| [ Index ] |
|
Code source de Drupal 5.3 |
1 <?php 2 // $Id: unicode.inc,v 1.23.2.2 2007/05/21 01:09:21 drumm Exp $ 3 4 /** 5 * Indicates an error during check for PHP unicode support. 6 */ 7 define('UNICODE_ERROR', -1); 8 9 /** 10 * Indicates that standard PHP (emulated) unicode support is being used. 11 */ 12 define('UNICODE_SINGLEBYTE', 0); 13 14 /** 15 * Indicates that full unicode support with the PHP mbstring extension is being 16 * used. 17 */ 18 define('UNICODE_MULTIBYTE', 1); 19 20 /** 21 * Wrapper around _unicode_check(). 22 */ 23 function unicode_check() { 24 list($GLOBALS['multibyte']) = _unicode_check(); 25 } 26 27 /** 28 * Perform checks about Unicode support in PHP, and set the right settings if 29 * needed. 30 * 31 * Because Drupal needs to be able to handle text in various encodings, we do 32 * not support mbstring function overloading. HTTP input/output conversion must 33 * be disabled for similar reasons. 34 * 35 * @param $errors 36 * Whether to report any fatal errors with form_set_error(). 37 */ 38 function _unicode_check() { 39 // Ensure translations don't break at install time 40 $t = get_t(); 41 42 // Set the standard C locale to ensure consistent, ASCII-only string handling. 43 setlocale(LC_CTYPE, 'C'); 44 45 // Check for outdated PCRE library 46 // Note: we check if U+E2 is in the range U+E0 - U+E1. This test returns TRUE on old PCRE versions. 47 if (preg_match('/[à -á]/u', 'â')) { 48 return array(UNICODE_ERROR, $t('The PCRE library in your PHP installation is outdated. This will cause problems when handling Unicode text. If you are running PHP 4.3.3 or higher, make sure you are using the PCRE library supplied by PHP. Please refer to the <a href="@url">PHP PCRE documentation</a> for more information.', array('@url' => 'http://www.php.net/pcre'))); 49 } 50 51 // Check for mbstring extension 52 if (!function_exists('mb_strlen')) { 53 return array(UNICODE_SINGLEBYTE, $t('Operations on Unicode strings are emulated on a best-effort basis. Install the <a href="@url">PHP mbstring extension</a> for improved Unicode support.', array('@url' => 'http://www.php.net/mbstring'))); 54 } 55 56 // Check mbstring configuration 57 if (ini_get('mbstring.func_overload') != 0) { 58 return array(UNICODE_ERROR, $t('Multibyte string function overloading in PHP is active and must be disabled. Check the php.ini <em>mbstring.func_overload</em> setting. Please refer to the <a href="@url">PHP mbstring documentation</a> for more information.', array('@url' => 'http://www.php.net/mbstring'))); 59 } 60 if (ini_get('mbstring.encoding_translation') != 0) { 61 return array(UNICODE_ERROR, $t('Multibyte string input conversion in PHP is active and must be disabled. Check the php.ini <em>mbstring.encoding_translation</em> setting. Please refer to the <a href="@url">PHP mbstring documentation</a> for more information.', array('@url' => 'http://www.php.net/mbstring'))); 62 } 63 if (ini_get('mbstring.http_input') != 'pass') { 64 return array(UNICODE_ERROR, $t('Multibyte string input conversion in PHP is active and must be disabled. Check the php.ini <em>mbstring.http_input</em> setting. Please refer to the <a href="@url">PHP mbstring documentation</a> for more information.', array('@url' => 'http://www.php.net/mbstring'))); 65 } 66 if (ini_get('mbstring.http_output') != 'pass') { 67 return array(UNICODE_ERROR, $t('Multibyte string output conversion in PHP is active and must be disabled. Check the php.ini <em>mbstring.http_output</em> setting. Please refer to the <a href="@url">PHP mbstring documentation</a> for more information.', array('@url' => 'http://www.php.net/mbstring'))); 68 } 69 70 // Set appropriate configuration 71 mb_internal_encoding('utf-8'); 72 mb_language('uni'); 73 return array(UNICODE_MULTIBYTE, ''); 74 } 75 76 /** 77 * Return Unicode library status and errors. 