@@ -22,7 +22,7 @@ discard block |
||
| 22 | 22 | |
| 23 | 23 | // securité |
| 24 | 24 | if (!defined('_ECRIRE_INC_VERSION')) { |
| 25 | - return; |
|
| 25 | + return; |
|
| 26 | 26 | } |
| 27 | 27 | |
| 28 | 28 | // se faciliter la lecture du charset |
@@ -42,45 +42,45 @@ discard block |
||
| 42 | 42 | * - false si le charset n'est pas décrit dans le répertoire charsets/ |
| 43 | 43 | **/ |
| 44 | 44 | function load_charset($charset = 'AUTO') { |
| 45 | - if ($charset == 'AUTO') { |
|
| 46 | - $charset = $GLOBALS['meta']['charset']; |
|
| 47 | - } |
|
| 48 | - $charset = trim(strtolower($charset)); |
|
| 49 | - if (isset($GLOBALS['CHARSET'][$charset])) { |
|
| 50 | - return $charset; |
|
| 51 | - } |
|
| 52 | - |
|
| 53 | - if ($charset == 'utf-8') { |
|
| 54 | - $GLOBALS['CHARSET'][$charset] = array(); |
|
| 55 | - |
|
| 56 | - return $charset; |
|
| 57 | - } |
|
| 58 | - |
|
| 59 | - // Quelques synonymes |
|
| 60 | - if ($charset == '') { |
|
| 61 | - $charset = 'iso-8859-1'; |
|
| 62 | - } else { |
|
| 63 | - if ($charset == 'windows-1250') { |
|
| 64 | - $charset = 'cp1250'; |
|
| 65 | - } else { |
|
| 66 | - if ($charset == 'windows-1251') { |
|
| 67 | - $charset = 'cp1251'; |
|
| 68 | - } else { |
|
| 69 | - if ($charset == 'windows-1256') { |
|
| 70 | - $charset = 'cp1256'; |
|
| 71 | - } |
|
| 72 | - } |
|
| 73 | - } |
|
| 74 | - } |
|
| 75 | - |
|
| 76 | - if (find_in_path($charset . '.php', 'charsets/', true)) { |
|
| 77 | - return $charset; |
|
| 78 | - } else { |
|
| 79 | - spip_log("Erreur: pas de fichier de conversion 'charsets/$charset'"); |
|
| 80 | - $GLOBALS['CHARSET'][$charset] = array(); |
|
| 81 | - |
|
| 82 | - return false; |
|
| 83 | - } |
|
| 45 | + if ($charset == 'AUTO') { |
|
| 46 | + $charset = $GLOBALS['meta']['charset']; |
|
| 47 | + } |
|
| 48 | + $charset = trim(strtolower($charset)); |
|
| 49 | + if (isset($GLOBALS['CHARSET'][$charset])) { |
|
| 50 | + return $charset; |
|
| 51 | + } |
|
| 52 | + |
|
| 53 | + if ($charset == 'utf-8') { |
|
| 54 | + $GLOBALS['CHARSET'][$charset] = array(); |
|
| 55 | + |
|
| 56 | + return $charset; |
|
| 57 | + } |
|
| 58 | + |
|
| 59 | + // Quelques synonymes |
|
| 60 | + if ($charset == '') { |
|
| 61 | + $charset = 'iso-8859-1'; |
|
| 62 | + } else { |
|
| 63 | + if ($charset == 'windows-1250') { |
|
| 64 | + $charset = 'cp1250'; |
|
| 65 | + } else { |
|
| 66 | + if ($charset == 'windows-1251') { |
|
| 67 | + $charset = 'cp1251'; |
|
| 68 | + } else { |
|
| 69 | + if ($charset == 'windows-1256') { |
|
| 70 | + $charset = 'cp1256'; |
|
| 71 | + } |
|
| 72 | + } |
|
| 73 | + } |
|
| 74 | + } |
|
| 75 | + |
|
| 76 | + if (find_in_path($charset . '.php', 'charsets/', true)) { |
|
| 77 | + return $charset; |
|
| 78 | + } else { |
|
| 79 | + spip_log("Erreur: pas de fichier de conversion 'charsets/$charset'"); |
|
| 80 | + $GLOBALS['CHARSET'][$charset] = array(); |
|
| 81 | + |
|
| 82 | + return false; |
|
| 83 | + } |
|
| 84 | 84 | } |
| 85 | 85 | |
| 86 | 86 | |
@@ -91,30 +91,30 @@ discard block |
||
| 91 | 91 | * true si toutes les fonctions mb nécessaires sont présentes |
| 92 | 92 | **/ |
| 93 | 93 | function init_mb_string() { |
| 94 | - static $mb; |
|
| 95 | - |
|
| 96 | - // verifier que tout est present (fonctions mb_string pour php >= 4.0.6) |
|
| 97 | - // et que le charset interne est connu de mb_string |
|
| 98 | - if (!$mb) { |
|
| 99 | - if (function_exists('mb_internal_encoding') |
|
| 100 | - and function_exists('mb_detect_order') |
|
| 101 | - and function_exists('mb_substr') |
|
| 102 | - and function_exists('mb_strlen') |
|
| 103 | - and function_exists('mb_strtolower') |
|
| 104 | - and function_exists('mb_strtoupper') |
|
| 105 | - and function_exists('mb_encode_mimeheader') |
|
| 106 | - and function_exists('mb_encode_numericentity') |
|
| 107 | - and function_exists('mb_decode_numericentity') |
|
| 108 | - and mb_detect_order(lire_config('charset', _DEFAULT_CHARSET)) |
|
| 109 | - ) { |
|
| 110 | - mb_internal_encoding('utf-8'); |
|
| 111 | - $mb = 1; |
|
| 112 | - } else { |
|
| 113 | - $mb = -1; |
|
| 114 | - } |
|
| 115 | - } |
|
| 116 | - |
|
| 117 | - return ($mb == 1); |
|
| 94 | + static $mb; |
|
| 95 | + |
|
| 96 | + // verifier que tout est present (fonctions mb_string pour php >= 4.0.6) |
|
| 97 | + // et que le charset interne est connu de mb_string |
|
| 98 | + if (!$mb) { |
|
| 99 | + if (function_exists('mb_internal_encoding') |
|
| 100 | + and function_exists('mb_detect_order') |
|
| 101 | + and function_exists('mb_substr') |
|
| 102 | + and function_exists('mb_strlen') |
|
| 103 | + and function_exists('mb_strtolower') |
|
| 104 | + and function_exists('mb_strtoupper') |
|
| 105 | + and function_exists('mb_encode_mimeheader') |
|
| 106 | + and function_exists('mb_encode_numericentity') |
|
| 107 | + and function_exists('mb_decode_numericentity') |
|
| 108 | + and mb_detect_order(lire_config('charset', _DEFAULT_CHARSET)) |
|
| 109 | + ) { |
|
| 110 | + mb_internal_encoding('utf-8'); |
|
| 111 | + $mb = 1; |
|
| 112 | + } else { |
|
| 113 | + $mb = -1; |
|
| 114 | + } |
|
| 115 | + } |
|
| 116 | + |
|
| 117 | + return ($mb == 1); |
|
| 118 | 118 | } |
| 119 | 119 | |
| 120 | 120 | /** |
@@ -129,21 +129,21 @@ discard block |
||
| 129 | 129 | * true si iconv fonctionne correctement |
| 130 | 130 | **/ |
| 131 | 131 | function test_iconv() { |
| 132 | - static $iconv_ok; |
|
| 133 | - |
|
| 134 | - if (!$iconv_ok) { |
|
| 135 | - if (!function_exists('iconv')) { |
|
| 136 | - $iconv_ok = -1; |
|
| 137 | - } else { |
|
| 138 | - if (utf_32_to_unicode(@iconv('utf-8', 'utf-32', 'chaine de test')) == 'chaine de test') { |
|
| 139 | - $iconv_ok = 1; |
|
| 140 | - } else { |
|
| 141 | - $iconv_ok = -1; |
|
| 142 | - } |
|
| 143 | - } |
|
| 144 | - } |
|
| 145 | - |
|
| 146 | - return ($iconv_ok == 1); |
|
| 132 | + static $iconv_ok; |
|
| 133 | + |
|
| 134 | + if (!$iconv_ok) { |
|
| 135 | + if (!function_exists('iconv')) { |
|
| 136 | + $iconv_ok = -1; |
|
| 137 | + } else { |
|
| 138 | + if (utf_32_to_unicode(@iconv('utf-8', 'utf-32', 'chaine de test')) == 'chaine de test') { |
|
| 139 | + $iconv_ok = 1; |
|
| 140 | + } else { |
|
| 141 | + $iconv_ok = -1; |
|
| 142 | + } |
|
| 143 | + } |
|
| 144 | + } |
|
| 145 | + |
|
| 146 | + return ($iconv_ok == 1); |
|
| 147 | 147 | } |
| 148 | 148 | |
| 149 | 149 | |
@@ -156,18 +156,18 @@ discard block |
||
| 156 | 156 | * true si PCRE supporte l'UTF-8 correctement |
| 157 | 157 | **/ |
| 158 | 158 | function test_pcre_unicode() { |
| 159 | - static $pcre_ok = 0; |
|
| 160 | - |
|
| 161 | - if (!$pcre_ok) { |
|
| 162 | - $s = " " . chr(195) . chr(169) . "t" . chr(195) . chr(169) . " "; |
|
| 163 | - if (preg_match(',\W...\W,u', $s)) { |
|
| 164 | - $pcre_ok = 1; |
|
| 165 | - } else { |
|
| 166 | - $pcre_ok = -1; |
|
| 167 | - } |
|
| 168 | - } |
|
| 169 | - |
|
| 170 | - return $pcre_ok == 1; |
|
| 159 | + static $pcre_ok = 0; |
|
| 160 | + |
|
| 161 | + if (!$pcre_ok) { |
|
| 162 | + $s = " " . chr(195) . chr(169) . "t" . chr(195) . chr(169) . " "; |
|
| 163 | + if (preg_match(',\W...\W,u', $s)) { |
|
| 164 | + $pcre_ok = 1; |
|
| 165 | + } else { |
|
| 166 | + $pcre_ok = -1; |
|
| 167 | + } |
|
| 168 | + } |
|
| 169 | + |
|
| 170 | + return $pcre_ok == 1; |
|
| 171 | 171 | } |
| 172 | 172 | |
| 173 | 173 | /** |
@@ -183,22 +183,22 @@ discard block |
||
| 183 | 183 | * Plage de caractères |
| 184 | 184 | **/ |
| 185 | 185 | function pcre_lettres_unicode() { |
| 186 | - static $plage_unicode; |
|
| 187 | - |
|
| 188 | - if (!$plage_unicode) { |
|
| 189 | - if (test_pcre_unicode()) { |
|
| 190 | - // cf. http://www.unicode.org/charts/ |
|
| 191 | - $plage_unicode = '\w' // iso-latin |
|
| 192 | - . '\x{100}-\x{24f}' // europeen etendu |
|
| 193 | - . '\x{300}-\x{1cff}' // des tas de trucs |
|
| 194 | - ; |
|
| 195 | - } else { |
|
| 196 | - // fallback a trois sous |
|
| 197 | - $plage_unicode = '\w'; |
|
| 198 | - } |
|
| 199 | - } |
