[*] BO : improved unicode characters replacement in URLs

This commit is contained in:
Damien Metzger
2013-06-28 12:19:02 +02:00
parent 97370946d2
commit 42f8a0a67a
2 changed files with 102 additions and 54 deletions
+54 -35
View File
@@ -1025,47 +1025,66 @@ class ToolsCore
*/
public static function replaceAccentedChars($str)
{
/* One source among others:
http://www.tachyonsoft.com/uc0000.htm
http://www.tachyonsoft.com/uc0001.htm
*/
$patterns = array(
/* Lowercase */
'/[\x{0105}\x{00E0}\x{00E1}\x{00E2}\x{00E3}\x{00E4}\x{00E5}]/u',
'/[\x{00E7}\x{010D}\x{0107}]/u',
'/[\x{010F}]/u',
'/[\x{00E8}\x{00E9}\x{00EA}\x{00EB}\x{011B}\x{0119}]/u',
'/[\x{00EC}\x{00ED}\x{00EE}\x{00EF}]/u',
'/[\x{011F}]/u',
'/[\x{0142}\x{013E}\x{013A}]/u',
'/[\x{00F1}\x{0148}]/u',
'/[\x{00F2}\x{00F3}\x{00F4}\x{00F5}\x{00F6}\x{00F8}]/u',
'/[\x{0159}\x{0155}]/u',
'/[\x{015B}\x{0161}\x{015F}]/u',
'/[\x{00DF}]/u',
'/[\x{0165}]/u',
'/[\x{00F9}\x{00FA}\x{00FB}\x{00FC}\x{016F}]/u',
'/[\x{00FD}\x{00FF}]/u',
'/[\x{017C}\x{017A}\x{017E}]/u',
'/[\x{00E6}]/u',
'/[\x{0153}]/u',
/* a */ '/[\x{00E0}\x{00E1}\x{00E2}\x{00E3}\x{00E4}\x{00E5}\x{0101}\x{0103}\x{0105}]/u',
/* c */ '/[\x{00E7}\x{0107}\x{0109}\x{010D}]/u',
/* d */ '/[\x{010F}\x{0111}]/u',
/* e */ '/[\x{00E8}\x{00E9}\x{00EA}\x{00EB}\x{0113}\x{0115}\x{0117}\x{0119}\x{011B}]/u',
/* g */ '/[\x{011F}\x{0121}\x{0123}]/u',
/* h */ '/[\x{0125}\x{0127}]/u',
/* i */ '/[\x{00EC}\x{00ED}\x{00EE}\x{00EF}\x{0129}\x{012B}\x{012D}\x{012F}\x{0131}]/u',
/* j */ '/[\x{0135}]/u',
/* k */ '/[\x{0137}\x{0138}]/u',
/* l */ '/[\x{013A}\x{013C}\x{013E}\x{0140}\x{0142}]/u',
/* n */ '/[\x{00F1}\x{0144}\x{0146}\x{0148}\x{0149}\x{014B}]/u',
/* o */ '/[\x{00F2}\x{00F3}\x{00F4}\x{00F5}\x{00F6}\x{00F8}\x{014D}\x{014F}\x{0151}]/u',
/* r */ '/[\x{0155}\x{0157}\x{0159}]/u',
/* s */ '/[\x{015B}\x{015D}\x{015F}\x{0161}]/u',
/* ss*/ '/[\x{00DF}]/u',
/* t */ '/[\x{0163}\x{0165}\x{0167}]/u',
/* u */ '/[\x{00F9}\x{00FA}\x{00FB}\x{00FC}\x{0169}\x{016B}\x{016D}\x{016F}\x{0171}\x{0173}]/u',
/* w */ '/[\x{0175}]/u',
/* y */ '/[\x{00FF}\x{0177}\x{00FD}]/u',
/* z */ '/[\x{017A}\x{017C}\x{017E}]/u',
/* ae*/ '/[\x{00E6}]/u',
/* oe*/ '/[\x{0153}]/u',
/* Uppercase */
'/[\x{0104}\x{00C0}\x{00C1}\x{00C2}\x{00C3}\x{00C4}\x{00C5}]/u',
'/[\x{00C7}\x{010C}\x{0106}]/u',
'/[\x{010E}]/u',
'/[\x{00C8}\x{00C9}\x{00CA}\x{00CB}\x{011A}\x{0118}]/u',
'/[\x{011E}]/u',
'/[\x{0141}\x{013D}\x{0139}]/u',
'/[\x{00D1}\x{0147}]/u',
'/[\x{00D3}]/u',
'/[\x{0158}\x{0154}]/u',
'/[\x{015A}\x{0160}\x{015E}]/u',
'/[\x{0164}]/u',
'/[\x{00D9}\x{00DA}\x{00DB}\x{00DC}\x{016E}]/u',