78 */ 79 function unicode_requirements() { 80 // Ensure translations don't break at install time 81 $t = get_t(); 82 83 $libraries = array( 84 UNICODE_SINGLEBYTE => $t('Standard PHP'), 85 UNICODE_MULTIBYTE => $t('PHP Mbstring Extension'), 86 UNICODE_ERROR => $t('Error'), 87 ); 88 $severities = array( 89 UNICODE_SINGLEBYTE => REQUIREMENT_WARNING, 90 UNICODE_MULTIBYTE => REQUIREMENT_OK, 91 UNICODE_ERROR => REQUIREMENT_ERROR, 92 ); 93 list($library, $description) = _unicode_check(); 94 95 $requirements['unicode'] = array( 96 'title' => $t('Unicode library'), 97 'value' => $libraries[$library], 98 ); 99 if ($description) { 100 $requirements['unicode']['description'] = $description; 101 } 102 103 $requirements['unicode']['severity'] = $severities[$library]; 104 105 return $requirements; 106 } 107 108 /** 109 * Prepare a new XML parser. 110 * 111 * This is a wrapper around xml_parser_create() which extracts the encoding from 112 * the XML data first and sets the output encoding to UTF-8. This function should 113 * be used instead of xml_parser_create(), because PHP 4's XML parser doesn't 114 * check the input encoding itself. "Starting from PHP 5, the input encoding is 115 * automatically detected, so that the encoding parameter specifies only the 116 * output encoding." 117 * 118 * This is also where unsupported encodings will be converted. Callers should 119 * take this into account: $data might have been changed after the call. 120 * 121 * @param &$data 122 * The XML data which will be parsed later. 123 * @return 124 * An XML parser object. 125 */ 126 function drupal_xml_parser_create(&$data) { 127 // Default XML encoding is UTF-8 128 $encoding = 'utf-8'; 129 $bom = FALSE; 130 131 // Check for UTF-8 byte order mark (PHP5's XML parser doesn't handle it). 132 if (!strncmp($data, "\xEF\xBB\xBF", 3)) { 133 $bom = TRUE; 134 $data = substr($data, 3); 135 } 136 137 // Check for an encoding declaration in the XML prolog if no BOM was found. 138 if (!$bom && ereg('^<\?xml[^>]+encoding="([^"]+)"', $data, $match)) { 139 $encoding = $match[1]; 140 } 141 142 // Unsupported encodings are converted here into UTF-8. 143 $php_supported = array('utf-8', 'iso-8859-1', 'us-ascii'); 144 if (!in_array(strtolower($encoding), $php_supported)) { 145 $out = drupal_convert_to_utf8($data, $encoding); 146 if ($out !== FALSE) { 147 $encoding = 'utf-8'; 148 $data = ereg_replace('^(<\?xml[^>]+encoding)="([^"]+)"', '\\1="utf-8"', $out); 149 } 150 else { 151 watchdog('php', t("Could not convert XML encoding %s to UTF-8.", array('%s' => $encoding)), WATCHDOG_WARNING); 152 return 0; 153 } 154 } 155 156 $xml_parser = xml_parser_create($encoding); 157 xml_parser_set_option($xml_parser, XML_OPTION_TARGET_ENCODING, 'utf-8'); 158 return $xml_parser; 159 } 160 161 /** 162 * Convert data to UTF-8 163 * 164 * Requires the iconv, GNU recode or mbstring PHP extension. 165 * 166 * @param $data 167 * The data to be converted. 168 * @param $encoding 169 * The encoding that the data is in 170 * @return 171 * Converted data or FALSE. 