PageRenderTime 52ms CodeModel.GetById 24ms RepoModel.GetById 0ms app.codeStats 0ms

/library/Doctrine/Doctrine/Inflector.php

https://github.com/ostric/e-learning
PHP | 259 lines | 164 code | 14 blank | 81 comment | 16 complexity | de5fc3a2cc4724dc4cca78285b2d9d2e MD5 | raw file
  1. <?php
  2. /*
  3. * $Id: Inflector.php 3189 2007-11-18 20:37:44Z meus $
  4. *
  5. * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
  6. * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  7. * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
  8. * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
  9. * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  10. * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
  11. * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  12. * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  13. * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  14. * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
  15. * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  16. *
  17. * This software consists of voluntary contributions made by many individuals
  18. * and is licensed under the LGPL. For more information, see
  19. * <http://www.phpdoctrine.org>.
  20. */
  21. /**
  22. * Doctrine inflector has static methods for inflecting text
  23. *
  24. * The methods in these classes are from several different sources collected
  25. * across several different php projects and several different authors. The
  26. * original author names and emails are not known
  27. *
  28. * @package Doctrine
  29. * @subpackage Inflector
  30. * @license http://www.opensource.org/licenses/lgpl-license.php LGPL
  31. * @link www.phpdoctrine.org
  32. * @since 1.0
  33. * @version $Revision: 3189 $
  34. * @author Konsta Vesterinen <kvesteri@cc.hut.fi>
  35. * @author Jonathan H. Wage <jonwage@gmail.com>
  36. */
  37. class Doctrine_Inflector
  38. {
  39. /**
  40. * Convert word in to the format for a Doctrine table name. Converts 'ModelName' to 'model_name'
  41. *
  42. * @param string $word Word to tableize
  43. * @return string $word Tableized word
  44. */
  45. public static function tableize($word)
  46. {
  47. return strtolower(preg_replace('~(?<=\\w)([A-Z])~', '_$1', $word));
  48. }
  49. /**
  50. * Convert a word in to the format for a Doctrine class name. Converts 'table_name' to 'TableName'
  51. *
  52. * @param string $word Word to classify
  53. * @return string $word Classified word
  54. */
  55. public static function classify($word)
  56. {
  57. return preg_replace_callback('~(_?)(_)([\w])~', array("Doctrine_Inflector", "classifyCallback"), ucfirst(strtolower($word)));
  58. }
  59. /**
  60. * Callback function to classify a classname properly.
  61. *
  62. * @param array $matches An array of matches from a pcre_replace call
  63. * @return string $string A string with matches 1 and mathces 3 in upper case.
  64. */
  65. public static function classifyCallback($matches)
  66. {
  67. return $matches[1] . strtoupper($matches[3]);
  68. }
  69. /**
  70. * Check if a string has utf7 characters in it
  71. *
  72. * By bmorel at ssi dot fr
  73. *
  74. * @param string $string
  75. * @return boolean $bool
  76. */
  77. public static function seemsUtf8($string)
  78. {
  79. for ($i = 0; $i < strlen($string); $i++) {
  80. if (ord($string[$i]) < 0x80) continue; # 0bbbbbbb
  81. elseif ((ord($string[$i]) & 0xE0) == 0xC0) $n=1; # 110bbbbb
  82. elseif ((ord($string[$i]) & 0xF0) == 0xE0) $n=2; # 1110bbbb
  83. elseif ((ord($string[$i]) & 0xF8) == 0xF0) $n=3; # 11110bbb
  84. elseif ((ord($string[$i]) & 0xFC) == 0xF8) $n=4; # 111110bb
  85. elseif ((ord($string[$i]) & 0xFE) == 0xFC) $n=5; # 1111110b
  86. else return false; # Does not match any model
  87. for ($j=0; $j<$n; $j++) { # n bytes matching 10bbbbbb follow ?
  88. if ((++$i == strlen($string)) || ((ord($string[$i]) & 0xC0) != 0x80))
  89. return false;
  90. }
  91. }
  92. return true;
  93. }
  94. /**
  95. * Remove any illegal characters, accents, etc.
