OssnTranslit.php
Go to the documentation of this file.
1 <?php
2 /**
3  * Open Source Social Network
4  *
5  * @package (Informatikon.com).ossn
6  * @author OSSN Core Team <[email protected]>
7  *
8  * For creating "friendly titles" for URLs
9  *
10  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
11  * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
12  * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
13  * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
14  * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
15  * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
16  * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
17  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
18  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
19  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
20  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
21  *
22  * This software consists of voluntary contributions made by many individuals
23  * and is licensed under the LGPL. For more information, see
24  * <http://www.doctrine-project.org>.
25  *
26  * @package
27  * @author Elgg <http://elgg.org>
28  * @author Konsta Vesterinen <[email protected]>
29  * @author Jonathan H. Wage <[email protected]>
30  * @author Steve Clay <[email protected]>
31  *
32  * @access private Components authors should not use this directly
33  */
34 class OssnTranslit {
35 
36  /**
37  * Create a version of a string for embedding in a URL
38  *
39  * @param string $string A UTF-8 string
40  * @param string $separator The character to separate words with
41  * @return string
42  */
43  static public function urlize($string, $separator = '-') {
44  // Iñtërnâtiônàlizætiøn, AND 日本語!
45 
46  // try to force combined chars because the translit map and others expect it
47  if (self::hasNormalizerSupport()) {
48  $nfc = normalizer_normalize($string);
49  if (is_string($nfc)) {
50  $string = $nfc;
51  }
52  }
53  // Internationalization, AND 日本語!
54  $string = self::transliterateAscii($string);
55 
56  // allow HTML tags in titles
57  $string = preg_replace('~<([a-zA-Z][^>]*)>~', ' $1 ', $string);
58 
59  // more substitutions
60  // @todo put these somewhere else
61  $string = strtr($string, array(
62  // currency
63  "\xE2\x82\xAC" /* € */ => ' E ',
64  "\xC2\xA3" /* £ */ => ' GBP ',
65  ));
66 
67  // remove all ASCII except 0-9a-zA-Z, hyphen, underscore, and whitespace
68  // note: "x" modifier did not work with this pattern.
69  $string = preg_replace('~['
70  . '\x00-\x08' // control chars
71  . '\x0b\x0c' // vert tab, form feed
72  . '\x0e-\x1f' // control chars
73  . '\x21-\x2c' // ! ... ,
74  . '\x2e\x2f' // . slash
75  . '\x3a-\x40' // : ... @
76  . '\x5b-\x5e' // [ ... ^
77  . '\x60' // `
78  . '\x7b-\x7f' // { ... DEL
79  . ']~', '', $string);
80  $string = strtr($string, '', '');
81 
82  // internationalization, and 日本語!
83  // note: not using elgg_strtolower to keep this class portable
84  $string = is_callable('mb_strtolower') ? mb_strtolower($string, 'UTF-8') : strtolower($string);
85 
86  // split by ASCII chars not in 0-9a-zA-Z
87  // note: we cannot use [^0-9a-zA-Z] because that matches multibyte chars.
88  // note: "x" modifier did not work with this pattern.
89  $pattern = '~['
90  . '\x00-\x2f' // controls ... slash
91  . '\x3a-\x40' // : ... @
92  . '\x5b-\x60' // [ ... `
93  . '\x7b-\x7f' // { ... DEL
94  . ']+~x';
95 
96  // ['internationalization', 'and', '日本語']
97  $words = preg_split($pattern, $string, -1, PREG_SPLIT_NO_EMPTY);
98 
99  // ['internationalization', 'and', '%E6%97%A5%E6%9C%AC%E8%AA%9E']
100  $words = array_map('urlencode', $words);
101 
102  // internationalization-and-%E6%97%A5%E6%9C%AC%E8%AA%9E
103  return implode($separator, $words);
104  }
105 
106  /**
107  * Transliterate Western multibyte chars to ASCII
108  *
109  * @param string $utf8 a UTF-8 string
110  * @return string
111  */
112  static public function transliterateAscii($utf8) {
113  static $map = null;
114  if (!preg_match('/[\x80-\xff]/', $utf8)) {
115  return $utf8;
116  }
117  if (null === $map) {
118  $map = self::getAsciiTranslitMap();
119  }
120  return strtr($utf8, $map);
121  }
122 
123  /**
124  * Get array of UTF-8 (NFC) character replacements.