|
| 200 | - |
|
| 201 | - return $plage_unicode; |
|
| 186 | + static $plage_unicode; |
|
| 187 | + |
|
| 188 | + if (!$plage_unicode) { |
|
| 189 | + if (test_pcre_unicode()) { |
|
| 190 | + // cf. http://www.unicode.org/charts/ |
|
| 191 | + $plage_unicode = '\w' // iso-latin |
|
| 192 | + . '\x{100}-\x{24f}' // europeen etendu |
|
| 193 | + . '\x{300}-\x{1cff}' // des tas de trucs |
|
| 194 | + ; |
|
| 195 | + } else { |
|
| 196 | + // fallback a trois sous |
|
| 197 | + $plage_unicode = '\w'; |
|
| 198 | + } |
|
| 199 | + } |
|
| 200 | + |
|
| 201 | + return $plage_unicode; |
|
| 202 | 202 | } |
| 203 | 203 | |
| 204 | 204 | |
@@ -216,7 +216,7 @@ discard block |
||
| 216 | 216 | * Plage de caractères |
| 217 | 217 | **/ |
| 218 | 218 | function plage_punct_unicode() { |
| 219 | - return '\xE2(\x80[\x80-\xBF]|\x81[\x80-\xAF])'; |
|
| 219 | + return '\xE2(\x80[\x80-\xBF]|\x81[\x80-\xAF])'; |
|
| 220 | 220 | } |
| 221 | 221 | |
| 222 | 222 | /** |
@@ -236,72 +236,72 @@ discard block |
||
| 236 | 236 | * Texte corrigé |
| 237 | 237 | **/ |
| 238 | 238 | function corriger_caracteres_windows($texte, $charset = 'AUTO', $charset_cible = 'unicode') { |
| 239 | - static $trans; |
|
| 240 | - |
|
| 241 | - if (is_array($texte)) { |
|
| 242 | - return array_map('corriger_caracteres_windows', $texte); |
|
| 243 | - } |
|
| 244 | - |
|
| 245 | - if ($charset == 'AUTO') { |
|
| 246 | - $charset = lire_config('charset', _DEFAULT_CHARSET); |
|
| 247 | - } |
|
| 248 | - if ($charset == 'utf-8') { |
|
| 249 | - $p = chr(194); |
|
| 250 | - if (strpos($texte, $p) == false) { |
|
| 251 | - return $texte; |
|
| 252 | - } |
|
| 253 | - } else { |
|
| 254 | - if ($charset == 'iso-8859-1') { |
|
| 255 | - $p = ''; |
|
| 256 | - } else { |
|
| 257 | - return $texte; |
|
| 258 | - } |
|
| 259 | - } |
|
| 260 | - |
|
| 261 | - if (!isset($trans[$charset][$charset_cible])) { |
|
| 262 | - $trans[$charset][$charset_cible] = array( |
|
| 263 | - $p . chr(128) => "€", |
|
| 264 | - $p . chr(129) => ' ', # pas affecte |
|
| 265 | - $p . chr(130) => "‚", |
|
| 266 | - $p . chr(131) => "ƒ", |
|
| 267 | - $p . chr(132) => "„", |
|
| 268 | - $p . chr(133) => "…", |
|
| 269 | - $p . chr(134) => "†", |
|
| 270 | - $p . chr(135) => "‡", |
|
| 271 | - $p . chr(136) => "ˆ", |
|
| 272 | - $p . chr(137) => "‰", |
|
| 273 | - $p . chr(138) => "Š", |
|
| 274 | - $p . chr(139) => "‹", |
|
| 275 | - $p . chr(140) => "Œ", |
|
| 276 | - $p . chr(141) => ' ', # pas affecte |
|
| 277 | - $p . chr(142) => "Ž", |
|
| 278 | - $p . chr(143) => ' ', # pas affecte |
|
| 279 | - $p . chr(144) => ' ', # pas affecte |
|
| 280 | - $p . chr(145) => "‘", |
|
| 281 | - $p . chr(146) => "’", |
|
| 282 | - $p . chr(147) => "“", |
|
| 283 | - $p . chr(148) => "”", |
|
| 284 | - $p . chr(149) => "•", |
|
| 285 | - $p . chr(150) => "–", |
|
| 286 | - $p . chr(151) => "—", |
|
| 287 | - $p . chr(152) => "˜", |
|
| 288 | - $p . chr(153) => "™", |
|
| 289 | - $p . chr(154) => "š", |
|
| 290 | - $p . chr(155) => "›", |
|
| 291 | - $p . chr(156) => "œ", |
|
| 292 | - $p . chr(157) => ' ', # pas affecte |
|
| 293 | - $p . chr(158) => "ž", |
|
| 294 | - $p . chr(159) => "Ÿ", |
|
| 295 | - ); |
|
| 296 | - if ($charset_cible != 'unicode') { |
|
| 297 | - foreach ($trans[$charset][$charset_cible] as $k => $c) { |
|
| 298 | - $trans[$charset][$charset_cible][$k] = unicode2charset($c, $charset_cible); |
|
| 299 | - } |
|
| 300 | - } |
|
| 301 | - } |
|
| 302 | - |
|
| 303 | - return @str_replace(array_keys($trans[$charset][$charset_cible]), |
|
| 304 | - array_values($trans[$charset][$charset_cible]), $texte); |
|
| 239 | + static $trans; |
|
| 240 | + |
|
| 241 | + if (is_array($texte)) { |
|
| 242 | + return array_map('corriger_caracteres_windows', $texte); |
|
| 243 | + } |
|
| 244 | + |
|
| 245 | + if ($charset == 'AUTO') { |
|
| 246 | + $charset = lire_config('charset', _DEFAULT_CHARSET); |
|
| 247 | + } |
|
| 248 | + if ($charset == 'utf-8') { |
|
| 249 | + $p = chr(194); |
|
| 250 | + if (strpos($texte, $p) == false) { |
|
| 251 | + return $texte; |
|
| 252 | + } |
|
| 253 | + } else { |
|
| 254 | + if ($charset == 'iso-8859-1') { |
|
| 255 | + $p = ''; |
|
| 256 | + } else { |
|
| 257 | + return $texte; |
|
| 258 | + } |
|
| 259 | + } |
|
| 260 | + |
|
| 261 | + if (!isset($trans[$charset][$charset_cible])) { |
|
| 262 | + $trans[$charset][$charset_cible] = array( |
|
| 263 | + $p . chr(128) => "€", |
|
| 264 | + $p . chr(129) => ' ', # pas affecte |
|
| 265 | + $p . chr(130) => "‚", |
|
| 266 | + $p . chr(131) => "ƒ", |
|
| 267 | + $p . chr(132) => "„", |
|
| 268 | + $p . chr(133) => "…", |
|
| 269 | + $p . chr(134) => "†", |
|
| 270 | + $p . chr(135) => "‡", |
|
| 271 | + $p . chr(136) => "ˆ", |
|
| 272 | + $p . chr(137) => "‰", |
|
| 273 | + $p . chr(138) => "Š", |
|
| 274 | + $p . chr(139) => "‹", |
|
| 275 | + $p . chr(140) => "Œ", |
|
| 276 | + $p . chr(141) => ' ', # pas affecte |
|
| 277 | + $p . chr(142) => "Ž", |
|
| 278 | + $p . chr(143) => ' ', # pas affecte |
|
| 279 | + $p . chr(144) => ' ', # pas affecte |
|
| 280 | + $p . chr(145) => "‘", |
|
| 281 | + $p . chr(146) => "’", |
|
| 282 | + $p . chr(147) => "“", |
|
| 283 | + $p . chr(148) => "”", |
|
| 284 | + $p . chr(149) => "•", |
|
| 285 | + $p . chr(150) => "–", |
|
| 286 | + $p . chr(151) => "—", |
|
| 287 | + $p . chr(152) => "˜", |
|
| 288 | + $p . chr(153) => "™", |
|
| 289 | + $p . chr(154) => "š", |
|
| 290 | + $p . chr(155) => "›", |
|
| 291 | + $p . chr(156) => "œ", |
|
| 292 | + $p . chr(157) => ' ', # pas affecte |
|
| 293 | + $p . chr(158) => "ž", |
|
| 294 | + $p . chr(159) => "Ÿ", |
|
| 295 | + ); |
|
| 296 | + if ($charset_cible != 'unicode') { |
|
| 297 | + foreach ($trans[$charset][$charset_cible] as $k => $c) { |
|
| 298 | + $trans[$charset][$charset_cible][$k] = unicode2charset($c, $charset_cible); |
|
| 299 | + } |
|
| 300 | + } |
|
| 301 | + } |
|
| 302 | + |
|
| 303 | + return @str_replace(array_keys($trans[$charset][$charset_cible]), |
|
| 304 | + array_values($trans[$charset][$charset_cible]), $texte); |
|
| 305 | 305 | } |
| 306 | 306 | |
| 307 | 307 | |
@@ -318,24 +318,24 @@ discard block |
||
| 318 | 318 | * Texte converti |
| 319 | 319 | **/ |
| 320 | 320 | function html2unicode($texte, $secure = false) { |
| 321 | - if (strpos($texte, '&') === false) { |
|
| 322 | - return $texte; |
|
| 323 | - } |
|
| 324 | - static $trans = array(); |
|
| 325 | - if (!$trans) { |
|
| 326 | - load_charset('html'); |
|
| 327 | - foreach ($GLOBALS['CHARSET']['html'] as $key => $val) { |
|
| 328 | - $trans["&$key;"] = $val; |
|
| 329 | - } |
|
| 330 | - } |
|
| 331 | - |
|
| 332 | - if ($secure) { |
|
| 333 | - return str_replace(array_keys($trans), array_values($trans), $texte); |
|
| 334 | - } else { |
|
| 335 | - return str_replace(array('&', '"', '<', '>'), array('&', '"', '<', '>'), |
|
| 336 | - str_replace(array_keys($trans), array_values($trans), $texte) |
|
| 337 | - ); |
|
| 338 | - } |
|
| 321 | + if (strpos($texte, '&') === false) { |
|
| 322 | + return $texte; |
|
| 323 | + } |
|
| 324 | + static $trans = array(); |
|
| 325 | + if (!$trans) { |
|
| 326 | + load_charset('html'); |
|
| 327 | + foreach ($GLOBALS['CHARSET']['html'] as $key => $val) { |
|
| 328 | + $trans["&$key;"] = $val; |
|
| 329 | + } |
|
| 330 | + } |
|
| 331 | + |
|
| 332 | + if ($secure) { |
|
| 333 | + return str_replace(array_keys($trans), array_values($trans), $texte); |
|
| 334 | + } else { |
|
| 335 | + return str_replace(array('&', '"', '<', '>'), array('&', '"', '<', '>'), |
|
| 336 | + str_replace(array_keys($trans), array_values($trans), $texte) |
|
| 337 | + ); |
|
| 338 | + } |
|
| 339 | 339 | } |
| 340 | 340 | |
| 341 | 341 | |
@@ -350,16 +350,16 @@ discard block |
||
| 350 | 350 | * Texte converti |
| 351 | 351 | **/ |
| 352 | 352 | function mathml2unicode($texte) { |
| 353 | - static $trans; |
|
| 354 | - if (!$trans) { |
|
| 355 | - load_charset('mathml'); |
|