'/[\x{017B}\x{0179}\x{017D}]/u',
'/[\x{00C6}]/u',
'/[\x{0152}]/u');
/* A */ '/[\x{0100}\x{0102}\x{0104}\x{00C0}\x{00C1}\x{00C2}\x{00C3}\x{00C4}\x{00C5}]/u',
/* C */ '/[\x{00C7}\x{0106}\x{0108}\x{010A}\x{010C}]/u',
/* D */ '/[\x{010E}\x{0110}]/u',
/* E */ '/[\x{00C8}\x{00C9}\x{00CA}\x{00CB}\x{0112}\x{0114}\x{0116}\x{0118}\x{011A}]/u',
/* G */ '/[\x{011C}\x{011E}\x{0120}\x{0122}]/u',
/* H */ '/[\x{0124}\x{0126}]/u',
/* I */ '/[\x{0128}\x{012A}\x{012C}\x{012E}\x{0130}]/u',
/* J */ '/[\x{0134}]/u',
/* K */ '/[\x{0136}]/u',
/* L */ '/[\x{0139}\x{013B}\x{013D}\x{0139}\x{0141}]/u',
/* N */ '/[\x{00D1}\x{0143}\x{0145}\x{0147}\x{014A}]/u',
/* O */ '/[\x{00D3}\x{014C}\x{014E}\x{0150}]/u',
/* R */ '/[\x{0154}\x{0156}\x{0158}]/u',
/* S */ '/[\x{015A}\x{015C}\x{015E}\x{0160}]/u',
/* T */ '/[\x{0162}\x{0164}\x{0166}]/u',
/* U */ '/[\x{00D9}\x{00DA}\x{00DB}\x{00DC}\x{0168}\x{016A}\x{016C}\x{016E}\x{0170}\x{0172}]/u',
/* W */ '/[\x{0174}]/u',
/* Y */ '/[\x{0176}]/u',
/* Z */ '/[\x{0179}\x{017B}\x{017D}]/u',
/* AE*/ '/[\x{00C6}]/u',
/* OE*/ '/[\x{0152}]/u');
// ö to oe
// å to aa
// ä to ae
$replacements = array(
'a', 'c', 'd', 'e', 'i', 'g', 'l', 'n', 'o', 'r', 's', 'ss', 't', 'u', 'y', 'z', 'ae', 'oe',
'A', 'C', 'D', 'E', 'G', 'L', 'N', 'O', 'R', 'S', 'T', 'U', 'Z', 'AE', 'OE'
'a', 'c', 'd', 'e', 'g', 'h', 'i', 'j', 'k', 'l', 'n', 'o', 'r', 's', 'ss', 't', 'u', 'y', 'w', 'z', 'ae', 'oe',
'A', 'C', 'D', 'E', 'G', 'H', 'I', 'J', 'K', 'L', 'N', 'O', 'R', 'S', 'T', 'U', 'Z', 'AE', 'OE'
);
return preg_replace($patterns, $replacements, $str);
+48 -19
View File
@@ -35,25 +35,54 @@ function str2url(str, encoding, ucfirst)
str = str.replace(/[^a-z0-9\s\'\:\/\[\]-]\\u00A1-\\uFFFF/g,'');
else
{
str = str.replace(/[\u0105\u0104\u00E0\u00E1\u00E2\u00E3\u00E4\u00E5]/g,'a');
str = str.replace(/[\u00E7\u010D\u0107\u0106]/g,'c');
str = str.replace(/[\u010F]/g,'d');
str = str.replace(/[\u00E8\u00E9\u00EA\u00EB\u011B\u0119\u0118\u0117]/g,'e');
str = str.replace(/[\u00EC\u00ED\u00EE\u00EF\u012F]/g,'i');
str = str.replace(/[\u0142\u0141]/g,'l');
str = str.replace(/[\u00F1\u0148]/g,'n');
str = str.replace(/[\u00F2\u00F3\u00F4\u00F5\u00F6\u00F8\u00D3]/g,'o');
str = str.replace(/[\u0159]/g,'r');
str = str.replace(/[\u015B\u015A\u0161]/g,'s');
str = str.replace(/[\u00DF]/g,'ss');
str = str.replace(/[\u0165]/g,'t');
str = str.replace(/[\u00F9\u00FA\u00FB\u00FC\u016F\u016B\u0173]/g,'u');
str = str.replace(/[\u00FD\u00FF]/g,'y');
str = str.replace(/[\u017C\u017A\u017B\u0179\u017E]/g,'z');
str = str.replace(/[\u00E6]/g,'ae');
str = str.replace(/[\u0153]/g,'oe');