125  *
126  * @return array
127  */
128  static public function getAsciiTranslitMap() {
129  return array(
130  // Decompositions for Latin-1 Supplement
131  "\xC2\xAA" /* ª */ => 'a', "\xC2\xBA" /* º */ => 'o', "\xC3\x80" /* À */ => 'A',
132  "\xC3\x81" /* Á */ => 'A', "\xC3\x82" /* Â */ => 'A', "\xC3\x83" /* Ã */ => 'A',
133  "\xC3\x84" /* Ä */ => 'A', "\xC3\x85" /* Å */ => 'A', "\xC3\x86" /* Æ */ => 'AE',
134  "\xC3\x87" /* Ç */ => 'C', "\xC3\x88" /* È */ => 'E', "\xC3\x89" /* É */ => 'E',
135  "\xC3\x8A" /* Ê */ => 'E', "\xC3\x8B" /* Ë */ => 'E', "\xC3\x8C" /* Ì */ => 'I',
136  "\xC3\x8D" /* Í */ => 'I', "\xC3\x8E" /* Î */ => 'I', "\xC3\x8F" /* Ï */ => 'I',
137  "\xC3\x90" /* Ð */ => 'D', "\xC3\x91" /* Ñ */ => 'N', "\xC3\x92" /* Ò */ => 'O',
138  "\xC3\x93" /* Ó */ => 'O', "\xC3\x94" /* Ô */ => 'O', "\xC3\x95" /* Õ */ => 'O',
139  "\xC3\x96" /* Ö */ => 'O', "\xC3\x99" /* Ù */ => 'U', "\xC3\x9A" /* Ú */ => 'U',
140  "\xC3\x9B" /* Û */ => 'U', "\xC3\x9C" /* Ü */ => 'U', "\xC3\x9D" /* Ý */ => 'Y',
141  "\xC3\x9E" /* Þ */ => 'TH', "\xC3\x9F" /* ß */ => 'ss', "\xC3\xA0" /* à */ => 'a',
142  "\xC3\xA1" /* á */ => 'a', "\xC3\xA2" /* â */ => 'a', "\xC3\xA3" /* ã */ => 'a',
143  "\xC3\xA4" /* ä */ => 'a', "\xC3\xA5" /* å */ => 'a', "\xC3\xA6" /* æ */ => 'ae',
144  "\xC3\xA7" /* ç */ => 'c', "\xC3\xA8" /* è */ => 'e', "\xC3\xA9" /* é */ => 'e',
145  "\xC3\xAA" /* ê */ => 'e', "\xC3\xAB" /* ë */ => 'e', "\xC3\xAC" /* ì */ => 'i',
146  "\xC3\xAD" /* í */ => 'i', "\xC3\xAE" /* î */ => 'i', "\xC3\xAF" /* ï */ => 'i',
147  "\xC3\xB0" /* ð */ => 'd', "\xC3\xB1" /* ñ */ => 'n', "\xC3\xB2" /* ò */ => 'o',
148  "\xC3\xB3" /* ó */ => 'o', "\xC3\xB4" /* ô */ => 'o', "\xC3\xB5" /* õ */ => 'o',
149  "\xC3\xB6" /* ö */ => 'o', "\xC3\xB8" /* ø */ => 'o', "\xC3\xB9" /* ù */ => 'u',
150  "\xC3\xBA" /* ú */ => 'u', "\xC3\xBB" /* û */ => 'u', "\xC3\xBC" /* ü */ => 'u',
151  "\xC3\xBD" /* ý */ => 'y', "\xC3\xBE" /* þ */ => 'th', "\xC3\xBF" /* ÿ */ => 'y',
152  "\xC3\x98" /* Ø */ => 'O',
153  // Decompositions for Latin Extended-A
154  "\xC4\x80" /* Ā */ => 'A', "\xC4\x81" /* ā */ => 'a', "\xC4\x82" /* Ă */ => 'A',
155  "\xC4\x83" /* ă */ => 'a', "\xC4\x84" /* Ą */ => 'A', "\xC4\x85" /* ą */ => 'a',
156  "\xC4\x86" /* Ć */ => 'C', "\xC4\x87" /* ć */ => 'c', "\xC4\x88" /* Ĉ */ => 'C',
157  "\xC4\x89" /* ĉ */ => 'c', "\xC4\x8A" /* Ċ */ => 'C', "\xC4\x8B" /* ċ */ => 'c',