  96. *
  97. * @param string $string String to unaccent
  98. * @return string $string Unaccented string
  99. */
  100. public static function unaccent($string)
  101. {
  102. if ( ! preg_match('/[\x80-\xff]/', $string) ) {
  103. return $string;
  104. }
  105. if (self::seemsUtf8($string)) {
  106. $chars = array(
  107. // Decompositions for Latin-1 Supplement
  108. chr(195).chr(128) => 'A', chr(195).chr(129) => 'A',
  109. chr(195).chr(130) => 'A', chr(195).chr(131) => 'A',
  110. chr(195).chr(132) => 'A', chr(195).chr(133) => 'A',
  111. chr(195).chr(135) => 'C', chr(195).chr(136) => 'E',
  112. chr(195).chr(137) => 'E', chr(195).chr(138) => 'E',
  113. chr(195).chr(139) => 'E', chr(195).chr(140) => 'I',
  114. chr(195).chr(141) => 'I', chr(195).chr(142) => 'I',
  115. chr(195).chr(143) => 'I', chr(195).chr(145) => 'N',
  116. chr(195).chr(146) => 'O', chr(195).chr(147) => 'O',
  117. chr(195).chr(148) => 'O', chr(195).chr(149) => 'O',
  118. chr(195).chr(150) => 'O', chr(195).chr(153) => 'U',
  119. chr(195).chr(154) => 'U', chr(195).chr(155) => 'U',
  120. chr(195).chr(156) => 'U', chr(195).chr(157) => 'Y',
  121. chr(195).chr(159) => 's', chr(195).chr(160) => 'a',
  122. chr(195).chr(161) => 'a', chr(195).chr(162) => 'a',
  123. chr(195).chr(163) => 'a', chr(195).chr(164) => 'a',
  124. chr(195).chr(165) => 'a', chr(195).chr(167) => 'c',
  125. chr(195).chr(168) => 'e', chr(195).chr(169) => 'e',
  126. chr(195).chr(170) => 'e', chr(195).chr(171) => 'e',
  127. chr(195).chr(172) => 'i', chr(195).chr(173) => 'i',
  128. chr(195).chr(174) => 'i', chr(195).chr(175) => 'i',
  129. chr(195).chr(177) => 'n', chr(195).chr(178) => 'o',
  130. chr(195).chr(179) => 'o', chr(195).chr(180) => 'o',
  131. chr(195).chr(181) => 'o', chr(195).chr(182) => 'o',
  132. chr(195).chr(182) => 'o', chr(195).chr(185) => 'u',
  133. chr(195).chr(186) => 'u', chr(195).chr(187) => 'u',
  134. chr(195).chr(188) => 'u', chr(195).chr(189) => 'y',
  135. chr(195).chr(191) => 'y',
  136. // Decompositions for Latin Extended-A
  137. chr(196).chr(128) => 'A', chr(196).chr(129) => 'a',
  138. chr(196).chr(130) => 'A', chr(196).chr(131) => 'a',
  139. chr(196).chr(132) => 'A', chr(196).chr(133) => 'a',
  140. chr(196).chr(134) => 'C', chr(196).chr(135) => 'c',
  141. chr(196).chr(136) => 'C', chr(196).chr(137) => 'c',
  142. chr(196).chr(138) => 'C', chr(196).chr(139) => 'c',
  143. chr(196).chr(140) => 'C', chr(196).chr(141) => 'c',
  144. chr(196).chr(142) => 'D', chr(196).chr(143) => 'd',
  145. chr(196).chr(144) => 'D', chr(196).chr(145) => 'd',
  146. chr(196).chr(146) => 'E', chr(196).chr(147) => 'e',
  147. chr(196).chr(148) => 'E', chr(196).chr(149) => 'e',
  148. chr(196).chr(150) => 'E', chr(196).chr(151) => 'e',
  149. chr(196).chr(152) => 'E', chr(196).chr(153) => 'e',
  150. chr(196).chr(154) => 'E', chr(196).chr(155) => 'e',
  151. chr(196).chr(156) => 'G', chr(196).chr(157) => 'g',
  152. chr(196).chr(158) => 'G', chr(196).chr(159) => 'g',
  153. chr(196).chr(160) => 'G', chr(196).chr(161) => 'g',
  154. chr(196).chr(162) => 'G', chr(196).chr(163) => 'g',
  155. chr(196).chr(164) => 'H', chr(196).chr(165) => 'h',
  156. chr(196).chr(166) => 'H', chr(196).chr(167) => 'h',
  157. chr(196).chr(168) => 'I', chr(196).chr(169) => 'i',
  158. chr(196).chr(170) => 'I', chr(196).chr(171) => 'i',
  159. chr(196).chr(172) => 'I', chr(196).chr(173) => 'i',
  160. chr(196).chr(174) => 'I', chr(196).chr(175) => 'i',
  161. chr(196).chr(176) => 'I', chr(196).chr(177) => 'i',
  162. chr(196).chr(178) => 'IJ',chr(196).chr(179) => 'ij',
  163. chr(196).chr(180) => 'J', chr(196).chr(181) => 'j',
  164. chr(196).chr(182) => 'K', chr(196).chr(183) => 'k',
  165. chr(196).chr(184) => 'k', chr(196).chr(185) => 'L',
  166. chr(196).chr(186) => 'l', chr(196).chr(187) => 'L',
  167. chr(196).chr(188) => 'l', chr(196).chr(189) => 'L',
  168. chr(196).chr(190) => 'l', chr(196).chr(191) => 'L',
  169. chr(197).chr(128) => 'l', chr(197).chr(129) => 'L',
