Make WordPress Core

Ticket #15248: allow-turkish-accents-in-slugs.php

File allow-turkish-accents-in-slugs.php, 7.8 KB (added by SergeyBiryukov, 15 years ago)

Allows to keep accented Turkish characters in post and term slugs, as well as in usernames.

Line 
1<?php
2/*
3Plugin Name: Allow Turkish Accents in Slugs
4Plugin URI: http://core.trac.wordpress.org/ticket/15248
5Description: Allows to keep accented Turkish characters in post and term slugs, as well as in usernames.
6Author: Sergey Biryukov
7Author URI: http://profiles.wordpress.org/sergeybiryukov/
8Version: 0.1
9*/
10
11function atas_restore_raw_title($title, $raw_title, $context) {
12 return ( 'save' == $context ) ? $raw_title : $title;
13}
14add_filter('sanitize_title', 'atas_restore_raw_title', 8, 3);
15
16function atas_remove_accents($string, $raw_string = '', $context = 'save') {
17 if ( 'save' != $context || !preg_match('/[\x80-\xff]/', $string) )
18 return $string;
19
20 if (seems_utf8($string)) {
21 $chars = array(
22 // Decompositions for Latin-1 Supplement
23 chr(195).chr(128) => 'A', chr(195).chr(129) => 'A',
24 chr(195).chr(130) => 'A', chr(195).chr(131) => 'A',
25 chr(195).chr(132) => 'A', chr(195).chr(133) => 'A',
26 chr(195).chr(134) => 'AE',chr(195).chr(136) => 'E',
27 chr(195).chr(137) => 'E', chr(195).chr(138) => 'E',
28 chr(195).chr(139) => 'E', chr(195).chr(140) => 'I',
29 chr(195).chr(141) => 'I', chr(195).chr(142) => 'I',
30 chr(195).chr(143) => 'I', chr(195).chr(144) => 'D',
31 chr(195).chr(145) => 'N', chr(195).chr(146) => 'O',
32 chr(195).chr(147) => 'O', chr(195).chr(148) => 'O',
33 chr(195).chr(149) => 'O', chr(195).chr(153) => 'U',
34 chr(195).chr(154) => 'U', chr(195).chr(155) => 'U',
35 chr(195).chr(157) => 'Y', chr(195).chr(158) => 'TH',
36 chr(195).chr(159) => 's', chr(195).chr(160) => 'a',
37 chr(195).chr(161) => 'a', chr(195).chr(162) => 'a',
38 chr(195).chr(163) => 'a', chr(195).chr(164) => 'a',
39 chr(195).chr(165) => 'a', chr(195).chr(166) => 'ae',
40 chr(195).chr(168) => 'e', chr(195).chr(169) => 'e',
41 chr(195).chr(170) => 'e', chr(195).chr(171) => 'e',
42 chr(195).chr(172) => 'i', chr(195).chr(173) => 'i',
43 chr(195).chr(174) => 'i', chr(195).chr(175) => 'i',
44 chr(195).chr(176) => 'd', chr(195).chr(177) => 'n',
45 chr(195).chr(178) => 'o', chr(195).chr(179) => 'o',
46 chr(195).chr(180) => 'o', chr(195).chr(181) => 'o',
47 chr(195).chr(184) => 'o', chr(195).chr(185) => 'u',
48 chr(195).chr(186) => 'u', chr(195).chr(187) => 'u',
49 chr(195).chr(189) => 'y', chr(195).chr(190) => 'th',
50 chr(195).chr(191) => 'y',
51 // Decompositions for Latin Extended-A
52 chr(196).chr(128) => 'A', chr(196).chr(129) => 'a',
53 chr(196).chr(130) => 'A', chr(196).chr(131) => 'a',
54 chr(196).chr(132) => 'A', chr(196).chr(133) => 'a',
