| 1 |
<?php
|
| 2 |
// $Id: accents.module,v 1.2 2006/08/16 01:50:27 canen Exp $
|
| 3 |
/**
|
| 4 |
* @file
|
| 5 |
* Remove accents from words before searching. Will require a re-indexing.
|
| 6 |
*
|
| 7 |
* The remove_accents and seems_utf8 functions were taken from Wordpress.
|
| 8 |
* They can be found here:
|
| 9 |
* http://trac.wordpress.org/browser/trunk/wp-includes/formatting.php
|
| 10 |
*/
|
| 11 |
|
| 12 |
function accents_search_preprocess($text) {
|
| 13 |
$text = _accents_remove_accents($text);
|
| 14 |
return $text;
|
| 15 |
}
|
| 16 |
|
| 17 |
/**
|
| 18 |
* From Wordpress:
|
| 19 |
* http://trac.wordpress.org/browser/trunk/wp-includes/formatting.php#L86
|
| 20 |
*/
|
| 21 |
function _accents_seems_utf8($str) { // by bmorel at ssi dot fr
|
| 22 |
for ($i=0; $i<strlen($str); $i++) {
|
| 23 |
if (ord($str[$i]) < 0x80) continue; // 0bbbbbbb
|
| 24 |
elseif ((ord($str[$i]) & 0xE0) == 0xC0) $n=1; // 110bbbbb
|
| 25 |
elseif ((ord($str[$i]) & 0xF0) == 0xE0) $n=2; // 1110bbbb
|
| 26 |
elseif ((ord($str[$i]) & 0xF8) == 0xF0) $n=3; // 11110bbb
|
| 27 |
elseif ((ord($str[$i]) & 0xFC) == 0xF8) $n=4; // 111110bb
|
| 28 |
elseif ((ord($str[$i]) & 0xFE) == 0xFC) $n=5; // 1111110b
|
| 29 |
else return false; // Does not match any model
|
| 30 |
for ($j=0; $j<$n; $j++) { // n bytes matching 10bbbbbb follow ?
|
| 31 |
if ((++$i == strlen($str)) || ((ord($str[$i]) & 0xC0) != 0x80))
|
| 32 |
return false;
|
| 33 |
}
|
| 34 |
}
|
| 35 |
return true;
|
| 36 |
}
|
| 37 |
|
| 38 |
/**
|
| 39 |
* From Wordpress:
|
| 40 |
* http://trac.wordpress.org/browser/trunk/wp-includes/formatting.php#L151
|
| 41 |
*/
|
| 42 |
function _accents_remove_accents($string) {
|
| 43 |
if (!preg_match('/[\x80-\xff]/', $string))
|
| 44 |
return $string;
|
| 45 |
|
| 46 |
if (_accents_seems_utf8($string)) {
|
| 47 |
$chars = array(
|
| 48 |
// Decompositions for Latin-1 Supplement
|
| 49 |
chr(195).chr(128) => 'A', chr(195).chr(129) => 'A',
|
| 50 |
chr(195).chr(130) => 'A', chr(195).chr(131) => 'A',
|
| 51 |
chr(195).chr(132) => 'A', chr(195).chr(133) => 'A',
|
| 52 |
chr(195).chr(135) => 'C', chr(195).chr(136) => 'E',
|
| 53 |
chr(195).chr(137) => 'E', chr(195).chr(138) => 'E',
|
| 54 |
chr(195).chr(139) => 'E', chr(195).chr(140) => 'I',
|
| 55 |
chr(195).chr(141) => 'I', chr(195).chr(142) => 'I',
|
| 56 |
chr(195).chr(143) => 'I', chr(195).chr(145) => 'N',
|
| 57 |
chr(195).chr(146) => 'O', chr(195).chr(147) => 'O',
|
| 58 |
chr(195).chr(148) => 'O', chr(195).chr(149) => 'O',
|
| 59 |
chr(195).chr(150) => 'O', chr(195).chr(153) => 'U',
|
| 60 |
chr(195).chr(154) => 'U', chr(195).chr(155) => 'U',
|
| 61 |
chr(195).chr(156) => 'U', chr(195).chr(157) => 'Y',
|
| 62 |
chr(195).chr(159) => 's', chr(195).chr(160) => 'a',
|
| 63 |
chr(195).chr(161) => 'a', chr(195).chr(162) => 'a',
|
| 64 |
chr(195).chr(163) => 'a', chr(195).chr(164) => 'a',
|
| 65 |
chr(195).chr(165) => 'a', chr(195).chr(167) => 'c',
|
| 66 |
chr(195).chr(168) => 'e', chr(195).chr(169) => 'e',
|
| 67 |