| 353 | + static $trans; |
|
| 354 | + if (!$trans) { |
|
| 355 | + load_charset('mathml'); |
|
| 356 | 356 | |
| 357 | - foreach ($GLOBALS['CHARSET']['mathml'] as $key => $val) { |
|
| 358 | - $trans["&$key;"] = $val; |
|
| 359 | - } |
|
| 360 | - } |
|
| 357 | + foreach ($GLOBALS['CHARSET']['mathml'] as $key => $val) { |
|
| 358 | + $trans["&$key;"] = $val; |
|
| 359 | + } |
|
| 360 | + } |
|
| 361 | 361 | |
| 362 | - return str_replace(array_keys($trans), array_values($trans), $texte); |
|
| 362 | + return str_replace(array_keys($trans), array_values($trans), $texte); |
|
| 363 | 363 | } |
| 364 | 364 | |
| 365 | 365 | |
@@ -381,76 +381,76 @@ discard block |
||
| 381 | 381 | * Texte converti en unicode |
| 382 | 382 | **/ |
| 383 | 383 | function charset2unicode($texte, $charset = 'AUTO' /* $forcer: obsolete*/) { |
| 384 | - static $trans; |
|
| 385 | - |
|
| 386 | - if ($charset == 'AUTO') { |
|
| 387 | - $charset = lire_config('charset', _DEFAULT_CHARSET); |
|
| 388 | - } |
|
| 389 | - |
|
| 390 | - if ($charset == '') { |
|
| 391 | - $charset = 'iso-8859-1'; |
|
| 392 | - } |
|
| 393 | - $charset = strtolower($charset); |
|
| 394 | - |
|
| 395 | - switch ($charset) { |
|
| 396 | - case 'utf-8': |
|
| 397 | - case 'utf8': |
|
| 398 | - return utf_8_to_unicode($texte); |
|
| 399 | - |
|
| 400 | - case 'iso-8859-1': |
|
| 401 | - $texte = corriger_caracteres_windows($texte, 'iso-8859-1'); |
|
| 402 | - // pas de break; ici, on suit sur default: |
|
| 403 | - |
|
| 404 | - default: |
|
| 405 | - // mbstring presente ? |
|
| 406 | - if (init_mb_string()) { |
|
| 407 | - $order = mb_detect_order(); |
|
| 408 | - try { |
|
| 409 | - # mb_string connait-il $charset? |
|
| 410 | - if ($order and mb_detect_order($charset)) { |
|
| 411 | - $s = mb_convert_encoding($texte, 'utf-8', $charset); |
|
| 412 | - if ($s && $s != $texte) { |
|
| 413 | - return utf_8_to_unicode($s); |
|
| 414 | - } |
|
| 415 | - } |
|
| 384 | + static $trans; |
|
| 385 | + |
|
| 386 | + if ($charset == 'AUTO') { |
|
| 387 | + $charset = lire_config('charset', _DEFAULT_CHARSET); |
|
| 388 | + } |
|
| 389 | + |
|
| 390 | + if ($charset == '') { |
|
| 391 | + $charset = 'iso-8859-1'; |
|
| 392 | + } |
|
| 393 | + $charset = strtolower($charset); |
|
| 394 | + |
|
| 395 | + switch ($charset) { |
|
| 396 | + case 'utf-8': |
|
| 397 | + case 'utf8': |
|
| 398 | + return utf_8_to_unicode($texte); |
|
| 399 | + |
|
| 400 | + case 'iso-8859-1': |
|
| 401 | + $texte = corriger_caracteres_windows($texte, 'iso-8859-1'); |
|
| 402 | + // pas de break; ici, on suit sur default: |
|
| 403 | + |
|
| 404 | + default: |
|
| 405 | + // mbstring presente ? |
|
| 406 | + if (init_mb_string()) { |
|
| 407 | + $order = mb_detect_order(); |
|
| 408 | + try { |
|
| 409 | + # mb_string connait-il $charset? |
|
| 410 | + if ($order and mb_detect_order($charset)) { |
|
| 411 | + $s = mb_convert_encoding($texte, 'utf-8', $charset); |
|
| 412 | + if ($s && $s != $texte) { |
|
| 413 | + return utf_8_to_unicode($s); |
|
| 414 | + } |
|
| 415 | + } |
|
| 416 | 416 | |
| 417 | - } catch (\Error $e) { |
|
| 418 | - // Le charset n'existe probablement pas |
|
| 419 | - } finally { |
|
| 420 | - mb_detect_order($order); # remettre comme precedemment |
|
| 421 | - } |
|
| 422 | - } |
|
| 423 | - |
|
| 424 | - // Sinon, peut-etre connaissons-nous ce charset ? |
|
| 425 | - if (!isset($trans[$charset])) { |
|
| 426 | - if ( |
|
| 427 | - $cset = load_charset($charset) |
|
| 428 | - and is_array($GLOBALS['CHARSET'][$cset]) |
|
| 429 | - ) { |
|
| 430 | - foreach ($GLOBALS['CHARSET'][$cset] as $key => $val) { |
|
| 431 | - $trans[$charset][chr($key)] = '&#' . $val . ';'; |
|
| 432 | - } |
|
| 433 | - } |
|
| 434 | - } |
|
| 435 | - if (isset($trans[$charset]) and count($trans[$charset])) { |
|
| 436 | - return str_replace(array_keys($trans[$charset]), array_values($trans[$charset]), $texte); |
|
| 437 | - } |
|
| 438 | - |
|
| 439 | - // Sinon demander a iconv (malgre le fait qu'il coupe quand un |
|
| 440 | - // caractere n'appartient pas au charset, mais c'est un probleme |
|
| 441 | - // surtout en utf-8, gere ci-dessus) |
|
| 442 | - if (test_iconv()) { |
|
| 443 | - $s = iconv($charset, 'utf-32le', $texte); |
|
| 444 | - if ($s) { |
|
| 445 | - return utf_32_to_unicode($s); |
|
| 446 | - } |
|
| 447 | - } |
|
| 448 | - |
|
| 449 | - // Au pire ne rien faire |
|
| 450 | - spip_log("erreur charset '$charset' non supporte"); |
|
| 451 | - |
|
| 452 | - return $texte; |
|
| 453 | - } |
|
| 417 | + } catch (\Error $e) { |
|
| 418 | + // Le charset n'existe probablement pas |
|
| 419 | + } finally { |
|
| 420 | + mb_detect_order($order); # remettre comme precedemment |
|
| 421 | + } |
|
| 422 | + } |
|
| 423 | + |
|
| 424 | + // Sinon, peut-etre connaissons-nous ce charset ? |
|
| 425 | + if (!isset($trans[$charset])) { |
|
| 426 | + if ( |
|
| 427 | + $cset = load_charset($charset) |
|
| 428 | + and is_array($GLOBALS['CHARSET'][$cset]) |
|
| 429 | + ) { |
|
| 430 | + foreach ($GLOBALS['CHARSET'][$cset] as $key => $val) { |
|
| 431 | + $trans[$charset][chr($key)] = '&#' . $val . ';'; |
|
| 432 | + } |
|
| 433 | + } |
|
| 434 | + } |
|
| 435 | + if (isset($trans[$charset]) and count($trans[$charset])) { |
|
| 436 | + return str_replace(array_keys($trans[$charset]), array_values($trans[$charset]), $texte); |
|
| 437 | + } |
|
| 438 | + |
|
| 439 | + // Sinon demander a iconv (malgre le fait qu'il coupe quand un |
|
| 440 | + // caractere n'appartient pas au charset, mais c'est un probleme |
|
| 441 | + // surtout en utf-8, gere ci-dessus) |
|
| 442 | + if (test_iconv()) { |
|
| 443 | + $s = iconv($charset, 'utf-32le', $texte); |
|
| 444 | + if ($s) { |
|
| 445 | + return utf_32_to_unicode($s); |
|
| 446 | + } |
|
| 447 | + } |
|
| 448 | + |
|
| 449 | + // Au pire ne rien faire |
|
| 450 | + spip_log("erreur charset '$charset' non supporte"); |
|
| 451 | + |
|
| 452 | + return $texte; |
|
| 453 | + } |
|
| 454 | 454 | } |
| 455 | 455 | |
| 456 | 456 | |
@@ -469,44 +469,44 @@ discard block |
||
| 469 | 469 | * Texte transformé dans le charset souhaité |
| 470 | 470 | **/ |
| 471 | 471 | function unicode2charset($texte, $charset = 'AUTO') { |
| 472 | - static $CHARSET_REVERSE = array(); |
|
| 473 | - static $trans = array(); |
|
| 474 | - |
|
| 475 | - if ($charset == 'AUTO') { |
|
| 476 | - $charset = lire_config('charset', _DEFAULT_CHARSET); |
|
| 477 | - } |
|
| 478 | - |
|
| 479 | - switch ($charset) { |
|
| 480 | - case 'utf-8': |
|
| 481 | - return unicode_to_utf_8($texte); |
|
| 482 | - break; |
|
| 483 | - |
|
| 484 | - default: |
|
| 485 | - $charset = load_charset($charset); |
|
| 486 | - |
|
| 487 | - if (empty($CHARSET_REVERSE[$charset])) { |
|
| 488 | - $CHARSET_REVERSE[$charset] = array_flip($GLOBALS['CHARSET'][$charset]); |
|
| 489 | - } |
|
| 490 | - |
|
| 491 | - if (!isset($trans[$charset])) { |
|
| 492 | - $trans[$charset] = array(); |
|
| 493 | - $t = &$trans[$charset]; |
|
| 494 | - for ($e = 128; $e < 255; $e++) { |
|
| 495 | - $h = dechex($e); |
|
| 496 | - if ($s = isset($CHARSET_REVERSE[$charset][$e])) { |
|
| 497 | - $s = $CHARSET_REVERSE[$charset][$e]; |
|
| 498 | - $t['&#' . $e . ';'] = $t['�' . $e . ';'] = $t['�' . $e . ';'] = chr($s); |
|
| 499 | - $t['&#x' . $h . ';'] = $t['�' . $h . ';'] = $t['�' . $h . ';'] = chr($s); |
|
| 500 | - } else { |
|
| 501 | - $t['&#' . $e . ';'] = $t['�' . $e . ';'] = $t['�' . $e . ';'] = chr($e); |
|
| 502 | - $t['&#x' . $h . ';'] = $t['�' . $h . ';'] = $t['�' . $h . ';'] = chr($e); |
|
| 503 | - } |
|
| 504 | - } |
|
| 505 | - } |
|
| 506 | - $texte = str_replace(array_keys($trans[$charset]), array_values($trans[$charset]), $texte); |
|
| 507 | - |
|
| 508 | - return $texte; |
|
| 509 | - } |
|
| 472 | + static $CHARSET_REVERSE = array(); |
|
| 473 | + static $trans = array(); |
|
| 474 | + |
|
| 475 | + if ($charset == 'AUTO') { |
|
| 476 | + $charset = lire_config('charset', _DEFAULT_CHARSET); |