  170. chr(197).chr(130) => 'l', chr(197).chr(131) => 'N',
  171. chr(197).chr(132) => 'n', chr(197).chr(133) => 'N',
  172. chr(197).chr(134) => 'n', chr(197).chr(135) => 'N',
  173. chr(197).chr(136) => 'n', chr(197).chr(137) => 'N',
  174. chr(197).chr(138) => 'n', chr(197).chr(139) => 'N',
  175. chr(197).chr(140) => 'O', chr(197).chr(141) => 'o',
  176. chr(197).chr(142) => 'O', chr(197).chr(143) => 'o',
  177. chr(197).chr(144) => 'O', chr(197).chr(145) => 'o',
  178. chr(197).chr(146) => 'OE',chr(197).chr(147) => 'oe',
  179. chr(197).chr(148) => 'R', chr(197).chr(149) => 'r',
  180. chr(197).chr(150) => 'R', chr(197).chr(151) => 'r',
  181. chr(197).chr(152) => 'R', chr(197).chr(153) => 'r',
  182. chr(197).chr(154) => 'S', chr(197).chr(155) => 's',
  183. chr(197).chr(156) => 'S', chr(197).chr(157) => 's',
  184. chr(197).chr(158) => 'S', chr(197).chr(159) => 's',
  185. chr(197).chr(160) => 'S', chr(197).chr(161) => 's',
  186. chr(197).chr(162) => 'T', chr(197).chr(163) => 't',
  187. chr(197).chr(164) => 'T', chr(197).chr(165) => 't',
  188. chr(197).chr(166) => 'T', chr(197).chr(167) => 't',
  189. chr(197).chr(168) => 'U', chr(197).chr(169) => 'u',
  190. chr(197).chr(170) => 'U', chr(197).chr(171) => 'u',
  191. chr(197).chr(172) => 'U', chr(197).chr(173) => 'u',
  192. chr(197).chr(174) => 'U', chr(197).chr(175) => 'u',
  193. chr(197).chr(176) => 'U', chr(197).chr(177) => 'u',
  194. chr(197).chr(178) => 'U', chr(197).chr(179) => 'u',
  195. chr(197).chr(180) => 'W', chr(197).chr(181) => 'w',
  196. chr(197).chr(182) => 'Y', chr(197).chr(183) => 'y',
  197. chr(197).chr(184) => 'Y', chr(197).chr(185) => 'Z',
  198. chr(197).chr(186) => 'z', chr(197).chr(187) => 'Z',
  199. chr(197).chr(188) => 'z', chr(197).chr(189) => 'Z',
  200. chr(197).chr(190) => 'z', chr(197).chr(191) => 's',
  201. // Euro Sign
  202. chr(226).chr(130).chr(172) => 'E',
  203. // GBP (Pound) Sign
  204. chr(194).chr(163) => '');
  205. $string = strtr($string, $chars);
  206. } else {
  207. // Assume ISO-8859-1 if not UTF-8
  208. $chars['in'] = chr(128).chr(131).chr(138).chr(142).chr(154).chr(158)
  209. .chr(159).chr(162).chr(165).chr(181).chr(192).chr(193).chr(194)
  210. .chr(195).chr(196).chr(197).chr(199).chr(200).chr(201).chr(202)
  211. .chr(203).chr(204).chr(205).chr(206).chr(207).chr(209).chr(210)
  212. .chr(211).chr(212).chr(213).chr(214).chr(216).chr(217).chr(218)
  213. .chr(219).chr(220).chr(221).chr(224).chr(225).chr(226).chr(227)
  214. .chr(228).chr(229).chr(231).chr(232).chr(233).chr(234).chr(235)
  215. .chr(236).chr(237).chr(238).chr(239).chr(241).chr(242).chr(243)
  216. .chr(244).chr(245).chr(246).chr(248).chr(249).chr(250).chr(251)
  217. .chr(252).chr(253).chr(255);
  218. $chars['out'] = "EfSZszYcYuAAAAAACEEEEIIIINOOOOOOUUUUYaaaaaaceeeeiiiinoooooouuuuyy";
  219. $string = strtr($string, $chars['in'], $chars['out']);
  220. $doubleChars['in'] = array(chr(140), chr(156), chr(198), chr(208), chr(222), chr(223), chr(230), chr(240), chr(254));
  221. $doubleChars['out'] = array('OE', 'oe', 'AE', 'DH', 'TH', 'ss', 'ae', 'dh', 'th');
  222. $string = str_replace($doubleChars['in'], $doubleChars['out'], $string);
  223. }
  224. return $string;
  225. }
  226. /**
  227. * Convert any passed string to a url friendly string. Converts 'My first blog post' to 'my-first-blog-post'
  228. *
  229. * @param string $text Text to urlize
  230. * @return string $text Urlized text
  231. */
  232. public static function urlize($text)
  233. {
  234. // Remove all non url friendly characters with the unaccent function
  235. $text = self::unaccent($text);
  236. // Remove all none word characters
  237. $text = preg_replace('/\W/', ' ', $text);
  238. // More stripping. Replace spaces with dashes
  239. $text = strtolower(preg_replace('/[^A-Z^a-z^0-9^\/]+/', '-',
  240. preg_replace('/([a-z\d])([A-Z])/', '\1_\2',
  241. preg_replace('/([A-Z]+)([A-Z][a-z])/', '\1_\2',
  242. preg_replace('/::/', '/', $text)))));
  243. return trim($text, '-');
  244. }
  245. }