PageRenderTime 53ms CodeModel.GetById 17ms RepoModel.GetById 0ms app.codeStats 0ms

/application/third_party/Gedmo/Sluggable/Util/Urlizer.php

https://github.com/mrcoco/codeigniter-hmvc-doctrine
PHP | 221 lines | 162 code | 13 blank | 46 comment | 18 complexity | c423a33ed9458df07690f69ab80b3bf6 MD5 | raw file
Possible License(s): LGPL-2.1
  1. <?php
  2. namespace Gedmo\Sluggable\Util;
  3. /**
  4. * This is the part taken from Doctrine 1.2.3
  5. * Doctrine inflector has static methods for inflecting text
  6. *
  7. * The methods in these classes are from several different sources collected
  8. * across several different php projects and several different authors. The
  9. * original author names and emails are not known
  10. *
  11. * @package Gedmo.Sluggable.Util
  12. * @subpackage Urlizer
  13. * @license http://www.opensource.org/licenses/lgpl-license.php LGPL
  14. * @link www.doctrine-project.org
  15. * @since 1.0
  16. * @version $Revision: 3189 $
  17. * @author Konsta Vesterinen <kvesteri@cc.hut.fi>
  18. * @author Jonathan H. Wage <jonwage@gmail.com>
  19. */
  20. class Urlizer
  21. {
  22. /**
  23. * Check if a string has utf7 characters in it
  24. *
  25. * By bmorel at ssi dot fr
  26. *
  27. * @param string $string
  28. * @return boolean $bool
  29. */
  30. public static function seemsUtf8($string)
  31. {
  32. for ($i = 0; $i < strlen($string); $i++) {
  33. if (ord($string[$i]) < 0x80) continue; # 0bbbbbbb
  34. elseif ((ord($string[$i]) & 0xE0) == 0xC0) $n=1; # 110bbbbb
  35. elseif ((ord($string[$i]) & 0xF0) == 0xE0) $n=2; # 1110bbbb
  36. elseif ((ord($string[$i]) & 0xF8) == 0xF0) $n=3; # 11110bbb
  37. elseif ((ord($string[$i]) & 0xFC) == 0xF8) $n=4; # 111110bb
  38. elseif ((ord($string[$i]) & 0xFE) == 0xFC) $n=5; # 1111110b
  39. else return false; # Does not match any model
  40. for ($j=0; $j<$n; $j++) { # n bytes matching 10bbbbbb follow ?
  41. if ((++$i == strlen($string)) || ((ord($string[$i]) & 0xC0) != 0x80))
  42. return false;
  43. }
  44. }
  45. return true;
  46. }
  47. /**
  48. * Remove any illegal characters, accents, etc.
  49. *
  50. * @param string $string String to unaccent
  51. * @return string $string Unaccented string
  52. */
  53. public static function unaccent($string)
  54. {
  55. if (!preg_match('/[\x80-\xff]/', $string)) {
  56. return $string;
  57. }
  58. if (self::seemsUtf8($string)) {
  59. $chars = array(
  60. // Decompositions for Latin-1 Supplement
  61. chr(195).chr(128) => 'A', chr(195).chr(129) => 'A',
  62. chr(195).chr(130) => 'A', chr(195).chr(131) => 'A',
  63. chr(195).chr(132) => 'A', chr(195).chr(133) => 'A',
  64. chr(195).chr(135) => 'C', chr(195).chr(136) => 'E',
  65. chr(195).chr(137) => 'E', chr(195).chr(138) => 'E',
  66. chr(195).chr(139) => 'E', chr(195).chr(140) => 'I',
  67. chr(195).chr(141) => 'I', chr(195).chr(142) => 'I',
  68. chr(195).chr(143) => 'I', chr(195).chr(145) => 'N',
  69. chr(195).chr(146) => 'O', chr(195).chr(147) => 'O',
  70. chr(195).chr(148) => 'O', chr(195).chr(149) => 'O',
  71. chr(195).chr(150) => 'O', chr(195).chr(153) => 'U',
  72. chr(195).chr(154) => 'U', chr(195).chr(155) => 'U',
  73. chr(195).chr(156) => 'U', chr(195).chr(157) => 'Y',
  74. chr(195).chr(159) => 's', chr(195).chr(160) => 'a',
  75. chr(195).chr(161) => 'a', chr(195).chr(162) => 'a',
  76. chr(195).chr(163) => 'a', chr(195).chr(164) => 'a',
  77. chr(195).chr(165) => 'a', chr(195).chr(167) => 'c',
  78. chr(195).chr(168) => 'e', chr(195).chr(169) => 'e',