158  "\xC4\x8C" /* Č */ => 'C', "\xC4\x8D" /* č */ => 'c', "\xC4\x8E" /* Ď */ => 'D',
159  "\xC4\x8F" /* ď */ => 'd', "\xC4\x90" /* Đ */ => 'D', "\xC4\x91" /* đ */ => 'd',
160  "\xC4\x92" /* Ē */ => 'E', "\xC4\x93" /* ē */ => 'e', "\xC4\x94" /* Ĕ */ => 'E',
161  "\xC4\x95" /* ĕ */ => 'e', "\xC4\x96" /* Ė */ => 'E', "\xC4\x97" /* ė */ => 'e',
162  "\xC4\x98" /* Ę */ => 'E', "\xC4\x99" /* ę */ => 'e', "\xC4\x9A" /* Ě */ => 'E',
163  "\xC4\x9B" /* ě */ => 'e', "\xC4\x9C" /* Ĝ */ => 'G', "\xC4\x9D" /* ĝ */ => 'g',
164  "\xC4\x9E" /* Ğ */ => 'G', "\xC4\x9F" /* ğ */ => 'g', "\xC4\xA0" /* Ġ */ => 'G',
165  "\xC4\xA1" /* ġ */ => 'g', "\xC4\xA2" /* Ģ */ => 'G', "\xC4\xA3" /* ģ */ => 'g',
166  "\xC4\xA4" /* Ĥ */ => 'H', "\xC4\xA5" /* ĥ */ => 'h', "\xC4\xA6" /* Ħ */ => 'H',
167  "\xC4\xA7" /* ħ */ => 'h', "\xC4\xA8" /* Ĩ */ => 'I', "\xC4\xA9" /* ĩ */ => 'i',
168  "\xC4\xAA" /* Ī */ => 'I', "\xC4\xAB" /* ī */ => 'i', "\xC4\xAC" /* Ĭ */ => 'I',
169  "\xC4\xAD" /* ĭ */ => 'i', "\xC4\xAE" /* Į */ => 'I', "\xC4\xAF" /* į */ => 'i',
170  "\xC4\xB0" /* İ */ => 'I', "\xC4\xB1" /* ı */ => 'i', "\xC4\xB2" /* IJ */ => 'IJ',
171  "\xC4\xB3" /* ij */ => 'ij', "\xC4\xB4" /* Ĵ */ => 'J', "\xC4\xB5" /* ĵ */ => 'j',
172  "\xC4\xB6" /* Ķ */ => 'K', "\xC4\xB7" /* ķ */ => 'k', "\xC4\xB8" /* ĸ */ => 'k',
173  "\xC4\xB9" /* Ĺ */ => 'L', "\xC4\xBA" /* ĺ */ => 'l', "\xC4\xBB" /* Ļ */ => 'L',
174  "\xC4\xBC" /* ļ */ => 'l', "\xC4\xBD" /* Ľ */ => 'L', "\xC4\xBE" /* ľ */ => 'l',
175  "\xC4\xBF" /* Ŀ */ => 'L', "\xC5\x80" /* ŀ */ => 'l', "\xC5\x81" /* Ł */ => 'L',
176  "\xC5\x82" /* ł */ => 'l', "\xC5\x83" /* Ń */ => 'N', "\xC5\x84" /* ń */ => 'n',
177  "\xC5\x85" /* Ņ */ => 'N', "\xC5\x86" /* ņ */ => 'n', "\xC5\x87" /* Ň */ => 'N',
178  "\xC5\x88" /* ň */ => 'n', "\xC5\x89" /* ʼn */ => 'N', "\xC5\x8A" /* Ŋ */ => 'n',
179  "\xC5\x8B" /* ŋ */ => 'N', "\xC5\x8C" /* Ō */ => 'O', "\xC5\x8D" /* ō */ => 'o',
180  "\xC5\x8E" /* Ŏ */ => 'O', "\xC5\x8F" /* ŏ */ => 'o', "\xC5\x90" /* Ő */ => 'O',
181  "\xC5\x91" /* ő */ => 'o', "\xC5\x92" /* Œ */ => 'OE', "\xC5\x93" /* œ */ => 'oe',
182  "\xC5\x94" /* Ŕ */ => 'R', "\xC5\x95" /* ŕ */ => 'r', "\xC5\x96" /* Ŗ */ => 'R',
183  "\xC5\x97" /* ŗ */ => 'r', "\xC5\x98" /* Ř */ => 'R', "\xC5\x99" /* ř */ => 'r',
184  "\xC5\x9A" /* Ś */ => 'S', "\xC5\x9B" /* ś */ => 's', "\xC5\x9C" /* Ŝ */ => 'S',
185  "\xC5\x9D" /* ŝ */ => 's', "\xC5\x9E" /* Ş */ => 'S', "\xC5\x9F" /* ş */ => 's',