str = str.replace(/[\u013E\u013A]/g,'l');
str = str.replace(/[\u0155]/g,'r');
/* Lowercase */
str = str.replace(/[\u00E0\u00E1\u00E2\u00E3\u00E4\u00E5\u0101\u0103\u0105]/g, 'a');
str = str.replace(/[\u00E7\u0107\u0109\u010D]/g, 'c');
str = str.replace(/[\u010F\u0111]/g, 'd');
str = str.replace(/[\u00E8\u00E9\u00EA\u00EB\u0113\u0115\u0117\u0119\u011B]/g, 'e');
str = str.replace(/[\u011F\u0121\u0123]/g, 'g');
str = str.replace(/[\u0125\u0127]/g, 'h');
str = str.replace(/[\u00EC\u00ED\u00EE\u00EF\u0129\u012B\u012D\u012F\u0131]/g, 'i');
str = str.replace(/[\u0135]/g, 'j');
str = str.replace(/[\u0137\u0138]/g, 'k');
str = str.replace(/[\u013A\u013C\u013E\u0140\u0142]/g, 'l');
str = str.replace(/[\u00F1\u0144\u0146\u0148\u0149\u014B]/g, 'n');
str = str.replace(/[\u00F2\u00F3\u00F4\u00F5\u00F6\u00F8\u014D\u014F\u0151]/g, 'o');
str = str.replace(/[\u0155\u0157\u0159]/g, 'r');
str = str.replace(/[\u015B\u015D\u015F\u0161]/g, 's');
str = str.replace(/[\u00DF]/g, 'ss');
str = str.replace(/[\u0163\u0165\u0167]/g, 't');
str = str.replace(/[\u00F9\u00FA\u00FB\u00FC\u0169\u016B\u016D\u016F\u0171\u0173]/g, 'u');
str = str.replace(/[\u0175]/g, 'w');
str = str.replace(/[\u00FF\u0177\u00FD]/g, 'y');
str = str.replace(/[\u017A\u017C\u017E]/g, 'z');
str = str.replace(/[\u00E6]/g, 'ae');
str = str.replace(/[\u0153]/g, 'oe');
/* Uppercase */
str = str.replace(/[\u0100\u0102\u0104\u00C0\u00C1\u00C2\u00C3\u00C4\u00C5]/g, 'A');
str = str.replace(/[\u00C7\u0106\u0108\u010A\u010C]/g, 'C');
str = str.replace(/[\u010E\u0110]/g, 'D');
str = str.replace(/[\u00C8\u00C9\u00CA\u00CB\u0112\u0114\u0116\u0118\u011A]/g, 'E');
str = str.replace(/[\u011C\u011E\u0120\u0122]/g, 'G');
str = str.replace(/[\u0124\u0126]/g, 'H');
str = str.replace(/[\u0128\u012A\u012C\u012E\u0130]/g, 'I');
str = str.replace(/[\u0134]/g, 'J');
str = str.replace(/[\u0136]/g, 'K');
str = str.replace(/[\u0139\u013B\u013D\u0139\u0141]/g, 'L');
str = str.replace(/[\u00D1\u0143\u0145\u0147\u014A]/g, 'N');
str = str.replace(/[\u00D3\u014C\u014E\u0150]/g, 'O');
str = str.replace(/[\u0154\u0156\u0158]/g, 'R');
str = str.replace(/[\u015A\u015C\u015E\u0160]/g, 'S');
str = str.replace(/[\u0162\u0164\u0166]/g, 'T');
str = str.replace(/[\u00D9\u00DA\u00DB\u00DC\u0168\u016A\u016C\u016E\u0170\u0172]/g, 'U');
str = str.replace(/[\u0174]/g, 'W');
str = str.replace(/[\u0176]/g, 'Y');
str = str.replace(/[\u0179\u017B\u017D]/g, 'Z');
str = str.replace(/[\u00C6]/g, 'AE');
str = str.replace(/[\u0152]/g, 'OE');
str = str.toLowerCase();
str = str.replace(/[^a-z0-9\s\'\:\/\[\]-]/g,'');
}