|
| 477 | + } |
|
| 478 | + |
|
| 479 | + switch ($charset) { |
|
| 480 | + case 'utf-8': |
|
| 481 | + return unicode_to_utf_8($texte); |
|
| 482 | + break; |
|
| 483 | + |
|
| 484 | + default: |
|
| 485 | + $charset = load_charset($charset); |
|
| 486 | + |
|
| 487 | + if (empty($CHARSET_REVERSE[$charset])) { |
|
| 488 | + $CHARSET_REVERSE[$charset] = array_flip($GLOBALS['CHARSET'][$charset]); |
|
| 489 | + } |
|
| 490 | + |
|
| 491 | + if (!isset($trans[$charset])) { |
|
| 492 | + $trans[$charset] = array(); |
|
| 493 | + $t = &$trans[$charset]; |
|
| 494 | + for ($e = 128; $e < 255; $e++) { |
|
| 495 | + $h = dechex($e); |
|
| 496 | + if ($s = isset($CHARSET_REVERSE[$charset][$e])) { |
|
| 497 | + $s = $CHARSET_REVERSE[$charset][$e]; |
|
| 498 | + $t['&#' . $e . ';'] = $t['�' . $e . ';'] = $t['�' . $e . ';'] = chr($s); |
|
| 499 | + $t['&#x' . $h . ';'] = $t['�' . $h . ';'] = $t['�' . $h . ';'] = chr($s); |
|
| 500 | + } else { |
|
| 501 | + $t['&#' . $e . ';'] = $t['�' . $e . ';'] = $t['�' . $e . ';'] = chr($e); |
|
| 502 | + $t['&#x' . $h . ';'] = $t['�' . $h . ';'] = $t['�' . $h . ';'] = chr($e); |
|
| 503 | + } |
|
| 504 | + } |
|
| 505 | + } |
|
| 506 | + $texte = str_replace(array_keys($trans[$charset]), array_values($trans[$charset]), $texte); |
|
| 507 | + |
|
| 508 | + return $texte; |
|
| 509 | + } |
|
| 510 | 510 | } |
| 511 | 511 | |
| 512 | 512 | |
@@ -524,37 +524,37 @@ discard block |
||
| 524 | 524 | * Texte transformé dans le charset site |
| 525 | 525 | **/ |
| 526 | 526 | function importer_charset($texte, $charset = 'AUTO') { |
| 527 | - static $trans = array(); |
|
| 528 | - // on traite le cas le plus frequent iso-8859-1 vers utf directement pour aller plus vite ! |
|
| 529 | - if (($charset == 'iso-8859-1') && ($GLOBALS['meta']['charset'] == 'utf-8')) { |
|
| 530 | - $texte = corriger_caracteres_windows($texte, 'iso-8859-1', $GLOBALS['meta']['charset']); |
|
| 531 | - if (init_mb_string()) { |
|
| 532 | - if ($order = mb_detect_order() # mb_string connait-il $charset? |
|
| 533 | - and mb_detect_order($charset) |
|
| 534 | - ) { |
|
| 535 | - $s = mb_convert_encoding($texte, 'utf-8', $charset); |
|
| 536 | - } |
|
| 537 | - mb_detect_order($order); # remettre comme precedemment |
|
| 538 | - return $s; |
|
| 539 | - } |
|
| 540 | - // Sinon, peut-etre connaissons-nous ce charset ? |
|
| 541 | - if (!isset($trans[$charset])) { |
|
| 542 | - if ($cset = load_charset($charset) |
|
| 543 | - and is_array($GLOBALS['CHARSET'][$cset]) |
|
| 544 | - ) { |
|
| 545 | - foreach ($GLOBALS['CHARSET'][$cset] as $key => $val) { |
|
| 546 | - $trans[$charset][chr($key)] = unicode2charset('&#' . $val . ';'); |
|
| 547 | - } |
|
| 548 | - } |
|
| 549 | - } |
|
| 550 | - if (count($trans[$charset])) { |
|
| 551 | - return str_replace(array_keys($trans[$charset]), array_values($trans[$charset]), $texte); |
|
| 552 | - } |
|
| 553 | - |
|
| 554 | - return $texte; |
|
| 555 | - } |
|
| 556 | - |
|
| 557 | - return unicode2charset(charset2unicode($texte, $charset)); |
|
| 527 | + static $trans = array(); |
|
| 528 | + // on traite le cas le plus frequent iso-8859-1 vers utf directement pour aller plus vite ! |
|
| 529 | + if (($charset == 'iso-8859-1') && ($GLOBALS['meta']['charset'] == 'utf-8')) { |
|
| 530 | + $texte = corriger_caracteres_windows($texte, 'iso-8859-1', $GLOBALS['meta']['charset']); |
|
| 531 | + if (init_mb_string()) { |
|
| 532 | + if ($order = mb_detect_order() # mb_string connait-il $charset? |
|
| 533 | + and mb_detect_order($charset) |
|
| 534 | + ) { |
|
| 535 | + $s = mb_convert_encoding($texte, 'utf-8', $charset); |
|
| 536 | + } |
|
| 537 | + mb_detect_order($order); # remettre comme precedemment |
|
| 538 | + return $s; |
|
| 539 | + } |
|
| 540 | + // Sinon, peut-etre connaissons-nous ce charset ? |
|
| 541 | + if (!isset($trans[$charset])) { |
|
| 542 | + if ($cset = load_charset($charset) |
|
| 543 | + and is_array($GLOBALS['CHARSET'][$cset]) |
|
| 544 | + ) { |
|
| 545 | + foreach ($GLOBALS['CHARSET'][$cset] as $key => $val) { |
|
| 546 | + $trans[$charset][chr($key)] = unicode2charset('&#' . $val . ';'); |
|
| 547 | + } |
|
| 548 | + } |
|
| 549 | + } |
|
| 550 | + if (count($trans[$charset])) { |
|
| 551 | + return str_replace(array_keys($trans[$charset]), array_values($trans[$charset]), $texte); |
|
| 552 | + } |
|
| 553 | + |
|
| 554 | + return $texte; |
|
| 555 | + } |
|
| 556 | + |
|
| 557 | + return unicode2charset(charset2unicode($texte, $charset)); |
|
| 558 | 558 | } |
| 559 | 559 | |
| 560 | 560 | |
@@ -570,92 +570,92 @@ discard block |
||
| 570 | 570 | **/ |
| 571 | 571 | function utf_8_to_unicode($source) { |
| 572 | 572 | |
| 573 | - // mb_string : methode rapide |
|
| 574 | - if (init_mb_string()) { |
|
| 575 | - $convmap = array(0x7F, 0xFFFFFF, 0x0, 0xFFFFFF); |
|
| 576 | - |
|
| 577 | - return mb_encode_numericentity($source, $convmap, 'UTF-8'); |
|
| 578 | - } |
|
| 579 | - |
|
| 580 | - // Sinon methode pas a pas |
|
| 581 | - static $decrement; |
|
| 582 | - static $shift; |
|
| 583 | - |
|
| 584 | - // Cf. php.net, par Ronen. Adapte pour compatibilite < php4 |
|
| 585 | - if (!is_array($decrement)) { |
|
| 586 | - // array used to figure what number to decrement from character order value |
|
| 587 | - // according to number of characters used to map unicode to ascii by utf-8 |
|
| 588 | - $decrement[4] = 240; |
|
| 589 | - $decrement[3] = 224; |
|
| 590 | - $decrement[2] = 192; |
|
| 591 | - $decrement[1] = 0; |
|
| 592 | - // the number of bits to shift each charNum by |
|
| 593 | - $shift[1][0] = 0; |
|
| 594 | - $shift[2][0] = 6; |
|
| 595 | - $shift[2][1] = 0; |
|
| 596 | - $shift[3][0] = 12; |
|
| 597 | - $shift[3][1] = 6; |
|
| 598 | - $shift[3][2] = 0; |
|
| 599 | - $shift[4][0] = 18; |
|
| 600 | - $shift[4][1] = 12; |
|
| 601 | - $shift[4][2] = 6; |
|
| 602 | - $shift[4][3] = 0; |
|
| 603 | - } |
|
| 604 | - |
|
| 605 | - $pos = 0; |
|
| 606 | - $len = strlen($source); |
|
| 607 | - $encodedString = ''; |
|
| 608 | - while ($pos < $len) { |
|
| 609 | - $char = ''; |
|
| 610 | - $ischar = false; |
|
| 611 | - $asciiPos = ord(substr($source, $pos, 1)); |
|
| 612 | - if (($asciiPos >= 240) && ($asciiPos <= 255)) { |
|
| 613 | - // 4 chars representing one unicode character |
|
| 614 | - $thisLetter = substr($source, $pos, 4); |
|
| 615 | - $pos += 4; |
|
| 616 | - } else { |
|
| 617 | - if (($asciiPos >= 224) && ($asciiPos <= 239)) { |
|
| 618 | - // 3 chars representing one unicode character |
|
| 619 | - $thisLetter = substr($source, $pos, 3); |
|
| 620 | - $pos += 3; |
|
| 621 | - } else { |
|
| 622 | - if (($asciiPos >= 192) && ($asciiPos <= 223)) { |
|
| 623 | - // 2 chars representing one unicode character |
|
| 624 | - $thisLetter = substr($source, $pos, 2); |
|
| 625 | - $pos += 2; |
|
| 626 | - } else { |
|
| 627 | - // 1 char (lower ascii) |
|
| 628 | - $thisLetter = substr($source, $pos, 1); |
|
| 629 | - $pos += 1; |
|
| 630 | - $char = $thisLetter; |
|
| 631 | - $ischar = true; |
|
| 632 | - } |
|
| 633 | - } |
|
| 634 | - } |
|
| 635 | - |
|
| 636 | - if ($ischar) { |
|
| 637 | - $encodedString .= $char; |
|
| 638 | - } else { // process the string representing the letter to a unicode entity |
|
| 639 | - $thisLen = strlen($thisLetter); |
|
| 640 | - $thisPos = 0; |
|
| 641 | - $decimalCode = 0; |
|
| 642 | - while ($thisPos < $thisLen) { |
|
| 643 | - $thisCharOrd = ord(substr($thisLetter, $thisPos, 1)); |
|
| 644 | - if ($thisPos == 0) { |
|
| 645 | - $charNum = intval($thisCharOrd - $decrement[$thisLen]); |
|
| 646 | - $decimalCode += ($charNum << $shift[$thisLen][$thisPos]); |
|
| 647 | - } else { |
|
| 648 | - $charNum = intval($thisCharOrd - 128); |
|
| 649 | - $decimalCode += ($charNum << $shift[$thisLen][$thisPos]); |
|
| 650 | - } |
|
| 651 | - $thisPos++; |
|
| 652 | - } |
|
| 653 | - $encodedLetter = "&#" . preg_replace('/^0+/', '', $decimalCode) . ';'; |
|