186  "\xC5\xA0" /* Š */ => 'S', "\xC5\xA1" /* š */ => 's', "\xC5\xA2" /* Ţ */ => 'T',
187  "\xC5\xA3" /* ţ */ => 't', "\xC5\xA4" /* Ť */ => 'T', "\xC5\xA5" /* ť */ => 't',
188  "\xC5\xA6" /* Ŧ */ => 'T', "\xC5\xA7" /* ŧ */ => 't', "\xC5\xA8" /* Ũ */ => 'U',
189  "\xC5\xA9" /* ũ */ => 'u', "\xC5\xAA" /* Ū */ => 'U', "\xC5\xAB" /* ū */ => 'u',
190  "\xC5\xAC" /* Ŭ */ => 'U', "\xC5\xAD" /* ŭ */ => 'u', "\xC5\xAE" /* Ů */ => 'U',
191  "\xC5\xAF" /* ů */ => 'u', "\xC5\xB0" /* Ű */ => 'U', "\xC5\xB1" /* ű */ => 'u',
192  "\xC5\xB2" /* Ų */ => 'U', "\xC5\xB3" /* ų */ => 'u', "\xC5\xB4" /* Ŵ */ => 'W',
193  "\xC5\xB5" /* ŵ */ => 'w', "\xC5\xB6" /* Ŷ */ => 'Y', "\xC5\xB7" /* ŷ */ => 'y',
194  "\xC5\xB8" /* Ÿ */ => 'Y', "\xC5\xB9" /* Ź */ => 'Z', "\xC5\xBA" /* ź */ => 'z',
195  "\xC5\xBB" /* Ż */ => 'Z', "\xC5\xBC" /* ż */ => 'z', "\xC5\xBD" /* Ž */ => 'Z',
196  "\xC5\xBE" /* ž */ => 'z', "\xC5\xBF" /* ſ */ => 's',
197  // Decompositions for Latin Extended-B
198  "\xC8\x98" /* Ș */ => 'S', "\xC8\x99" /* ș */ => 's',
199  "\xC8\x9A" /* Ț */ => 'T', "\xC8\x9B" /* ț */ => 't',
200  // unmarked
201  "\xC6\xA0" /* Ơ */ => 'O', "\xC6\xA1" /* ơ */ => 'o',
202  "\xC6\xAF" /* Ư */ => 'U', "\xC6\xB0" /* ư */ => 'u',
203  // grave accent
204  "\xE1\xBA\xA6" /* Ầ */ => 'A', "\xE1\xBA\xA7" /* ầ */ => 'a',
205  "\xE1\xBA\xB0" /* Ằ */ => 'A', "\xE1\xBA\xB1" /* ằ */ => 'a',
206  "\xE1\xBB\x80" /* Ề */ => 'E', "\xE1\xBB\x81" /* ề */ => 'e',
207  "\xE1\xBB\x92" /* Ồ */ => 'O', "\xE1\xBB\x93" /* ồ */ => 'o',
208  "\xE1\xBB\x9C" /* Ờ */ => 'O', "\xE1\xBB\x9D" /* ờ */ => 'o',
209  "\xE1\xBB\xAA" /* Ừ */ => 'U', "\xE1\xBB\xAB" /* ừ */ => 'u',
210  "\xE1\xBB\xB2" /* Ỳ */ => 'Y', "\xE1\xBB\xB3" /* ỳ */ => 'y',
211  // hook
212  "\xE1\xBA\xA2" /* Ả */ => 'A', "\xE1\xBA\xA3" /* ả */ => 'a',
213  "\xE1\xBA\xA8" /* Ẩ */ => 'A', "\xE1\xBA\xA9" /* ẩ */ => 'a',
214  "\xE1\xBA\xB2" /* Ẳ */ => 'A', "\xE1\xBA\xB3" /* ẳ */ => 'a',
215  "\xE1\xBA\xBA" /* Ẻ */ => 'E', "\xE1\xBA\xBB" /* ẻ */ => 'e',
216  "\xE1\xBB\x82" /* Ể */ => 'E', "\xE1\xBB\x83" /* ể */ => 'e',
217  "\xE1\xBB\x88" /* Ỉ */ => 'I', "\xE1\xBB\x89" /* ỉ */ => 'i',
218  "\xE1\xBB\x8E" /* Ỏ */ => 'O', "\xE1\xBB\x8F" /* ỏ */ => 'o',
219  "\xE1\xBB\x94" /* Ổ */ => 'O', "\xE1\xBB\x95" /* ổ */ => 'o',
220  "\xE1\xBB\x9E" /* Ở */ => 'O', "\xE1\xBB\x9F" /* ở */ => 'o',
221  "\xE1\xBB\xA6" /* Ủ */ => 'U', "\xE1\xBB\xA7" /* ủ */ => 'u',
222  "\xE1\xBB\xAC" /* Ử */ => 'U', "\xE1\xBB\xAD" /* ử */ => 'u',