chr(195).chr(170) => 'e', chr(195).chr(171) => 'e',
|
| 68 |
chr(195).chr(172) => 'i', chr(195).chr(173) => 'i',
|
| 69 |
chr(195).chr(174) => 'i', chr(195).chr(175) => 'i',
|
| 70 |
chr(195).chr(177) => 'n', chr(195).chr(178) => 'o',
|
| 71 |
chr(195).chr(179) => 'o', chr(195).chr(180) => 'o',
|
| 72 |
chr(195).chr(181) => 'o', chr(195).chr(182) => 'o',
|
| 73 |
chr(195).chr(182) => 'o', chr(195).chr(185) => 'u',
|
| 74 |
chr(195).chr(186) => 'u', chr(195).chr(187) => 'u',
|
| 75 |
chr(195).chr(188) => 'u', chr(195).chr(189) => 'y',
|
| 76 |
chr(195).chr(191) => 'y',
|
| 77 |
// Decompositions for Latin Extended-A
|
| 78 |
chr(196).chr(128) => 'A', chr(196).chr(129) => 'a',
|
| 79 |
chr(196).chr(130) => 'A', chr(196).chr(131) => 'a',
|
| 80 |
chr(196).chr(132) => 'A', chr(196).chr(133) => 'a',
|
| 81 |
chr(196).chr(134) => 'C', chr(196).chr(135) => 'c',
|
| 82 |
chr(196).chr(136) => 'C', chr(196).chr(137) => 'c',
|
| 83 |
chr(196).chr(138) => 'C', chr(196).chr(139) => 'c',
|
| 84 |
chr(196).chr(140) => 'C', chr(196).chr(141) => 'c',
|
| 85 |
chr(196).chr(142) => 'D', chr(196).chr(143) => 'd',
|
| 86 |
chr(196).chr(144) => 'D', chr(196).chr(145) => 'd',
|
| 87 |
chr(196).chr(146) => 'E', chr(196).chr(147) => 'e',
|
| 88 |
chr(196).chr(148) => 'E', chr(196).chr(149) => 'e',
|
| 89 |
chr(196).chr(150) => 'E', chr(196).chr(151) => 'e',
|
| 90 |
chr(196).chr(152) => 'E', chr(196).chr(153) => 'e',
|
| 91 |
chr(196).chr(154) => 'E', chr(196).chr(155) => 'e',
|
| 92 |
chr(196).chr(156) => 'G', chr(196).chr(157) => 'g',
|
| 93 |
chr(196).chr(158) => 'G', chr(196).chr(159) => 'g',
|
| 94 |
chr(196).chr(160) => 'G', chr(196).chr(161) => 'g',
|
| 95 |
chr(196).chr(162) => 'G', chr(196).chr(163) => 'g',
|
| 96 |
chr(196).chr(164) => 'H', chr(196).chr(165) => 'h',
|
| 97 |
chr(196).chr(166) => 'H', chr(196).chr(167) => 'h',
|
| 98 |
chr(196).chr(168) => 'I', chr(196).chr(169) => 'i',
|
| 99 |
chr(196).chr(170) => 'I', chr(196).chr(171) => 'i',
|
| 100 |
chr(196).chr(172) => 'I', chr(196).chr(173) => 'i',
|
| 101 |
chr(196).chr(174) => 'I', chr(196).chr(175) => 'i',
|
| 102 |
chr(196).chr(176) => 'I', chr(196).chr(177) => 'i',
|
| 103 |
chr(196).chr(178) => 'IJ',chr(196).chr(179) => 'ij',
|
| 104 |
chr(196).chr(180) => 'J', chr(196).chr(181) => 'j',
|
| 105 |
chr(196).chr(182) => 'K', chr(196).chr(183) => 'k',
|
| 106 |
chr(196).chr(184) => 'k', chr(196).chr(185) => 'L',
|
| 107 |
chr(196).chr(186) => 'l', chr(196).chr(187) => 'L',
|
| 108 |
chr(196).chr(188) => 'l', chr(196).chr(189) => 'L',
|
| 109 |
chr(196).chr(190) => 'l', chr(196).chr(191) => 'L',
|
| 110 |
chr(197).chr(128) => 'l', chr(197).chr(129) => 'L',
|
| 111 |
chr(197).chr(130) => 'l', chr(197).chr(131) => 'N',
|
| 112 |
chr(197).chr(132) => 'n', chr(197).chr(133) => 'N',
|
| 113 |
chr(197).chr(134) => 'n', chr(197).chr(135) => 'N',
|
| 114 |
chr(197).chr(136) => 'n', chr(197).chr(137) => 'N',
|
| 115 |
chr(197).chr(138) => 'n', chr(197).chr(139) => 'N',
|
| 116 |
chr(197).chr(140) => 'O', chr(197).chr(141) => 'o',
|
| 117 |