55 chr(196).chr(134) => 'C', chr(196).chr(135) => 'c',
56 chr(196).chr(136) => 'C', chr(196).chr(137) => 'c',
57 chr(196).chr(138) => 'C', chr(196).chr(139) => 'c',
58 chr(196).chr(140) => 'C', chr(196).chr(141) => 'c',
59 chr(196).chr(142) => 'D', chr(196).chr(143) => 'd',
60 chr(196).chr(144) => 'D', chr(196).chr(145) => 'd',
61 chr(196).chr(146) => 'E', chr(196).chr(147) => 'e',
62 chr(196).chr(148) => 'E', chr(196).chr(149) => 'e',
63 chr(196).chr(150) => 'E', chr(196).chr(151) => 'e',
64 chr(196).chr(152) => 'E', chr(196).chr(153) => 'e',
65 chr(196).chr(154) => 'E', chr(196).chr(155) => 'e',
66 chr(196).chr(156) => 'G', chr(196).chr(157) => 'g',
67 chr(196).chr(160) => 'G', chr(196).chr(161) => 'g',
68 chr(196).chr(162) => 'G', chr(196).chr(163) => 'g',
69 chr(196).chr(164) => 'H', chr(196).chr(165) => 'h',
70 chr(196).chr(166) => 'H', chr(196).chr(167) => 'h',
71 chr(196).chr(168) => 'I', chr(196).chr(169) => 'i',
72 chr(196).chr(170) => 'I', chr(196).chr(171) => 'i',
73 chr(196).chr(172) => 'I', chr(196).chr(173) => 'i',
74 chr(196).chr(174) => 'I', chr(196).chr(175) => 'i',
75 chr(196).chr(178) => 'IJ',chr(196).chr(179) => 'ij',
76 chr(196).chr(180) => 'J', chr(196).chr(181) => 'j',
77 chr(196).chr(182) => 'K', chr(196).chr(183) => 'k',
78 chr(196).chr(184) => 'k', chr(196).chr(185) => 'L',
79 chr(196).chr(186) => 'l', chr(196).chr(187) => 'L',
80 chr(196).chr(188) => 'l', chr(196).chr(189) => 'L',
81 chr(196).chr(190) => 'l', chr(196).chr(191) => 'L',
82 chr(197).chr(128) => 'l', chr(197).chr(129) => 'L',
83 chr(197).chr(130) => 'l', chr(197).chr(131) => 'N',
84 chr(197).chr(132) => 'n', chr(197).chr(133) => 'N',
85 chr(197).chr(134) => 'n', chr(197).chr(135) => 'N',
86 chr(197).chr(136) => 'n', chr(197).chr(137) => 'N',
87 chr(197).chr(138) => 'n', chr(197).chr(139) => 'N',
88 chr(197).chr(140) => 'O', chr(197).chr(141) => 'o',
89 chr(197).chr(142) => 'O', chr(197).chr(143) => 'o',
90 chr(197).chr(144) => 'O', chr(197).chr(145) => 'o',
91 chr(197).chr(146) => 'OE',chr(197).chr(147) => 'oe',
92 chr(197).chr(148) => 'R',chr(197).chr(149) => 'r',
93 chr(197).chr(150) => 'R',chr(197).chr(151) => 'r',
94 chr(197).chr(152) => 'R',chr(197).chr(153) => 'r',
95 chr(197).chr(154) => 'S',chr(197).chr(155) => 's',
96 chr(197).chr(156) => 'S',chr(197).chr(157) => 's',
97 chr(197).chr(160) => 'S', chr(197).chr(161) => 's',
98 chr(197).chr(162) => 'T', chr(197).chr(163) => 't',
99 chr(197).chr(164) => 'T', chr(197).chr(165) => 't',
100 chr(197).chr(166) => 'T', chr(197).chr(167) => 't',
101 chr(197).chr(168) => 'U', chr(197).chr(169) => 'u',
102 chr(197).chr(170) => 'U', chr(197).chr(171) => 'u',
103 chr(197).chr(172) => 'U', chr(197).chr(173) => 'u',
104 chr(197).chr(174) => 'U', chr(197).chr(175) => 'u',