172 */ 173 function drupal_convert_to_utf8($data, $encoding) { 174 if (function_exists('iconv')) { 175 $out = @iconv($encoding, 'utf-8', $data); 176 } 177 else if (function_exists('mb_convert_encoding')) { 178 $out = @mb_convert_encoding($data, 'utf-8', $encoding); 179 } 180 else if (function_exists('recode_string')) { 181 $out = @recode_string($encoding .'..utf-8', $data); 182 } 183 else { 184 watchdog('php', t("Unsupported encoding %s. Please install iconv, GNU recode or mbstring for PHP.", array('%s' => $encoding)), WATCHDOG_ERROR); 185 return FALSE; 186 } 187 188 return $out; 189 } 190 191 /** 192 * Truncate a UTF-8-encoded string safely to a number of bytes. 193 * 194 * If the end position is in the middle of a UTF-8 sequence, it scans backwards 195 * until the beginning of the byte sequence. 196 * 197 * Use this function whenever you want to chop off a string at an unsure 198 * location. On the other hand, if you're sure that you're splitting on a 199 * character boundary (e.g. after using strpos() or similar), you can safely use 200 * substr() instead. 201 * 202 * @param $string 203 * The string to truncate. 204 * @param $len 205 * An upper limit on the returned string length. 206 * @param $wordsafe 207 * Flag to truncate at nearest space. Defaults to FALSE. 208 * @return 209 * The truncated string. 210 */ 211 function truncate_utf8($string, $len, $wordsafe = FALSE, $dots = FALSE) { 212 $slen = strlen($string); 213 if ($slen <= $len) { 214 return $string; 215 } 216 if ($wordsafe) { 217 $end = $len; 218 while (($string[--$len] != ' ') && ($len > 0)) {}; 219 if ($len == 0) { 220 $len = $end; 221 } 222 } 223 if ((ord($string[$len]) < 0x80) || (ord($string[$len]) >= 0xC0)) { 224 return substr($string, 0, $len) . ($dots ? ' ...' : ''); 225 } 226 while (--$len >= 0 && ord($string[$len]) >= 0x80 && ord($string[$len]) < 0xC0) {}; 227 return substr($string, 0, $len) . ($dots ? ' ...' : ''); 228 } 229 230 /** 231 * Encodes MIME/HTTP header values that contain non-ASCII, UTF-8 encoded 232 * characters. 233 * 234 * For example, mime_header_encode('tést.txt') returns "=?UTF-8?B?dMOpc3QudHh0?=". 235 * 236 * See http://www.rfc-editor.org/rfc/rfc2047.txt for more information. 237 * 238 * Notes: 239 * - Only encode strings that contain non-ASCII characters. 240 * - We progressively cut-off a chunk with truncate_utf8(). This is to ensure 241 * each chunk starts and ends on a character boundary. 242 * - Using \n as the chunk separator may cause problems on some systems and may 243 * have to be changed to \r\n or \r. 244 */ 245 function mime_header_encode($string) { 246 if (preg_match('/[^\x20-\x7E]/', $string)) { 247 $chunk_size = 47; // floor((75 - strlen("=?UTF-8?B??=")) * 0.75); 248 $len = strlen($string); 249 $output = ''; 250 while ($len > 0) { 251 $chunk = truncate_utf8($string, $chunk_size); 252 $output .= ' =?UTF-8?B?'. base64_encode($chunk) ."?=\n"; 253 $c = strlen($chunk); 254 $string = substr($string, $c); 255 $len -= $c; 256 } 257 return trim($output); 258 } 259 return $string; 260 } 261 262 /** 263 * Complement to mime_header_encode 264 */ 265 function mime_header_decode($header) { 266 // First step: encoded chunks followed by other encoded chunks (need to collapse whitespace) 267 $header = preg_replace_callback('/=\?([^?]+)\?(Q|B)\?([^?]+|\?(?!=))\?=\s+(?==\?)/', '_mime_header_decode', $header); 268 // Second step: remaining chunks (do not collapse whitespace) 269 return preg_replace_callback('/=\?([^?]+)\?(Q|B)\?([^?]+|\?(?!=))\?