chr(197).chr(142) => 'O', chr(197).chr(143) => 'o',
|
| 118 |
chr(197).chr(144) => 'O', chr(197).chr(145) => 'o',
|
| 119 |
chr(197).chr(146) => 'OE',chr(197).chr(147) => 'oe',
|
| 120 |
chr(197).chr(148) => 'R',chr(197).chr(149) => 'r',
|
| 121 |
chr(197).chr(150) => 'R',chr(197).chr(151) => 'r',
|
| 122 |
chr(197).chr(152) => 'R',chr(197).chr(153) => 'r',
|
| 123 |
chr(197).chr(154) => 'S',chr(197).chr(155) => 's',
|
| 124 |
chr(197).chr(156) => 'S',chr(197).chr(157) => 's',
|
| 125 |
chr(197).chr(158) => 'S',chr(197).chr(159) => 's',
|
| 126 |
chr(197).chr(160) => 'S', chr(197).chr(161) => 's',
|
| 127 |
chr(197).chr(162) => 'T', chr(197).chr(163) => 't',
|
| 128 |
chr(197).chr(164) => 'T', chr(197).chr(165) => 't',
|
| 129 |
chr(197).chr(166) => 'T', chr(197).chr(167) => 't',
|
| 130 |
chr(197).chr(168) => 'U', chr(197).chr(169) => 'u',
|
| 131 |
chr(197).chr(170) => 'U', chr(197).chr(171) => 'u',
|
| 132 |
chr(197).chr(172) => 'U', chr(197).chr(173) => 'u',
|
| 133 |
chr(197).chr(174) => 'U', chr(197).chr(175) => 'u',
|
| 134 |
chr(197).chr(176) => 'U', chr(197).chr(177) => 'u',
|
| 135 |
chr(197).chr(178) => 'U', chr(197).chr(179) => 'u',
|
| 136 |
chr(197).chr(180) => 'W', chr(197).chr(181) => 'w',
|
| 137 |
chr(197).chr(182) => 'Y', chr(197).chr(183) => 'y',
|
| 138 |
chr(197).chr(184) => 'Y', chr(197).chr(185) => 'Z',
|
| 139 |
chr(197).chr(186) => 'z', chr(197).chr(187) => 'Z',
|
| 140 |
chr(197).chr(188) => 'z', chr(197).chr(189) => 'Z',
|
| 141 |
chr(197).chr(190) => 'z', chr(197).chr(191) => 's',
|
| 142 |
// Euro Sign
|
| 143 |
chr(226).chr(130).chr(172) => 'E'
|
| 144 |
);
|
| 145 |
|
| 146 |
$string = strtr($string, $chars);
|
| 147 |
}
|
| 148 |
else {
|
| 149 |
// Assume ISO-8859-1 if not UTF-8
|
| 150 |
$chars['in'] = chr(128).chr(131).chr(138).chr(142).chr(154).chr(158)
|
| 151 |
.chr(159).chr(162).chr(165).chr(181).chr(192).chr(193).chr(194)
|
| 152 |
.chr(195).chr(196).chr(197).chr(199).chr(200).chr(201).chr(202)
|
| 153 |
.chr(203).chr(204).chr(205).chr(206).chr(207).chr(209).chr(210)
|
| 154 |
.chr(211).chr(212).chr(213).chr(214).chr(216).chr(217).chr(218)
|
| 155 |
.chr(219).chr(220).chr(221).chr(224).chr(225).chr(226).chr(227)
|
| 156 |
.chr(228).chr(229).chr(231).chr(232).chr(233).chr(234).chr(235)
|
| 157 |
.chr(236).chr(237).chr(238).chr(239).chr(241).chr(242).chr(243)
|
| 158 |
.chr(244).chr(245).chr(246).chr(248).chr(249).chr(250).chr(251)
|
| 159 |
.chr(252).chr(253).chr(255);
|
| 160 |
|
| 161 |
$chars['out'] = "EfSZszYcYuAAAAAACEEEEIIIINOOOOOOUUUUYaaaaaaceeeeiiiinoooooouuuuyy";
|
| 162 |
|
| 163 |
$string = strtr($string, $chars['in'], $chars['out']);
|
| 164 |
$double_chars['in'] = array(chr(140), chr(156), chr(198), chr(208), chr(222), chr(223), chr(230), chr(240), chr(254));
|
| 165 |
$double_chars['out'] = array('OE', 'oe', 'AE', 'DH', 'TH', 'ss', 'ae', 'dh', 'th');
|
| 166 |
$string = str_replace($double_chars['in'], $double_chars['out'], $string);
|
| 167 |
}
|
| 168 |
|
| 169 |
return $string;
|
| 170 |
}
|
| 171 |
|