xA2\xCD\x85",
1719 "\xE1\xBE\xA3" => "\xE1\xBD\xA3\xCD\x85",
1720 "\xE1\xBE\xA4" => "\xE1\xBD\xA4\xCD\x85",
1721 "\xE1\xBE\xA5" => "\xE1\xBD\xA5\xCD\x85",
1722 "\xE1\xBE\xA6" => "\xE1\xBD\xA6\xCD\x85",
1723 "\xE1\xBE\xA7" => "\xE1\xBD\xA7\xCD\x85",
1724 "\xE1\xBE\xA8" => "\xE1\xBD\xA8\xCD\x85",
1725 "\xE1\xBE\xA9" => "\xE1\xBD\xA9\xCD\x85",
1726 "\xE1\xBE\xAA" => "\xE1\xBD\xAA\xCD\x85",
1727 "\xE1\xBE\xAB" => "\xE1\xBD\xAB\xCD\x85",
1728 "\xE1\xBE\xAC" => "\xE1\xBD\xAC\xCD\x85",
1729 "\xE1\xBE\xAD" => "\xE1\xBD\xAD\xCD\x85",
1730 "\xE1\xBE\xAE" => "\xE1\xBD\xAE\xCD\x85",
1731 "\xE1\xBE\xAF" => "\xE1\xBD\xAF\xCD\x85",
1732 "\xE1\xBE\xB2" => "\xE1\xBD\xB0\xCD\x85",
1733 "\xE1\xBE\xB3" => "\xCE\xB1\xCD\x85",
1734 "\xE1\xBE\xB4" => "\xCE\xAC\xCD\x85",
1735 "\xE1\xBE\xB7" => "\xE1\xBE\xB6\xCD\x85",
1736 "\xE1\xBE\xBC" => "\xCE\x91\xCD\x85",
1737 "\xE1\xBF\x82" => "\xE1\xBD\xB4\xCD\x85",
1738 "\xE1\xBF\x83" => "\xCE\xB7\xCD\x85",
1739 "\xE1\xBF\x84" => "\xCE\xAE\xCD\x85",
1740 "\xE1\xBF\x87" => "\xE1\xBF\x86\xCD\x85",
1741 "\xE1\xBF\x8C" => "\xCE\x97\xCD\x85",
1742 "\xE1\xBF\xB2" => "\xE1\xBD\xBC\xCD\x85",
1743 "\xE1\xBF\xB3" => "\xCF\x89\xCD\x85",
1744 "\xE1\xBF\xB4" => "\xCF\x8E\xCD\x85",
1745 "\xE1\xBF\xB7" => "\xE1\xBF\xB6\xCD\x85",
1746 "\xE1\xBF\xBC" => "\xCE\xA9\xCD\x85",
1747 );
1748 global $phpbb_root_path, $phpEx;
1749
1750 // perform a small trick, avoid further normalization on composed points that contain U+0345 in their decomposition
1751 $text = strtr($text, $ypogegrammeni);
1752
1753 // do the case fold
1754 $text = utf8_case_fold($text, $option);
1755
1756 return $text;
1757 }
1758
1759 /**
1760 * A wrapper function for the normalizer which takes care of including the class if required and modifies the passed strings
1761 * to be in NFC (Normalization Form Composition).
1762 *
1763 * @param mixed $strings a string or an array of strings to normalize
1764 * @return mixed the normalized content, preserving array keys if array given.
1765 */
1766 function utf8_normalize_nfc($strings)
1767 {
1768 if (empty($strings))
1769 {
1770 return $strings;
1771 }
1772
1773 if (!class_exists('utf_normalizer'))
1774 {
1775 global $phpbb_root_path, $phpEx;
1776 include($phpbb_root_path . 'includes/utf/utf_normalizer.' . $phpEx);
1777 }
1778
1779 if (!is_array($strings))
1780 {
1781 utf_normalizer::nfc($strings);
1782 }
1783 else if (is_array($strings))
1784 {
1785 foreach ($strings as $key => $string)
1786 {
1787 if (is_array($string))
1788 {
1789 foreach ($string as $_key => $_string)
1790 {
1791 utf_normalizer::nfc($strings[$key][$_key]);
1792 }
1793 }
1794 else
1795 {
1796 utf_normalizer::nfc($strings[$key]);
1797 }
1798 }
1799 }
1800
1801 return $strings;
1802 }
1803
1804 /**
1805 * This function is used to generate a "clean" version of a string.