=/', '_mime_header_decode', $header); 270 } 271 272 /** 273 * Helper function to mime_header_decode 274 */ 275 function _mime_header_decode($matches) { 276 // Regexp groups: 277 // 1: Character set name 278 // 2: Escaping method (Q or B) 279 // 3: Encoded data 280 $data = ($matches[2] == 'B') ? base64_decode($matches[3]) : str_replace('_', ' ', quoted_printable_decode($matches[3])); 281 if (strtolower($matches[1]) != 'utf-8') { 282 $data = drupal_convert_to_utf8($data, $matches[1]); 283 } 284 return $data; 285 } 286 287 /** 288 * Decode all HTML entities (including numerical ones) to regular UTF-8 bytes. 289 * Double-escaped entities will only be decoded once ("&lt;" becomes "<", not "<"). 290 * 291 * @param $text 292 * The text to decode entities in. 293 * @param $exclude 294 * An array of characters which should not be decoded. For example, 295 * array('<', '&', '"'). This affects both named and numerical entities. 296 */ 297 function decode_entities($text, $exclude = array()) { 298 static $table; 299 // We store named entities in a table for quick processing. 300 if (!isset($table)) { 301 // Get all named HTML entities. 302 $table = array_flip(get_html_translation_table(HTML_ENTITIES)); 303 // PHP gives us ISO-8859-1 data, we need UTF-8. 304 $table = array_map('utf8_encode', $table); 305 // Add apostrophe (XML) 306 $table['''] = "'"; 307 } 308 $newtable = array_diff($table, $exclude); 309 310 // Use a regexp to select all entities in one pass, to avoid decoding double-escaped entities twice. 311 return preg_replace('/&(#x?)?([A-Za-z0-9]+);/e', '_decode_entities("$1", "$2", "$0", $newtable, $exclude)', $text); 312 } 313 314 /** 315 * Helper function for decode_entities 316 */ 317 function _decode_entities($prefix, $codepoint, $original, &$table, &$exclude) { 318 // Named entity 319 if (!$prefix) { 320 if (isset($table[$original])) { 321 return $table[$original]; 322 } 323 else { 324 return $original; 325 } 326 } 327 // Hexadecimal numerical entity 328 if ($prefix == '#x') { 329 $codepoint = base_convert($codepoint, 16, 10); 330 } 331 // Decimal numerical entity (strip leading zeros to avoid PHP octal notation) 332 else { 333 $codepoint = preg_replace('/^0+/', '', $codepoint); 334 } 335 // Encode codepoint as UTF-8 bytes 336 if ($codepoint < 0x80) { 337 $str = chr($codepoint); 338 } 339 else if ($codepoint < 0x800) { 340 $str = chr(0xC0 | ($codepoint >> 6)) 341 . chr(0x80 | ($codepoint & 0x3F)); 342 } 343 else if ($codepoint < 0x10000) { 344 $str = chr(0xE0 | ( $codepoint >> 12)) 345 . chr(0x80 | (($codepoint >> 6) & 0x3F)) 346 . chr(0x80 | ( $codepoint & 0x3F)); 347 } 348 else if ($codepoint < 0x200000) { 349 $str = chr(0xF0 | ( $codepoint >> 18)) 350 . chr(0x80 | (($codepoint >> 12) & 0x3F)) 351 . chr(0x80 | (($codepoint >> 6) & 0x3F)) 352 . chr(0x80 | ( $codepoint & 0x3F)); 353 } 354 // Check for excluded characters 355 if (in_array($str, $exclude)) { 356 return $original; 357 } 358 else { 359 return $str; 360 } 361 } 362 363 /** 364 * Count the amount of characters in a UTF-8 string. This is less than or 365 * equal to the byte count. 366 */ 367 function drupal_strlen($text) { 368 global $multibyte; 369 if ($multibyte == UNICODE_MULTIBYTE) { 370 return mb_strlen($text); 371 } 372 else { 373 // Do not count UTF-8 continuation bytes. 374 return strlen(preg_replace("/[\x80-\xBF]/", '', $text)); 375 } 376 } 377 378 /** 379 * Uppercase a UTF-8 string. 380 */ 381 function drupal_strtoupper($text) { 382 global $multibyte; 383 if ($multibyte == UNICODE_MULTIBYTE) { 384 return mb_strtoupper($text); 385 } 386 else { 387 // Use C-locale for ASCII-only uppercase 388 $text = strtoupper($text); 389 // Case flip Latin-1 accented letters 390 $text = preg_replace_callback('/\xC3[\xA0-\xB6\xB8-\xBE]/', '_unicode_caseflip', $text); 391 return $text; 392 } 393 } 394 395 /** 396 * Lowercase a UTF-8 string. 