| 654 | - $encodedString .= $encodedLetter; |
|
| 655 | - } |
|
| 656 | - } |
|
| 657 | - |
|
| 658 | - return $encodedString; |
|
| 573 | + // mb_string : methode rapide |
|
| 574 | + if (init_mb_string()) { |
|
| 575 | + $convmap = array(0x7F, 0xFFFFFF, 0x0, 0xFFFFFF); |
|
| 576 | + |
|
| 577 | + return mb_encode_numericentity($source, $convmap, 'UTF-8'); |
|
| 578 | + } |
|
| 579 | + |
|
| 580 | + // Sinon methode pas a pas |
|
| 581 | + static $decrement; |
|
| 582 | + static $shift; |
|
| 583 | + |
|
| 584 | + // Cf. php.net, par Ronen. Adapte pour compatibilite < php4 |
|
| 585 | + if (!is_array($decrement)) { |
|
| 586 | + // array used to figure what number to decrement from character order value |
|
| 587 | + // according to number of characters used to map unicode to ascii by utf-8 |
|
| 588 | + $decrement[4] = 240; |
|
| 589 | + $decrement[3] = 224; |
|
| 590 | + $decrement[2] = 192; |
|
| 591 | + $decrement[1] = 0; |
|
| 592 | + // the number of bits to shift each charNum by |
|
| 593 | + $shift[1][0] = 0; |
|
| 594 | + $shift[2][0] = 6; |
|
| 595 | + $shift[2][1] = 0; |
|
| 596 | + $shift[3][0] = 12; |
|
| 597 | + $shift[3][1] = 6; |
|
| 598 | + $shift[3][2] = 0; |
|
| 599 | + $shift[4][0] = 18; |
|
| 600 | + $shift[4][1] = 12; |
|
| 601 | + $shift[4][2] = 6; |
|
| 602 | + $shift[4][3] = 0; |
|
| 603 | + } |
|
| 604 | + |
|
| 605 | + $pos = 0; |
|
| 606 | + $len = strlen($source); |
|
| 607 | + $encodedString = ''; |
|
| 608 | + while ($pos < $len) { |
|
| 609 | + $char = ''; |
|
| 610 | + $ischar = false; |
|
| 611 | + $asciiPos = ord(substr($source, $pos, 1)); |
|
| 612 | + if (($asciiPos >= 240) && ($asciiPos <= 255)) { |
|
| 613 | + // 4 chars representing one unicode character |
|
| 614 | + $thisLetter = substr($source, $pos, 4); |
|
| 615 | + $pos += 4; |
|
| 616 | + } else { |
|
| 617 | + if (($asciiPos >= 224) && ($asciiPos <= 239)) { |
|
| 618 | + // 3 chars representing one unicode character |
|
| 619 | + $thisLetter = substr($source, $pos, 3); |
|
| 620 | + $pos += 3; |
|
| 621 | + } else { |
|
| 622 | + if (($asciiPos >= 192) && ($asciiPos <= 223)) { |
|
| 623 | + // 2 chars representing one unicode character |
|
| 624 | + $thisLetter = substr($source, $pos, 2); |
|
| 625 | + $pos += 2; |
|
| 626 | + } else { |
|
| 627 | + // 1 char (lower ascii) |
|
| 628 | + $thisLetter = substr($source, $pos, 1); |
|
| 629 | + $pos += 1; |
|
| 630 | + $char = $thisLetter; |
|
| 631 | + $ischar = true; |
|
| 632 | + } |
|
| 633 | + } |
|
| 634 | + } |
|
| 635 | + |
|
| 636 | + if ($ischar) { |
|
| 637 | + $encodedString .= $char; |
|
| 638 | + } else { // process the string representing the letter to a unicode entity |
|
| 639 | + $thisLen = strlen($thisLetter); |
|
| 640 | + $thisPos = 0; |
|
| 641 | + $decimalCode = 0; |
|
| 642 | + while ($thisPos < $thisLen) { |
|
| 643 | + $thisCharOrd = ord(substr($thisLetter, $thisPos, 1)); |
|
| 644 | + if ($thisPos == 0) { |
|
| 645 | + $charNum = intval($thisCharOrd - $decrement[$thisLen]); |
|
| 646 | + $decimalCode += ($charNum << $shift[$thisLen][$thisPos]); |
|
| 647 | + } else { |
|
| 648 | + $charNum = intval($thisCharOrd - 128); |
|
| 649 | + $decimalCode += ($charNum << $shift[$thisLen][$thisPos]); |
|
| 650 | + } |
|
| 651 | + $thisPos++; |
|
| 652 | + } |
|
| 653 | + $encodedLetter = "&#" . preg_replace('/^0+/', '', $decimalCode) . ';'; |
|
| 654 | + $encodedString .= $encodedLetter; |
|
| 655 | + } |
|
| 656 | + } |
|
| 657 | + |
|
| 658 | + return $encodedString; |
|
| 659 | 659 | } |
| 660 | 660 | |
| 661 | 661 | /** |
@@ -674,32 +674,32 @@ discard block |
||
| 674 | 674 | **/ |
| 675 | 675 | function utf_32_to_unicode($source) { |
| 676 | 676 | |
| 677 | - // mb_string : methode rapide |
|
| 678 | - if (init_mb_string()) { |
|
| 679 | - $convmap = array(0x7F, 0xFFFFFF, 0x0, 0xFFFFFF); |
|
| 680 | - $source = mb_encode_numericentity($source, $convmap, 'UTF-32LE'); |
|
| 681 | - |
|
| 682 | - return str_replace(chr(0), '', $source); |
|
| 683 | - } |
|
| 684 | - |
|
| 685 | - // Sinon methode lente |
|
| 686 | - $texte = ''; |
|
| 687 | - while ($source) { |
|
| 688 | - $words = unpack("V*", substr($source, 0, 1024)); |
|
| 689 | - $source = substr($source, 1024); |
|
| 690 | - foreach ($words as $word) { |
|
| 691 | - if ($word < 128) { |
|
| 692 | - $texte .= chr($word); |
|
| 693 | - } // ignorer le BOM - http://www.unicode.org/faq/utf_bom.html |
|
| 694 | - else { |
|
| 695 | - if ($word != 65279) { |
|
| 696 | - $texte .= '&#' . $word . ';'; |
|
| 697 | - } |
|
| 698 | - } |
|
| 699 | - } |
|
| 700 | - } |
|
| 701 | - |
|
| 702 | - return $texte; |
|
| 677 | + // mb_string : methode rapide |
|
| 678 | + if (init_mb_string()) { |
|
| 679 | + $convmap = array(0x7F, 0xFFFFFF, 0x0, 0xFFFFFF); |
|
| 680 | + $source = mb_encode_numericentity($source, $convmap, 'UTF-32LE'); |
|
| 681 | + |
|
| 682 | + return str_replace(chr(0), '', $source); |
|
| 683 | + } |
|
| 684 | + |
|
| 685 | + // Sinon methode lente |
|
| 686 | + $texte = ''; |
|
| 687 | + while ($source) { |
|
| 688 | + $words = unpack("V*", substr($source, 0, 1024)); |
|
| 689 | + $source = substr($source, 1024); |
|
| 690 | + foreach ($words as $word) { |
|
| 691 | + if ($word < 128) { |
|
| 692 | + $texte .= chr($word); |
|
| 693 | + } // ignorer le BOM - http://www.unicode.org/faq/utf_bom.html |
|
| 694 | + else { |
|
| 695 | + if ($word != 65279) { |
|
| 696 | + $texte .= '&#' . $word . ';'; |
|
| 697 | + } |
|
| 698 | + } |
|
| 699 | + } |
|
| 700 | + } |
|
| 701 | + |
|
| 702 | + return $texte; |
|
| 703 | 703 | |
| 704 | 704 | } |
| 705 | 705 | |
@@ -717,21 +717,21 @@ discard block |
||
| 717 | 717 | * Caractère utf8 si trouvé, '' sinon |
| 718 | 718 | **/ |
| 719 | 719 | function caractere_utf_8($num) { |
| 720 | - $num = intval($num); |
|
| 721 | - if ($num < 128) { |
|
| 722 | - return chr($num); |
|
| 723 | - } |
|
| 724 | - if ($num < 2048) { |
|
| 725 | - return chr(($num >> 6) + 192) . chr(($num & 63) + 128); |
|
| 726 | - } |
|
| 727 | - if ($num < 65536) { |
|
| 728 | - return chr(($num >> 12) + 224) . chr((($num >> 6) & 63) + 128) . chr(($num & 63) + 128); |
|
| 729 | - } |
|
| 730 | - if ($num < 1114112) { |
|
| 731 | - return chr(($num >> 18) + 240) . chr((($num >> 12) & 63) + 128) . chr((($num >> 6) & 63) + 128) . chr(($num & 63) + 128); |
|
| 732 | - } |
|
| 733 | - |
|
| 734 | - return ''; |
|
| 720 | + $num = intval($num); |
|
| 721 | + if ($num < 128) { |
|
| 722 | + return chr($num); |
|
| 723 | + } |
|
| 724 | + if ($num < 2048) { |
|
| 725 | + return chr(($num >> 6) + 192) . chr(($num & 63) + 128); |
|
| 726 | + } |
|
| 727 | + if ($num < 65536) { |
|
| 728 | + return chr(($num >> 12) + 224) . chr((($num >> 6) & 63) + 128) . chr(($num & 63) + 128); |
|
| 729 | + } |
|
| 730 | + if ($num < 1114112) { |
|
| 731 | + return chr(($num >> 18) + 240) . chr((($num >> 12) & 63) + 128) . chr((($num >> 6) & 63) + 128) . chr(($num & 63) + 128); |
|
| 732 | + } |
|
| 733 | + |
|
| 734 | + return ''; |
|
| 735 | 735 | } |
| 736 | 736 | |
| 737 | 737 | /** |
@@ -744,30 +744,30 @@ discard block |
||
| 744 | 744 | **/ |
| 745 | 745 | function unicode_to_utf_8($texte) { |
| 746 | 746 | |
| 747 | - // 1. Entites € et suivantes |
|
| 748 | - $vu = array(); |
|
| 749 | - if (preg_match_all(',�*([1-9][0-9][0-9]+);,S', |
|
| 750 | - $texte, $regs, PREG_SET_ORDER)) { |
|
| 751 | - foreach ($regs as $reg) { |
|
| 752 | - if ($reg[1] > 127 and !isset($vu[$reg[0]])) { |
|
| 753 | - $vu[$reg[0]] = caractere_utf_8($reg[1]); |
|
| 754 | - } |
|
| 755 | - } |
|
| 756 | - } |
|
| 757 | - //$texte = str_replace(array_keys($vu), array_values($vu), $texte); |
|
| 758 | - |
|
| 759 | - // 2. Entites > ÿ |
|
| 760 | - //$vu = array(); |
|
| 761 | - if (preg_match_all(',�*([1-9a-f][0-9a-f][0-9a-f]+);,iS', |
|
| 762 | - $texte, $regs, PREG_SET_ORDER)) { |
|
| 763 | - foreach ($regs as $reg) { |
|
| 764 | - if (!isset($vu[$reg[0]])) { |
|
| 765 | - $vu[$reg[0]] = caractere_utf_8(hexdec($reg[1])); |