  79. chr(195).chr(170) => 'e', chr(195).chr(171) => 'e',
  80. chr(195).chr(172) => 'i', chr(195).chr(173) => 'i',
  81. chr(195).chr(174) => 'i', chr(195).chr(175) => 'i',
  82. chr(195).chr(177) => 'n', chr(195).chr(178) => 'o',
  83. chr(195).chr(179) => 'o', chr(195).chr(180) => 'o',
  84. chr(195).chr(181) => 'o', chr(195).chr(182) => 'o',
  85. chr(195).chr(182) => 'o', chr(195).chr(185) => 'u',
  86. chr(195).chr(186) => 'u', chr(195).chr(187) => 'u',
  87. chr(195).chr(188) => 'u', chr(195).chr(189) => 'y',
  88. chr(195).chr(191) => 'y',
  89. // Decompositions for Latin Extended-A
  90. chr(196).chr(128) => 'A', chr(196).chr(129) => 'a',
  91. chr(196).chr(130) => 'A', chr(196).chr(131) => 'a',
  92. chr(196).chr(132) => 'A', chr(196).chr(133) => 'a',
  93. chr(196).chr(134) => 'C', chr(196).chr(135) => 'c',
  94. chr(196).chr(136) => 'C', chr(196).chr(137) => 'c',
  95. chr(196).chr(138) => 'C', chr(196).chr(139) => 'c',
  96. chr(196).chr(140) => 'C', chr(196).chr(141) => 'c',
  97. chr(196).chr(142) => 'D', chr(196).chr(143) => 'd',
  98. chr(196).chr(144) => 'D', chr(196).chr(145) => 'd',
  99. chr(196).chr(146) => 'E', chr(196).chr(147) => 'e',
  100. chr(196).chr(148) => 'E', chr(196).chr(149) => 'e',
  101. chr(196).chr(150) => 'E', chr(196).chr(151) => 'e',
  102. chr(196).chr(152) => 'E', chr(196).chr(153) => 'e',
  103. chr(196).chr(154) => 'E', chr(196).chr(155) => 'e',
  104. chr(196).chr(156) => 'G', chr(196).chr(157) => 'g',
  105. chr(196).chr(158) => 'G', chr(196).chr(159) => 'g',
  106. chr(196).chr(160) => 'G', chr(196).chr(161) => 'g',
  107. chr(196).chr(162) => 'G', chr(196).chr(163) => 'g',
  108. chr(196).chr(164) => 'H', chr(196).chr(165) => 'h',
  109. chr(196).chr(166) => 'H', chr(196).chr(167) => 'h',
  110. chr(196).chr(168) => 'I', chr(196).chr(169) => 'i',
  111. chr(196).chr(170) => 'I', chr(196).chr(171) => 'i',
  112. chr(196).chr(172) => 'I', chr(196).chr(173) => 'i',
  113. chr(196).chr(174) => 'I', chr(196).chr(175) => 'i',
  114. chr(196).chr(176) => 'I', chr(196).chr(177) => 'i',
  115. chr(196).chr(178) => 'IJ',chr(196).chr(179) => 'ij',
  116. chr(196).chr(180) => 'J', chr(196).chr(181) => 'j',
  117. chr(196).chr(182) => 'K', chr(196).chr(183) => 'k',
  118. chr(196).chr(184) => 'k', chr(196).chr(185) => 'L',
  119. chr(196).chr(186) => 'l', chr(196).chr(187) => 'L',
  120. chr(196).chr(188) => 'l', chr(196).chr(189) => 'L',
  121. chr(196).chr(190) => 'l', chr(196).chr(191) => 'L',
  122. chr(197).chr(128) => 'l', chr(197).chr(129) => 'L',
  123. chr(197).chr(130) => 'l', chr(197).chr(131) => 'N',
  124. chr(197).chr(132) => 'n', chr(197).chr(133) => 'N',
  125. chr(197).chr(134) => 'n', chr(197).chr(135) => 'N',
  126. chr(197).chr(136) => 'n', chr(197).chr(137) => 'N',
  127. chr(197).chr(138) => 'n', chr(197).chr(139) => 'N',
  128. chr(197).chr(140) => 'O', chr(197).chr(141) => 'o',
  129. chr(197).chr(142) => 'O', chr(197).chr(143) => 'o',
  130. chr(197).chr(144) => 'O', chr(197).chr(145) => 'o',
  131. chr(197).chr(146) => 'OE',chr(197).chr(147) => 'oe',
  132. chr(197).chr(148) => 'R', chr(197).chr(149) => 'r',
  133. chr(197).chr(150) => 'R', chr(197).chr(151) => 'r',
  134. chr(197).chr(152) => 'R', chr(197).chr(153) => 'r',
  135. chr(197).chr(154) => 'S', chr(197).chr(155) => 's',
  136. chr(197).chr(156) => 'S', chr(197).chr(157) => 's',
  137. chr(197).chr(158) => 'S', chr(197).chr(159) => 's',
  138. chr(197).chr(160) => 'S', chr(197).chr(161) => 's',