397 */ 398 function drupal_strtolower($text) { 399 global $multibyte; 400 if ($multibyte == UNICODE_MULTIBYTE) { 401 return mb_strtolower($text); 402 } 403 else { 404 // Use C-locale for ASCII-only lowercase 405 $text = strtolower($text); 406 // Case flip Latin-1 accented letters 407 $text = preg_replace_callback('/\xC3[\x80-\x96\x98-\x9E]/', '_unicode_caseflip', $text); 408 return $text; 409 } 410 } 411 412 /** 413 * Helper function for case conversion of Latin-1. 414 * Used for flipping U+C0-U+DE to U+E0-U+FD and back. 415 */ 416 function _unicode_caseflip($matches) { 417 return $matches[0][0] . chr(ord($matches[0][1]) ^ 32); 418 } 419 420 /** 421 * Capitalize the first letter of a UTF-8 string. 422 */ 423 function drupal_ucfirst($text) { 424 // Note: no mbstring equivalent! 425 return drupal_strtoupper(drupal_substr($text, 0, 1)) . drupal_substr($text, 1); 426 } 427 428 /** 429 * Cut off a piece of a string based on character indices and counts. Follows 430 * the same behaviour as PHP's own substr() function. 431 * 432 * Note that for cutting off a string at a known character/substring 433 * location, the usage of PHP's normal strpos/substr is safe and 434 * much faster. 435 */ 436 function drupal_substr($text, $start, $length = NULL) { 437 global $multibyte; 438 if ($multibyte == UNICODE_MULTIBYTE) { 439 return $length === NULL ? mb_substr($text, $start) : mb_substr($text, $start, $length); 440 } 441 else { 442 $strlen = strlen($text); 443 // Find the starting byte offset 444 $bytes = 0; 445 if ($start > 0) { 446 // Count all the continuation bytes from the start until we have found 447 // $start characters 448 $bytes = -1; $chars = -1; 449 while ($bytes < $strlen && $chars < $start) { 450 $bytes++; 451 $c = ord($text[$bytes]); 452 if ($c < 0x80 || $c >= 0xC0) { 453 $chars++; 454 } 455 } 456 } 457 else if ($start < 0) { 458 // Count all the continuation bytes from the end until we have found 459 // abs($start) characters 460 $start = abs($start); 461 $bytes = $strlen; $chars = 0; 462 while ($bytes > 0 && $chars < $start) { 463 $bytes--; 464 $c = ord($text[$bytes]); 465 if ($c < 0x80 || $c >= 0xC0) { 466 $chars++; 467 } 468 } 469 } 470 $istart = $bytes; 471 472 // Find the ending byte offset 473 if ($length === NULL) { 474 $bytes = $strlen - 1; 475 } 476 else if ($length > 0) { 477 // Count all the continuation bytes from the starting index until we have 478 // found $length + 1 characters. Then backtrack one byte. 479 $bytes = $istart; $chars = 0; 480 while ($bytes < $strlen && $chars < $length) { 481 $bytes++; 482 $c = ord($text[$bytes]); 483 if ($c < 0x80 || $c >= 0xC0) { 484 $chars++; 485 } 486 } 487 $bytes--; 488 } 489 else if ($length < 0) { 490 // Count all the continuation bytes from the end until we have found 491 // abs($length) characters 492 $length = abs($length); 493 $bytes = $strlen - 1; $chars = 0; 494 while ($bytes >= 0 && $chars < $length) { 495 $c = ord($text[$bytes]); 496 if ($c < 0x80 || $c >= 0xC0) { 497 $chars++; 498 } 499 $bytes--; 500 } 501 } 502 $iend = $bytes; 503 504 return substr($text, $istart, max(0, $iend - $istart + 1)); 505 } 506 } 507 508
titre
Description
Corps
titre
Description
Corps
titre
Description
Corps
titre
Corps
| Généré le : Fri Nov 30 16:20:15 2007 | par Balluche grâce à PHPXref 0.7 |
|
|