|
| 766 | - } |
|
| 767 | - } |
|
| 768 | - } |
|
| 769 | - |
|
| 770 | - return str_replace(array_keys($vu), array_values($vu), $texte); |
|
| 747 | + // 1. Entites € et suivantes |
|
| 748 | + $vu = array(); |
|
| 749 | + if (preg_match_all(',�*([1-9][0-9][0-9]+);,S', |
|
| 750 | + $texte, $regs, PREG_SET_ORDER)) { |
|
| 751 | + foreach ($regs as $reg) { |
|
| 752 | + if ($reg[1] > 127 and !isset($vu[$reg[0]])) { |
|
| 753 | + $vu[$reg[0]] = caractere_utf_8($reg[1]); |
|
| 754 | + } |
|
| 755 | + } |
|
| 756 | + } |
|
| 757 | + //$texte = str_replace(array_keys($vu), array_values($vu), $texte); |
|
| 758 | + |
|
| 759 | + // 2. Entites > ÿ |
|
| 760 | + //$vu = array(); |
|
| 761 | + if (preg_match_all(',�*([1-9a-f][0-9a-f][0-9a-f]+);,iS', |
|
| 762 | + $texte, $regs, PREG_SET_ORDER)) { |
|
| 763 | + foreach ($regs as $reg) { |
|
| 764 | + if (!isset($vu[$reg[0]])) { |
|
| 765 | + $vu[$reg[0]] = caractere_utf_8(hexdec($reg[1])); |
|
| 766 | + } |
|
| 767 | + } |
|
| 768 | + } |
|
| 769 | + |
|
| 770 | + return str_replace(array_keys($vu), array_values($vu), $texte); |
|
| 771 | 771 | |
| 772 | 772 | } |
| 773 | 773 | |
@@ -780,15 +780,15 @@ discard block |
||
| 780 | 780 | * Texte converti |
| 781 | 781 | **/ |
| 782 | 782 | function unicode_to_javascript($texte) { |
| 783 | - $vu = array(); |
|
| 784 | - while (preg_match(',�*([0-9]+);,S', $texte, $regs) and !isset($vu[$regs[1]])) { |
|
| 785 | - $num = $regs[1]; |
|
| 786 | - $vu[$num] = true; |
|
| 787 | - $s = '\u' . sprintf("%04x", $num); |
|
| 788 | - $texte = str_replace($regs[0], $s, $texte); |
|
| 789 | - } |
|
| 790 | - |
|
| 791 | - return $texte; |
|
| 783 | + $vu = array(); |
|
| 784 | + while (preg_match(',�*([0-9]+);,S', $texte, $regs) and !isset($vu[$regs[1]])) { |
|
| 785 | + $num = $regs[1]; |
|
| 786 | + $vu[$num] = true; |
|
| 787 | + $s = '\u' . sprintf("%04x", $num); |
|
| 788 | + $texte = str_replace($regs[0], $s, $texte); |
|
| 789 | + } |
|
| 790 | + |
|
| 791 | + return $texte; |
|
| 792 | 792 | } |
| 793 | 793 | |
| 794 | 794 | /** |
@@ -800,11 +800,11 @@ discard block |
||
| 800 | 800 | * Texte converti |
| 801 | 801 | **/ |
| 802 | 802 | function javascript_to_unicode($texte) { |
| 803 | - while (preg_match(",%u([0-9A-F][0-9A-F][0-9A-F][0-9A-F]),", $texte, $regs)) { |
|
| 804 | - $texte = str_replace($regs[0], "&#" . hexdec($regs[1]) . ";", $texte); |
|
| 805 | - } |
|
| 803 | + while (preg_match(",%u([0-9A-F][0-9A-F][0-9A-F][0-9A-F]),", $texte, $regs)) { |
|
| 804 | + $texte = str_replace($regs[0], "&#" . hexdec($regs[1]) . ";", $texte); |
|
| 805 | + } |
|
| 806 | 806 | |
| 807 | - return $texte; |
|
| 807 | + return $texte; |
|
| 808 | 808 | } |
| 809 | 809 | |
| 810 | 810 | /** |
@@ -816,11 +816,11 @@ discard block |
||
| 816 | 816 | * Texte converti |
| 817 | 817 | **/ |
| 818 | 818 | function javascript_to_binary($texte) { |
| 819 | - while (preg_match(",%([0-9A-F][0-9A-F]),", $texte, $regs)) { |
|
| 820 | - $texte = str_replace($regs[0], chr(hexdec($regs[1])), $texte); |
|
| 821 | - } |
|
| 819 | + while (preg_match(",%([0-9A-F][0-9A-F]),", $texte, $regs)) { |
|
| 820 | + $texte = str_replace($regs[0], chr(hexdec($regs[1])), $texte); |
|
| 821 | + } |
|
| 822 | 822 | |
| 823 | - return $texte; |
|
| 823 | + return $texte; |
|
| 824 | 824 | } |
| 825 | 825 | |
| 826 | 826 | |
@@ -838,26 +838,26 @@ discard block |
||
| 838 | 838 | * @return string |
| 839 | 839 | */ |
| 840 | 840 | function translitteration_rapide($texte, $charset = 'AUTO', $complexe = '') { |
| 841 | - static $trans = []; |
|
| 842 | - if ($charset == 'AUTO') { |
|
| 843 | - $charset = $GLOBALS['meta']['charset']; |
|
| 844 | - } |
|
| 845 | - if (!strlen($texte)) { |
|
| 846 | - return $texte; |
|
| 847 | - } |
|
| 848 | - |
|
| 849 | - $table_translit = 'translit' . $complexe; |
|
| 850 | - |
|
| 851 | - // 2. Translitterer grace a la table predefinie |
|
| 852 | - if (!isset($trans[$complexe])) { |
|
| 853 | - $trans[$complexe] = []; |
|
| 854 | - load_charset($table_translit); |
|
| 855 | - foreach ($GLOBALS['CHARSET'][$table_translit] as $key => $val) { |
|
| 856 | - $trans[$complexe][caractere_utf_8($key)] = $val; |
|
| 857 | - } |
|
| 858 | - } |
|
| 859 | - |
|
| 860 | - return str_replace(array_keys($trans[$complexe]), array_values($trans[$complexe]), $texte); |
|
| 841 | + static $trans = []; |
|
| 842 | + if ($charset == 'AUTO') { |
|
| 843 | + $charset = $GLOBALS['meta']['charset']; |
|
| 844 | + } |
|
| 845 | + if (!strlen($texte)) { |
|
| 846 | + return $texte; |
|
| 847 | + } |
|
| 848 | + |
|
| 849 | + $table_translit = 'translit' . $complexe; |
|
| 850 | + |
|
| 851 | + // 2. Translitterer grace a la table predefinie |
|
| 852 | + if (!isset($trans[$complexe])) { |
|
| 853 | + $trans[$complexe] = []; |
|
| 854 | + load_charset($table_translit); |
|
| 855 | + foreach ($GLOBALS['CHARSET'][$table_translit] as $key => $val) { |
|
| 856 | + $trans[$complexe][caractere_utf_8($key)] = $val; |
|
| 857 | + } |
|
| 858 | + } |
|
| 859 | + |
|
| 860 | + return str_replace(array_keys($trans[$complexe]), array_values($trans[$complexe]), $texte); |
|
| 861 | 861 | } |
| 862 | 862 | |
| 863 | 863 | /** |
@@ -880,14 +880,14 @@ discard block |
||
| 880 | 880 | * @return string |
| 881 | 881 | */ |
| 882 | 882 | function translitteration($texte, $charset = 'AUTO', $complexe = '') { |
| 883 | - // 0. Supprimer les caracteres illegaux |
|
| 884 | - include_spip('inc/filtres'); |
|
| 885 | - $texte = corriger_caracteres($texte); |
|
| 883 | + // 0. Supprimer les caracteres illegaux |
|
| 884 | + include_spip('inc/filtres'); |
|
| 885 | + $texte = corriger_caracteres($texte); |
|
| 886 | 886 | |
| 887 | - // 1. Passer le charset et les é en utf-8 |
|
| 888 | - $texte = unicode_to_utf_8(html2unicode(charset2unicode($texte, $charset, true))); |
|
| 887 | + // 1. Passer le charset et les é en utf-8 |
|
| 888 | + $texte = unicode_to_utf_8(html2unicode(charset2unicode($texte, $charset, true))); |
|
| 889 | 889 | |
| 890 | - return translitteration_rapide($texte, $charset, $complexe); |
|
| 890 | + return translitteration_rapide($texte, $charset, $complexe); |
|
| 891 | 891 | } |
| 892 | 892 | |
| 893 | 893 | /** |
@@ -902,17 +902,17 @@ discard block |
||
| 902 | 902 | * @return string |
| 903 | 903 | */ |
| 904 | 904 | function translitteration_complexe($texte, $chiffres = false) { |
| 905 | - $texte = translitteration($texte, 'AUTO', 'complexe'); |
|
| 905 | + $texte = translitteration($texte, 'AUTO', 'complexe'); |
|
| 906 | 906 | |
| 907 | - if ($chiffres) { |
|
| 908 | - $texte = preg_replace_callback( |
|
| 909 | - "/[aeiuoyd]['`?~.^+(-]{1,2}/S", |
|
| 910 | - function($m) { return translitteration_chiffree($m[0]); }, |
|
| 911 | - $texte |
|
| 912 | - ); |
|
| 913 | - } |
|
| 907 | + if ($chiffres) { |
|
| 908 | + $texte = preg_replace_callback( |
|
| 909 | + "/[aeiuoyd]['`?~.^+(-]{1,2}/S", |
|
| 910 | + function($m) { return translitteration_chiffree($m[0]); }, |
|
| 911 | + $texte |
|
| 912 | + ); |
|
| 913 | + } |
|
| 914 | 914 | |
| 915 | - return $texte; |
|
| 915 | + return $texte; |
|
| 916 | 916 | } |
| 917 | 917 | |
| 918 | 918 | /** |
@@ -924,7 +924,7 @@ discard block |
||
| 924 | 924 | * @return string |
| 925 | 925 | */ |
| 926 | 926 | function translitteration_chiffree($car) { |
| 927 | - return strtr($car, "'`?~.^+(-", "123456789"); |
|
| 927 | + return strtr($car, "'`?~.^+(-", "123456789"); |
|
| 928 | 928 | } |
| 929 | 929 | |
| 930 | 930 | |
@@ -937,7 +937,7 @@ discard block |
||
| 937 | 937 | * true s'il a un BOM |
| 938 | 938 | **/ |
| 939 | 939 | function bom_utf8($texte) { |
| 940 | - return (substr($texte, 0, 3) == chr(0xEF) . chr(0xBB) . chr(0xBF)); |
|
| 940 | + return (substr($texte, 0, 3) == chr(0xEF) . chr(0xBB) . chr(0xBF)); |
|
| 941 | 941 | } |
| 942 | 942 | |
| 943 | 943 | /** |
@@ -954,18 +954,18 @@ discard block |
||
| 954 | 954 | * true si c'est le cas |
| 955 | 955 | **/ |