  139. chr(197).chr(162) => 'T', chr(197).chr(163) => 't',
  140. chr(197).chr(164) => 'T', chr(197).chr(165) => 't',
  141. chr(197).chr(166) => 'T', chr(197).chr(167) => 't',
  142. chr(197).chr(168) => 'U', chr(197).chr(169) => 'u',
  143. chr(197).chr(170) => 'U', chr(197).chr(171) => 'u',
  144. chr(197).chr(172) => 'U', chr(197).chr(173) => 'u',
  145. chr(197).chr(174) => 'U', chr(197).chr(175) => 'u',
  146. chr(197).chr(176) => 'U', chr(197).chr(177) => 'u',
  147. chr(197).chr(178) => 'U', chr(197).chr(179) => 'u',
  148. chr(197).chr(180) => 'W', chr(197).chr(181) => 'w',
  149. chr(197).chr(182) => 'Y', chr(197).chr(183) => 'y',
  150. chr(197).chr(184) => 'Y', chr(197).chr(185) => 'Z',
  151. chr(197).chr(186) => 'z', chr(197).chr(187) => 'Z',
  152. chr(197).chr(188) => 'z', chr(197).chr(189) => 'Z',
  153. chr(197).chr(190) => 'z', chr(197).chr(191) => 's',
  154. // Euro Sign
  155. chr(226).chr(130).chr(172) => 'E',
  156. // GBP (Pound) Sign
  157. chr(194).chr(163) => '',
  158. 'Ä' => 'Ae', 'ä' => 'ae', 'Ü' => 'Ue', 'ü' => 'ue',
  159. 'Ö' => 'Oe', 'ö' => 'oe', 'ß' => 'ss',
  160. // Norwegian characters
  161. 'Å'=>'Aa','Æ'=>'Ae','Ø'=>'O','æ'=>'a','ø'=>'o','å'=>'aa'
  162. );
  163. $string = strtr($string, $chars);
  164. } else {
  165. // Assume ISO-8859-1 if not UTF-8
  166. $chars['in'] = chr(128).chr(131).chr(138).chr(142).chr(154).chr(158)
  167. .chr(159).chr(162).chr(165).chr(181).chr(192).chr(193).chr(194)
  168. .chr(195).chr(196).chr(197).chr(199).chr(200).chr(201).chr(202)
  169. .chr(203).chr(204).chr(205).chr(206).chr(207).chr(209).chr(210)
  170. .chr(211).chr(212).chr(213).chr(214).chr(216).chr(217).chr(218)
  171. .chr(219).chr(220).chr(221).chr(224).chr(225).chr(226).chr(227)
  172. .chr(228).chr(229).chr(231).chr(232).chr(233).chr(234).chr(235)
  173. .chr(236).chr(237).chr(238).chr(239).chr(241).chr(242).chr(243)
  174. .chr(244).chr(245).chr(246).chr(248).chr(249).chr(250).chr(251)
  175. .chr(252).chr(253).chr(255);
  176. $chars['out'] = "EfSZszYcYuAAAAAACEEEEIIIINOOOOOOUUUUYaaaaaaceeeeiiiinoooooouuuuyy";
  177. $string = strtr($string, $chars['in'], $chars['out']);
  178. $doubleChars['in'] = array(chr(140), chr(156), chr(198), chr(208), chr(222), chr(223), chr(230), chr(240), chr(254));
  179. $doubleChars['out'] = array('OE', 'oe', 'AE', 'DH', 'TH', 'ss', 'ae', 'dh', 'th');
  180. $string = str_replace($doubleChars['in'], $doubleChars['out'], $string);
  181. }
  182. return $string;
  183. }
  184. /**
  185. * Convert any passed string to a url friendly string. Converts 'My first blog post' to 'my-first-blog-post'
  186. *
  187. * @param string $text Text to urlize
  188. * @return string $text Urlized text
  189. */
  190. public static function urlize($text, $separator = '-')
  191. {
  192. // Remove all non url friendly characters with the unaccent function
  193. $text = self::unaccent($text);
  194. if (function_exists('mb_strtolower')) {
  195. $text = mb_strtolower($text);
  196. } else {
  197. $text = strtolower($text);
  198. }
  199. // Remove all none word characters
  200. $text = preg_replace('/\W/', ' ', $text);
  201. // More stripping. Replace spaces with dashes
  202. $text = strtolower(preg_replace('/[^A-Z^a-z^0-9^\/]+/', $separator,
  203. preg_replace('/([a-z\d])([A-Z])/', '\1_\2',
  204. preg_replace('/([A-Z]+)([A-Z][a-z])/', '\1_\2',
  205. preg_replace('/::/', '/', $text)))));
  206. return trim($text, $separator);
  207. }
  208. }