105 chr(197).chr(176) => 'U', chr(197).chr(177) => 'u',
106 chr(197).chr(178) => 'U', chr(197).chr(179) => 'u',
107 chr(197).chr(180) => 'W', chr(197).chr(181) => 'w',
108 chr(197).chr(182) => 'Y', chr(197).chr(183) => 'y',
109 chr(197).chr(184) => 'Y', chr(197).chr(185) => 'Z',
110 chr(197).chr(186) => 'z', chr(197).chr(187) => 'Z',
111 chr(197).chr(188) => 'z', chr(197).chr(189) => 'Z',
112 chr(197).chr(190) => 'z', chr(197).chr(191) => 's',
113 // Decompositions for Latin Extended-B
114 chr(200).chr(152) => 'S', chr(200).chr(153) => 's',
115 chr(200).chr(154) => 'T', chr(200).chr(155) => 't',
116 // Euro Sign
117 chr(226).chr(130).chr(172) => 'E',
118 // GBP (Pound) Sign
119 chr(194).chr(163) => '');
120
121 $string = strtr($string, $chars);
122 } else {
123 // Assume ISO-8859-1 if not UTF-8
124 $chars['in'] = chr(128).chr(131).chr(138).chr(142).chr(154).chr(158)
125 .chr(159).chr(162).chr(165).chr(181).chr(192).chr(193).chr(194)
126 .chr(195).chr(196).chr(197).chr(199).chr(200).chr(201).chr(202)
127 .chr(203).chr(204).chr(205).chr(206).chr(207).chr(209).chr(210)
128 .chr(211).chr(212).chr(213).chr(214).chr(216).chr(217).chr(218)
129 .chr(219).chr(220).chr(221).chr(224).chr(225).chr(226).chr(227)
130 .chr(228).chr(229).chr(231).chr(232).chr(233).chr(234).chr(235)
131 .chr(236).chr(237).chr(238).chr(239).chr(241).chr(242).chr(243)
132 .chr(244).chr(245).chr(246).chr(248).chr(249).chr(250).chr(251)
133 .chr(252).chr(253).chr(255);
134
135 $chars['out'] = "EfSZszYcYuAAAAAACEEEEIIIINOOOOOOUUUUYaaaaaaceeeeiiiinoooooouuuuyy";
136
137 $string = strtr($string, $chars['in'], $chars['out']);
138 $double_chars['in'] = array(chr(140), chr(156), chr(198), chr(208), chr(222), chr(223), chr(230), chr(240), chr(254));
139 $double_chars['out'] = array('OE', 'oe', 'AE', 'DH', 'TH', 'ss', 'ae', 'dh', 'th');
140 $string = str_replace($double_chars['in'], $double_chars['out'], $string);
141 }
142
143 return $string;
144}
145add_filter('sanitize_title', 'atas_remove_accents', 9, 3);
146
147function atas_sanitize_user($username, $raw_username, $strict) {
148 $username = wp_strip_all_tags( $raw_username );
149 $username = atas_remove_accents( $username );
150 // Kill octets
151 $username = preg_replace( '|%([a-fA-F0-9][a-fA-F0-9])|', '', $username );
152 $username = preg_replace( '/&.+?;/', '', $username ); // Kill entities
153
154 // If strict, reduce to ASCII and Turkish characters for max portability.
155 if ( $strict )
156 $username = preg_replace( '|[^a-zŞşİıĞğÜüÖöÇç0-9 _.\-@]|iu', '', $username );
157
158 $username = trim( $username );
159 // Consolidate contiguous whitespace
160 $username = preg_replace( '|\s+|', ' ', $username );
161
162 return $username;
163}
164add_filter('sanitize_user', 'atas_sanitize_user', 10, 3);
165?>