223  "\xE1\xBB\xB6" /* Ỷ */ => 'Y', "\xE1\xBB\xB7" /* ỷ */ => 'y',
224  // tilde
225  "\xE1\xBA\xAA" /* Ẫ */ => 'A', "\xE1\xBA\xAB" /* ẫ */ => 'a',
226  "\xE1\xBA\xB4" /* Ẵ */ => 'A', "\xE1\xBA\xB5" /* ẵ */ => 'a',
227  "\xE1\xBA\xBC" /* Ẽ */ => 'E', "\xE1\xBA\xBD" /* ẽ */ => 'e',
228  "\xE1\xBB\x84" /* Ễ */ => 'E', "\xE1\xBB\x85" /* ễ */ => 'e',
229  "\xE1\xBB\x96" /* Ỗ */ => 'O', "\xE1\xBB\x97" /* ỗ */ => 'o',
230  "\xE1\xBB\xA0" /* Ỡ */ => 'O', "\xE1\xBB\xA1" /* ỡ */ => 'o',
231  "\xE1\xBB\xAE" /* Ữ */ => 'U', "\xE1\xBB\xAF" /* ữ */ => 'u',
232  "\xE1\xBB\xB8" /* Ỹ */ => 'Y', "\xE1\xBB\xB9" /* ỹ */ => 'y',
233  // acute accent
234  "\xE1\xBA\xA4" /* Ấ */ => 'A', "\xE1\xBA\xA5" /* ấ */ => 'a',
235  "\xE1\xBA\xAE" /* Ắ */ => 'A', "\xE1\xBA\xAF" /* ắ */ => 'a',
236  "\xE1\xBA\xBE" /* Ế */ => 'E', "\xE1\xBA\xBF" /* ế */ => 'e',
237  "\xE1\xBB\x90" /* Ố */ => 'O', "\xE1\xBB\x91" /* ố */ => 'o',
238  "\xE1\xBB\x9A" /* Ớ */ => 'O', "\xE1\xBB\x9B" /* ớ */ => 'o',
239  "\xE1\xBB\xA8" /* Ứ */ => 'U', "\xE1\xBB\xA9" /* ứ */ => 'u',
240  // dot below
241  "\xE1\xBA\xA0" /* Ạ */ => 'A', "\xE1\xBA\xA1" /* ạ */ => 'a',
242  "\xE1\xBA\xAC" /* Ậ */ => 'A', "\xE1\xBA\xAD" /* ậ */ => 'a',
243  "\xE1\xBA\xB6" /* Ặ */ => 'A', "\xE1\xBA\xB7" /* ặ */ => 'a',
244  "\xE1\xBA\xB8" /* Ẹ */ => 'E', "\xE1\xBA\xB9" /* ẹ */ => 'e',
245  "\xE1\xBB\x86" /* Ệ */ => 'E', "\xE1\xBB\x87" /* ệ */ => 'e',
246  "\xE1\xBB\x8A" /* Ị */ => 'I', "\xE1\xBB\x8B" /* ị */ => 'i',
247  "\xE1\xBB\x8C" /* Ọ */ => 'O', "\xE1\xBB\x8D" /* ọ */ => 'o',
248  "\xE1\xBB\x98" /* Ộ */ => 'O', "\xE1\xBB\x99" /* ộ */ => 'o',
249  "\xE1\xBB\xA2" /* Ợ */ => 'O', "\xE1\xBB\xA3" /* ợ */ => 'o',
250  "\xE1\xBB\xA4" /* Ụ */ => 'U', "\xE1\xBB\xA5" /* ụ */ => 'u',
251  "\xE1\xBB\xB0" /* Ự */ => 'U', "\xE1\xBB\xB1" /* ự */ => 'u',
252  "\xE1\xBB\xB4" /* Ỵ */ => 'Y', "\xE1\xBB\xB5" /* ỵ */ => 'y',
253  );
254  }
255 
256  /**
257  * Tests that "normalizer_normalize" exists and works
258  *
259  * @return bool
260  */
261  static public function hasNormalizerSupport() {
262  static $ret = null;
263  if (null === $ret) {
264  $form_c = "\xC3\x85"; // 'LATIN CAPITAL LETTER A WITH RING ABOVE' (U+00C5)
265  $form_d = "A\xCC\x8A"; // A followed by 'COMBINING RING ABOVE' (U+030A)
266  $ret = (function_exists('normalizer_normalize')
267  && $form_c === normalizer_normalize($form_d));
268  }
269  return $ret;
270  }
271 }