| 956 | 956 | function is_utf8($string) { |
| 957 | - return !strlen( |
|
| 958 | - preg_replace( |
|
| 959 | - ',[\x09\x0A\x0D\x20-\x7E]' # ASCII |
|
| 960 | - . '|[\xC2-\xDF][\x80-\xBF]' # non-overlong 2-byte |
|
| 961 | - . '|\xE0[\xA0-\xBF][\x80-\xBF]' # excluding overlongs |
|
| 962 | - . '|[\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}' # straight 3-byte |
|
| 963 | - . '|\xED[\x80-\x9F][\x80-\xBF]' # excluding surrogates |
|
| 964 | - . '|\xF0[\x90-\xBF][\x80-\xBF]{2}' # planes 1-3 |
|
| 965 | - . '|[\xF1-\xF3][\x80-\xBF]{3}' # planes 4-15 |
|
| 966 | - . '|\xF4[\x80-\x8F][\x80-\xBF]{2}' # plane 16 |
|
| 967 | - . ',sS', |
|
| 968 | - '', $string)); |
|
| 957 | + return !strlen( |
|
| 958 | + preg_replace( |
|
| 959 | + ',[\x09\x0A\x0D\x20-\x7E]' # ASCII |
|
| 960 | + . '|[\xC2-\xDF][\x80-\xBF]' # non-overlong 2-byte |
|
| 961 | + . '|\xE0[\xA0-\xBF][\x80-\xBF]' # excluding overlongs |
|
| 962 | + . '|[\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}' # straight 3-byte |
|
| 963 | + . '|\xED[\x80-\x9F][\x80-\xBF]' # excluding surrogates |
|
| 964 | + . '|\xF0[\x90-\xBF][\x80-\xBF]{2}' # planes 1-3 |
|
| 965 | + . '|[\xF1-\xF3][\x80-\xBF]{3}' # planes 4-15 |
|
| 966 | + . '|\xF4[\x80-\x8F][\x80-\xBF]{2}' # plane 16 |
|
| 967 | + . ',sS', |
|
| 968 | + '', $string)); |
|
| 969 | 969 | } |
| 970 | 970 | |
| 971 | 971 | /** |
@@ -977,10 +977,10 @@ discard block |
||
| 977 | 977 | * true si c'est le cas |
| 978 | 978 | **/ |
| 979 | 979 | function is_ascii($string) { |
| 980 | - return !strlen( |
|
| 981 | - preg_replace( |
|
| 982 | - ',[\x09\x0A\x0D\x20-\x7E],sS', |
|
| 983 | - '', $string)); |
|
| 980 | + return !strlen( |
|
| 981 | + preg_replace( |
|
| 982 | + ',[\x09\x0A\x0D\x20-\x7E],sS', |
|
| 983 | + '', $string)); |
|
| 984 | 984 | } |
| 985 | 985 | |
| 986 | 986 | /** |
@@ -999,53 +999,53 @@ discard block |
||
| 999 | 999 | **/ |
| 1000 | 1000 | function transcoder_page($texte, $headers = '') { |
| 1001 | 1001 | |
| 1002 | - // Si tout est < 128 pas la peine d'aller plus loin |
|
| 1003 | - if (is_ascii($texte)) { |
|
| 1004 | - #spip_log('charset: ascii'); |
|
| 1005 | - return $texte; |
|
| 1006 | - } |
|
| 1007 | - |
|
| 1008 | - if (bom_utf8($texte)) { |
|
| 1009 | - // Reconnaitre le BOM utf-8 (0xEFBBBF) |
|
| 1010 | - $charset = 'utf-8'; |
|
| 1011 | - $texte = substr($texte, 3); |
|
| 1012 | - } elseif (preg_match(',<[?]xml[^>]*encoding[^>]*=[^>]*([-_a-z0-9]+?),UimsS', $texte, $regs)) { |
|
| 1013 | - // charset precise par le contenu (xml) |
|
| 1014 | - $charset = trim(strtolower($regs[1])); |
|
| 1015 | - } elseif ( |
|
| 1016 | - // charset precise par le contenu (html) |
|
| 1017 | - preg_match(',<(meta|html|body)[^>]*charset[^>]*=[^>]*([#-_a-z0-9]+?),UimsS', $texte, $regs) |
|
| 1018 | - # eviter toute balise SPIP tel que #CHARSET ou #CONFIG d'un squelette |
|
| 1019 | - and false === strpos($regs[2], '#') |
|
| 1020 | - and $tmp = trim(strtolower($regs[2])) |
|
| 1021 | - ) { |
|
| 1022 | - $charset = $tmp; |
|
| 1023 | - } elseif (preg_match(',charset=([-_a-z0-9]+),i', $headers, $regs)) { |
|
| 1024 | - // charset de la reponse http |
|
| 1025 | - $charset = trim(strtolower($regs[1])); |
|
| 1026 | - } else { |
|
| 1027 | - $charset = ''; |
|
| 1028 | - } |
|
| 1029 | - |
|
| 1030 | - |
|
| 1031 | - // normaliser les noms du shif-jis japonais |
|
| 1032 | - if (preg_match(',^(x|shift)[_-]s?jis$,i', $charset)) { |
|
| 1033 | - $charset = 'shift-jis'; |
|
| 1034 | - } |
|
| 1035 | - |
|
| 1036 | - if ($charset) { |
|
| 1037 | - spip_log("charset: $charset"); |
|
| 1038 | - } else { |
|
| 1039 | - // valeur par defaut |
|
| 1040 | - if (is_utf8($texte)) { |
|
| 1041 | - $charset = 'utf-8'; |
|
| 1042 | - } else { |
|
| 1043 | - $charset = 'iso-8859-1'; |
|
| 1044 | - } |
|
| 1045 | - spip_log("charset probable: $charset"); |
|
| 1046 | - } |
|
| 1047 | - |
|
| 1048 | - return importer_charset($texte, $charset); |
|
| 1002 | + // Si tout est < 128 pas la peine d'aller plus loin |
|
| 1003 | + if (is_ascii($texte)) { |
|
| 1004 | + #spip_log('charset: ascii'); |
|
| 1005 | + return $texte; |
|
| 1006 | + } |
|
| 1007 | + |
|
| 1008 | + if (bom_utf8($texte)) { |
|
| 1009 | + // Reconnaitre le BOM utf-8 (0xEFBBBF) |
|
| 1010 | + $charset = 'utf-8'; |
|
| 1011 | + $texte = substr($texte, 3); |
|
| 1012 | + } elseif (preg_match(',<[?]xml[^>]*encoding[^>]*=[^>]*([-_a-z0-9]+?),UimsS', $texte, $regs)) { |
|
| 1013 | + // charset precise par le contenu (xml) |
|
| 1014 | + $charset = trim(strtolower($regs[1])); |
|
| 1015 | + } elseif ( |
|
| 1016 | + // charset precise par le contenu (html) |
|
| 1017 | + preg_match(',<(meta|html|body)[^>]*charset[^>]*=[^>]*([#-_a-z0-9]+?),UimsS', $texte, $regs) |
|
| 1018 | + # eviter toute balise SPIP tel que #CHARSET ou #CONFIG d'un squelette |
|
| 1019 | + and false === strpos($regs[2], '#') |
|
| 1020 | + and $tmp = trim(strtolower($regs[2])) |
|
| 1021 | + ) { |
|
| 1022 | + $charset = $tmp; |
|
| 1023 | + } elseif (preg_match(',charset=([-_a-z0-9]+),i', $headers, $regs)) { |
|
| 1024 | + // charset de la reponse http |
|
| 1025 | + $charset = trim(strtolower($regs[1])); |
|
| 1026 | + } else { |
|
| 1027 | + $charset = ''; |
|
| 1028 | + } |
|
| 1029 | + |
|
| 1030 | + |
|
| 1031 | + // normaliser les noms du shif-jis japonais |
|
| 1032 | + if (preg_match(',^(x|shift)[_-]s?jis$,i', $charset)) { |
|
| 1033 | + $charset = 'shift-jis'; |
|
| 1034 | + } |
|
| 1035 | + |
|
| 1036 | + if ($charset) { |
|
| 1037 | + spip_log("charset: $charset"); |
|
| 1038 | + } else { |
|
| 1039 | + // valeur par defaut |
|
| 1040 | + if (is_utf8($texte)) { |
|
| 1041 | + $charset = 'utf-8'; |
|
| 1042 | + } else { |
|
| 1043 | + $charset = 'iso-8859-1'; |
|
| 1044 | + } |
|
| 1045 | + spip_log("charset probable: $charset"); |
|
| 1046 | + } |
|
| 1047 | + |
|
| 1048 | + return importer_charset($texte, $charset); |
|
| 1049 | 1049 | } |
| 1050 | 1050 | |
| 1051 | 1051 | |
@@ -1069,26 +1069,26 @@ discard block |
||
| 1069 | 1069 | * Le texte coupé |
| 1070 | 1070 | **/ |
| 1071 | 1071 | function spip_substr($c, $start = 0, $length = null) { |
| 1072 | - // Si ce n'est pas utf-8, utiliser substr |
|
| 1073 | - if ($GLOBALS['meta']['charset'] != 'utf-8') { |
|
| 1074 | - if ($length) { |
|
| 1075 | - return substr($c, $start, $length); |
|
| 1076 | - } else { |
|
| 1077 | - substr($c, $start); |
|
| 1078 | - } |
|
| 1079 | - } |
|
| 1080 | - |
|
| 1081 | - // Si utf-8, voir si on dispose de mb_string |
|
| 1082 | - if (init_mb_string()) { |
|
| 1083 | - if ($length) { |
|
| 1084 | - return mb_substr($c, $start, $length); |
|
| 1085 | - } else { |
|
| 1086 | - return mb_substr($c, $start); |
|
| 1087 | - } |
|
| 1088 | - } |
|
| 1089 | - |
|
| 1090 | - // Version manuelle (cf. ci-dessous) |
|
| 1091 | - return spip_substr_manuelle($c, $start, $length); |
|
| 1072 | + // Si ce n'est pas utf-8, utiliser substr |
|
| 1073 | + if ($GLOBALS['meta']['charset'] != 'utf-8') { |
|
| 1074 | + if ($length) { |
|
| 1075 | + return substr($c, $start, $length); |
|
| 1076 | + } else { |
|
| 1077 | + substr($c, $start); |
|
| 1078 | + } |
|
| 1079 | + } |
|
| 1080 | + |
|
| 1081 | + // Si utf-8, voir si on dispose de mb_string |
|
| 1082 | + if (init_mb_string()) { |
|
| 1083 | + if ($length) { |
|
| 1084 | + return mb_substr($c, $start, $length); |
|
| 1085 | + } else { |
|
| 1086 | + return mb_substr($c, $start); |
|
| 1087 | + } |
|
| 1088 | + } |
|
| 1089 | + |
|
| 1090 | + // Version manuelle (cf. ci-dessous) |
|
| 1091 | + return spip_substr_manuelle($c, $start, $length); |
|
| 1092 | 1092 | } |
| 1093 | 1093 | |
| 1094 | 1094 | |
@@ -1107,40 +1107,40 @@ discard block |
||
| 1107 | 1107 | **/ |
| 1108 | 1108 | function spip_substr_manuelle($c, $start, $length = null) { |
| 1109 | 1109 | |
| 1110 | - // Cas pathologique |
|
| 1111 | - if ($length === 0) { |
|
| 1112 | - return ''; |
|
| 1113 | - } |
|
| 1114 | - |
|