1806 * Clean means that it is a case insensitive form (case folding) and that it is normalized (NFC).
1807 * Additionally a homographs of one character are transformed into one specific character (preferably ASCII
1808 * if it is an ASCII character).
1809 *
1810 * Please be aware that if you change something within this function or within
1811 * functions used here you need to rebuild/update the username_clean column in the users table. And all other
1812 * columns that store a clean string otherwise you will break this functionality.
1813 *
1814 * @param string $text An unclean string, mabye user input (has to be valid UTF-8!)
1815 * @return string Cleaned up version of the input string
1816 */
1817 function utf8_clean_string($text)
1818 {
1819 global $phpbb_root_path, $phpEx;
1820
1821 static $homographs = array();
1822 if (empty($homographs))
1823 {
1824 $homographs = include($phpbb_root_path . 'includes/utf/data/confusables.' . $phpEx);
1825 }
1826
1827 $text = utf8_case_fold_nfkc($text);
1828 $text = strtr($text, $homographs);
1829 // Other control characters
1830 $text = preg_replace('#(?:[\x00-\x1F\x7F]+|(?:\xC2[\x80-\x9F])+)#', '', $text);
1831
1832 // we need to reduce multiple spaces to a single one
1833 $text = preg_replace('# {2,}#', ' ', $text);
1834
1835 // we can use trim here as all the other space characters should have been turned
1836 // into normal ASCII spaces by now
1837 return trim($text);
1838 }
1839
1840 /**
1841 * A wrapper for htmlspecialchars($value, ENT_COMPAT, 'UTF-8')
1842 */
1843 function utf8_htmlspecialchars(&$value)
1844 {
1845 return htmlspecialchars($value, ENT_COMPAT, 'UTF-8');
1846 }
1847
1848 /**
1849 * Trying to convert returned system message to utf8
1850 *
1851 * PHP assumes such messages are ISO-8859-1 so we'll do that too
1852 * and if it breaks messages we'll blame it on them ;-)
1853 */
1854 function utf8_convert_message($message)
1855 {
1856 // First of all check if conversion is neded at all, as there is no point
1857 // in converting ASCII messages from ISO-8859-1 to UTF-8
1858 if (!preg_match('/[\x80-\xFF]/', $message))
1859 {
1860 return utf8_htmlspecialchars($message);
1861 }
1862
1863 // else we need to convert some part of the message
1864 return utf8_htmlspecialchars(utf8_recode($message, 'ISO-8859-1'));
1865 }
1866
1867 /**
1868 * UTF8-compatible wordwrap replacement
1869 *
1870 * @param string $string The input string
1871 * @param int $width The column width. Defaults to 75.
1872 * @param string $break The line is broken using the optional break parameter. Defaults to '\n'.
1873 * @param bool $cut If the cut is set to TRUE, the string is always wrapped at the specified width. So if you have a word that is larger than the given width, it is broken apart.
1874 *
1875 * @return string the given string wrapped at the specified column.
1876 *
1877 */
1878 function utf8_wordwrap($string, $width = 75, $break = "\n", $cut = false)
1879 {
1880 // We first need to explode on $break, not destroying existing (intended) breaks
1881 $lines = explode($break, $string);
1882 $new_lines = array(0 => '');
1883 $index = 0;
1884
1885 foreach ($lines as $line)
1886 {
1887 $words = explode(' ', $line);
1888
1889 for ($i = 0, $size = sizeof($words); $i < $size; $i++)
1890 {
1891 $word = $words[$i];
1892
1893 // If cut is true we need to cut the word if it is > width chars
1894 if ($cut && utf8_strlen($word) > $width)
1895 {
1896 $words[$i] = utf8_substr($word, $width);
1897 $word = utf8_substr($word, 0, $width);
1898 $i--;
1899 }
1900
1901 if (utf8_strlen($new_lines[$index] . $word) > $width)
1902 {
1903 $new_lines[$index] = substr($new_lines[$index], 0, -1);
1904 $index++;
1905 $new_lines[$index] = '';
1906 }
1907
1908 $new_lines[$index] .= $word . ' ';
1909 }
1910
1911 $new_lines[$index] = substr($new_lines[$index], 0, -1);
1912 $index++;
1913 $new_lines[$index] = '';
1914 }
1915
1916 unset($new_lines[$index]);
1917 return implode($break, $new_lines);
1918 }
1919
1920 ?>