| 1115 | - // S'il y a un demarrage, on se positionne |
|
| 1116 | - if ($start > 0) { |
|
| 1117 | - $c = substr($c, strlen(spip_substr_manuelle($c, 0, $start))); |
|
| 1118 | - } elseif ($start < 0) { |
|
| 1119 | - return spip_substr_manuelle($c, spip_strlen($c) + $start, $length); |
|
| 1120 | - } |
|
| 1121 | - |
|
| 1122 | - if (!$length) { |
|
| 1123 | - return $c; |
|
| 1124 | - } |
|
| 1125 | - |
|
| 1126 | - if ($length > 0) { |
|
| 1127 | - // on prend n fois la longueur desiree, pour etre surs d'avoir tout |
|
| 1128 | - // (un caractere utf-8 prenant au maximum n bytes) |
|
| 1129 | - $n = 0; |
|
| 1130 | - while (preg_match(',[\x80-\xBF]{' . (++$n) . '},', $c)) { |
|
| 1131 | - ; |
|
| 1132 | - } |
|
| 1133 | - $c = substr($c, 0, $n * $length); |
|
| 1134 | - // puis, tant qu'on est trop long, on coupe... |
|
| 1135 | - while (($l = spip_strlen($c)) > $length) { |
|
| 1136 | - $c = substr($c, 0, $length - $l); |
|
| 1137 | - } |
|
| 1138 | - |
|
| 1139 | - return $c; |
|
| 1140 | - } |
|
| 1141 | - |
|
| 1142 | - // $length < 0 |
|
| 1143 | - return spip_substr_manuelle($c, 0, spip_strlen($c) + $length); |
|
| 1110 | + // Cas pathologique |
|
| 1111 | + if ($length === 0) { |
|
| 1112 | + return ''; |
|
| 1113 | + } |
|
| 1114 | + |
|
| 1115 | + // S'il y a un demarrage, on se positionne |
|
| 1116 | + if ($start > 0) { |
|
| 1117 | + $c = substr($c, strlen(spip_substr_manuelle($c, 0, $start))); |
|
| 1118 | + } elseif ($start < 0) { |
|
| 1119 | + return spip_substr_manuelle($c, spip_strlen($c) + $start, $length); |
|
| 1120 | + } |
|
| 1121 | + |
|
| 1122 | + if (!$length) { |
|
| 1123 | + return $c; |
|
| 1124 | + } |
|
| 1125 | + |
|
| 1126 | + if ($length > 0) { |
|
| 1127 | + // on prend n fois la longueur desiree, pour etre surs d'avoir tout |
|
| 1128 | + // (un caractere utf-8 prenant au maximum n bytes) |
|
| 1129 | + $n = 0; |
|
| 1130 | + while (preg_match(',[\x80-\xBF]{' . (++$n) . '},', $c)) { |
|
| 1131 | + ; |
|
| 1132 | + } |
|
| 1133 | + $c = substr($c, 0, $n * $length); |
|
| 1134 | + // puis, tant qu'on est trop long, on coupe... |
|
| 1135 | + while (($l = spip_strlen($c)) > $length) { |
|
| 1136 | + $c = substr($c, 0, $length - $l); |
|
| 1137 | + } |
|
| 1138 | + |
|
| 1139 | + return $c; |
|
| 1140 | + } |
|
| 1141 | + |
|
| 1142 | + // $length < 0 |
|
| 1143 | + return spip_substr_manuelle($c, 0, spip_strlen($c) + $length); |
|
| 1144 | 1144 | } |
| 1145 | 1145 | |
| 1146 | 1146 | /** |
@@ -1154,14 +1154,14 @@ discard block |
||
| 1154 | 1154 | * La chaîne avec une majuscule sur le premier mot |
| 1155 | 1155 | */ |
| 1156 | 1156 | function spip_ucfirst($c) { |
| 1157 | - // Si on n'a pas mb_* ou si ce n'est pas utf-8, utiliser ucfirst |
|
| 1158 | - if (!init_mb_string() or $GLOBALS['meta']['charset'] != 'utf-8') { |
|
| 1159 | - return ucfirst($c); |
|
| 1160 | - } |
|
| 1157 | + // Si on n'a pas mb_* ou si ce n'est pas utf-8, utiliser ucfirst |
|
| 1158 | + if (!init_mb_string() or $GLOBALS['meta']['charset'] != 'utf-8') { |
|
| 1159 | + return ucfirst($c); |
|
| 1160 | + } |
|
| 1161 | 1161 | |
| 1162 | - $lettre1 = mb_strtoupper(spip_substr($c, 0, 1)); |
|
| 1162 | + $lettre1 = mb_strtoupper(spip_substr($c, 0, 1)); |
|
| 1163 | 1163 | |
| 1164 | - return $lettre1 . spip_substr($c, 1); |
|
| 1164 | + return $lettre1 . spip_substr($c, 1); |
|
| 1165 | 1165 | } |
| 1166 | 1166 | |
| 1167 | 1167 | /** |
@@ -1175,12 +1175,12 @@ discard block |
||
| 1175 | 1175 | * La chaîne en minuscules |
| 1176 | 1176 | */ |
| 1177 | 1177 | function spip_strtolower($c) { |
| 1178 | - // Si on n'a pas mb_* ou si ce n'est pas utf-8, utiliser strtolower |
|
| 1179 | - if (!init_mb_string() or $GLOBALS['meta']['charset'] != 'utf-8') { |
|
| 1180 | - return strtolower($c); |
|
| 1181 | - } |
|
| 1178 | + // Si on n'a pas mb_* ou si ce n'est pas utf-8, utiliser strtolower |
|
| 1179 | + if (!init_mb_string() or $GLOBALS['meta']['charset'] != 'utf-8') { |
|
| 1180 | + return strtolower($c); |
|
| 1181 | + } |
|
| 1182 | 1182 | |
| 1183 | - return mb_strtolower($c); |
|
| 1183 | + return mb_strtolower($c); |
|
| 1184 | 1184 | } |
| 1185 | 1185 | |
| 1186 | 1186 | /** |
@@ -1194,23 +1194,23 @@ discard block |
||
| 1194 | 1194 | * Longueur de la chaîne |
| 1195 | 1195 | */ |
| 1196 | 1196 | function spip_strlen($c) { |
| 1197 | - // On transforme les sauts de ligne pour ne pas compter deux caractères |
|
| 1198 | - $c = str_replace("\r\n", "\n", $c); |
|
| 1199 | - |
|
| 1200 | - // Si ce n'est pas utf-8, utiliser strlen |
|
| 1201 | - if ($GLOBALS['meta']['charset'] != 'utf-8') { |
|
| 1202 | - return strlen($c); |
|
| 1203 | - } |
|
| 1204 | - |
|
| 1205 | - // Sinon, utiliser mb_strlen() si disponible |
|
| 1206 | - if (init_mb_string()) { |
|
| 1207 | - return mb_strlen($c); |
|
| 1208 | - } |
|
| 1209 | - |
|
| 1210 | - // Methode manuelle : on supprime les bytes 10......, |
|
| 1211 | - // on compte donc les ascii (0.......) et les demarrages |
|
| 1212 | - // de caracteres utf-8 (11......) |
|
| 1213 | - return strlen(preg_replace(',[\x80-\xBF],S', '', $c)); |
|
| 1197 | + // On transforme les sauts de ligne pour ne pas compter deux caractères |
|
| 1198 | + $c = str_replace("\r\n", "\n", $c); |
|
| 1199 | + |
|
| 1200 | + // Si ce n'est pas utf-8, utiliser strlen |
|
| 1201 | + if ($GLOBALS['meta']['charset'] != 'utf-8') { |
|
| 1202 | + return strlen($c); |
|
| 1203 | + } |
|
| 1204 | + |
|
| 1205 | + // Sinon, utiliser mb_strlen() si disponible |
|
| 1206 | + if (init_mb_string()) { |
|
| 1207 | + return mb_strlen($c); |
|
| 1208 | + } |
|
| 1209 | + |
|
| 1210 | + // Methode manuelle : on supprime les bytes 10......, |
|
| 1211 | + // on compte donc les ascii (0.......) et les demarrages |
|
| 1212 | + // de caracteres utf-8 (11......) |
|
| 1213 | + return strlen(preg_replace(',[\x80-\xBF],S', '', $c)); |
|
| 1214 | 1214 | } |
| 1215 | 1215 | |
| 1216 | 1216 | // Initialisation |
@@ -1220,14 +1220,14 @@ discard block |
||
| 1220 | 1220 | // dans les preg_replace pour ne pas casser certaines lettres accentuees : |
| 1221 | 1221 | // en utf-8 chr(195).chr(160) = a` alors qu'en iso-latin chr(160) = nbsp |
| 1222 | 1222 | if (!isset($GLOBALS['meta']['pcre_u']) |
| 1223 | - or (isset($_GET['var_mode']) and !isset($_GET['var_profile'])) |
|
| 1223 | + or (isset($_GET['var_mode']) and !isset($_GET['var_profile'])) |
|
| 1224 | 1224 | ) { |
| 1225 | - include_spip('inc/meta'); |
|
| 1226 | - ecrire_meta('pcre_u', |
|
| 1227 | - $u = (lire_config('charset', _DEFAULT_CHARSET) == 'utf-8' |
|
| 1228 | - and test_pcre_unicode()) |
|
| 1229 | - ? 'u' : '' |
|
| 1230 | - ); |
|
| 1225 | + include_spip('inc/meta'); |
|
| 1226 | + ecrire_meta('pcre_u', |
|
| 1227 | + $u = (lire_config('charset', _DEFAULT_CHARSET) == 'utf-8' |
|
| 1228 | + and test_pcre_unicode()) |
|
| 1229 | + ? 'u' : '' |
|
| 1230 | + ); |
|
| 1231 | 1231 | } |
| 1232 | 1232 | |
| 1233 | 1233 | |
@@ -1243,17 +1243,17 @@ discard block |
||
| 1243 | 1243 | * en unicode : 💩 |
| 1244 | 1244 | */ |
| 1245 | 1245 | function utf8_noplanes($x) { |
| 1246 | - $regexp_utf8_4bytes = '/( |
|
| 1246 | + $regexp_utf8_4bytes = '/( |
|
| 1247 | 1247 | \xF0[\x90-\xBF][\x80-\xBF]{2} # planes 1-3 |
| 1248 | 1248 | | [\xF1-\xF3][\x80-\xBF]{3} # planes 4-15 |
| 1249 | 1249 | | \xF4[\x80-\x8F][\x80-\xBF]{2} # plane 16 |
| 1250 | 1250 | )/xS'; |
| 1251 | - if (preg_match_all($regexp_utf8_4bytes, $x, $z, PREG_PATTERN_ORDER)) { |
|
| 1252 | - foreach ($z[0] as $k) { |
|
| 1253 | - $ku = utf_8_to_unicode($k); |
|
| 1254 | - $x = str_replace($k, $ku, $x); |
|
| 1255 | - } |
|
| 1256 | - } |
|
| 1257 | - |
|
| 1258 | - return $x; |
|
| 1251 | + if (preg_match_all($regexp_utf8_4bytes, $x, $z, PREG_PATTERN_ORDER)) { |
|
| 1252 | + foreach ($z[0] as $k) { |
|
| 1253 | + $ku = utf_8_to_unicode($k); |
|
| 1254 | + $x = str_replace($k, $ku, $x); |
|
| 1255 | + } |
|
| 1256 | + } |
|
| 1257 | + |
|
| 1258 | + return $x; |
|
| 1259 | 1259 | } |