index.js 9.0 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316
  1. exports.remove = removeDiacritics;
  2. var replacementList = [
  3. {
  4. base: ' ',
  5. chars: "\u00A0",
  6. }, {
  7. base: '0',
  8. chars: "\u07C0",
  9. }, {
  10. base: 'A',
  11. chars: "\u24B6\uFF21\u00C0\u00C1\u00C2\u1EA6\u1EA4\u1EAA\u1EA8\u00C3\u0100\u0102\u1EB0\u1EAE\u1EB4\u1EB2\u0226\u01E0\u00C4\u01DE\u1EA2\u00C5\u01FA\u01CD\u0200\u0202\u1EA0\u1EAC\u1EB6\u1E00\u0104\u023A\u2C6F",
  12. }, {
  13. base: 'AA',
  14. chars: "\uA732",
  15. }, {
  16. base: 'AE',
  17. chars: "\u00C6\u01FC\u01E2",
  18. }, {
  19. base: 'AO',
  20. chars: "\uA734",
  21. }, {
  22. base: 'AU',
  23. chars: "\uA736",
  24. }, {
  25. base: 'AV',
  26. chars: "\uA738\uA73A",
  27. }, {
  28. base: 'AY',
  29. chars: "\uA73C",
  30. }, {
  31. base: 'B',
  32. chars: "\u24B7\uFF22\u1E02\u1E04\u1E06\u0243\u0181",
  33. }, {
  34. base: 'C',
  35. chars: "\u24b8\uff23\uA73E\u1E08\u0106\u0043\u0108\u010A\u010C\u00C7\u0187\u023B",
  36. }, {
  37. base: 'D',
  38. chars: "\u24B9\uFF24\u1E0A\u010E\u1E0C\u1E10\u1E12\u1E0E\u0110\u018A\u0189\u1D05\uA779",
  39. }, {
  40. base: 'Dh',
  41. chars: "\u00D0",
  42. }, {
  43. base: 'DZ',
  44. chars: "\u01F1\u01C4",
  45. }, {
  46. base: 'Dz',
  47. chars: "\u01F2\u01C5",
  48. }, {
  49. base: 'E',
  50. chars: "\u025B\u24BA\uFF25\u00C8\u00C9\u00CA\u1EC0\u1EBE\u1EC4\u1EC2\u1EBC\u0112\u1E14\u1E16\u0114\u0116\u00CB\u1EBA\u011A\u0204\u0206\u1EB8\u1EC6\u0228\u1E1C\u0118\u1E18\u1E1A\u0190\u018E\u1D07",
  51. }, {
  52. base: 'F',
  53. chars: "\uA77C\u24BB\uFF26\u1E1E\u0191\uA77B",
  54. }, {
  55. base: 'G',
  56. chars: "\u24BC\uFF27\u01F4\u011C\u1E20\u011E\u0120\u01E6\u0122\u01E4\u0193\uA7A0\uA77D\uA77E\u0262",
  57. }, {
  58. base: 'H',
  59. chars: "\u24BD\uFF28\u0124\u1E22\u1E26\u021E\u1E24\u1E28\u1E2A\u0126\u2C67\u2C75\uA78D",
  60. }, {
  61. base: 'I',
  62. chars: "\u24BE\uFF29\xCC\xCD\xCE\u0128\u012A\u012C\u0130\xCF\u1E2E\u1EC8\u01CF\u0208\u020A\u1ECA\u012E\u1E2C\u0197",
  63. }, {
  64. base: 'J',
  65. chars: "\u24BF\uFF2A\u0134\u0248\u0237",
  66. }, {
  67. base: 'K',
  68. chars: "\u24C0\uFF2B\u1E30\u01E8\u1E32\u0136\u1E34\u0198\u2C69\uA740\uA742\uA744\uA7A2",
  69. }, {
  70. base: 'L',
  71. chars: "\u24C1\uFF2C\u013F\u0139\u013D\u1E36\u1E38\u013B\u1E3C\u1E3A\u0141\u023D\u2C62\u2C60\uA748\uA746\uA780",
  72. }, {
  73. base: 'LJ',
  74. chars: "\u01C7",
  75. }, {
  76. base: 'Lj',
  77. chars: "\u01C8",
  78. }, {
  79. base: 'M',
  80. chars: "\u24C2\uFF2D\u1E3E\u1E40\u1E42\u2C6E\u019C\u03FB",
  81. }, {
  82. base: 'N',
  83. chars: "\uA7A4\u0220\u24C3\uFF2E\u01F8\u0143\xD1\u1E44\u0147\u1E46\u0145\u1E4A\u1E48\u019D\uA790\u1D0E",
  84. }, {
  85. base: 'NJ',
  86. chars: "\u01CA",
  87. }, {
  88. base: 'Nj',
  89. chars: "\u01CB",
  90. }, {
  91. base: 'O',
  92. chars: "\u24C4\uFF2F\xD2\xD3\xD4\u1ED2\u1ED0\u1ED6\u1ED4\xD5\u1E4C\u022C\u1E4E\u014C\u1E50\u1E52\u014E\u022E\u0230\xD6\u022A\u1ECE\u0150\u01D1\u020C\u020E\u01A0\u1EDC\u1EDA\u1EE0\u1EDE\u1EE2\u1ECC\u1ED8\u01EA\u01EC\xD8\u01FE\u0186\u019F\uA74A\uA74C",
  93. }, {
  94. base: 'OE',
  95. chars: "\u0152",
  96. }, {
  97. base: 'OI',
  98. chars: "\u01A2",
  99. }, {
  100. base: 'OO',
  101. chars: "\uA74E",
  102. }, {
  103. base: 'OU',
  104. chars: "\u0222",
  105. }, {
  106. base: 'P',
  107. chars: "\u24C5\uFF30\u1E54\u1E56\u01A4\u2C63\uA750\uA752\uA754",
  108. }, {
  109. base: 'Q',
  110. chars: "\u24C6\uFF31\uA756\uA758\u024A",
  111. }, {
  112. base: 'R',
  113. chars: "\u24C7\uFF32\u0154\u1E58\u0158\u0210\u0212\u1E5A\u1E5C\u0156\u1E5E\u024C\u2C64\uA75A\uA7A6\uA782",
  114. }, {
  115. base: 'S',
  116. chars: "\u24C8\uFF33\u1E9E\u015A\u1E64\u015C\u1E60\u0160\u1E66\u1E62\u1E68\u0218\u015E\u2C7E\uA7A8\uA784",
  117. }, {
  118. base: 'T',
  119. chars: "\u24C9\uFF34\u1E6A\u0164\u1E6C\u021A\u0162\u1E70\u1E6E\u0166\u01AC\u01AE\u023E\uA786",
  120. }, {
  121. base: 'Th',
  122. chars: "\u00DE",
  123. }, {
  124. base: 'TZ',
  125. chars: "\uA728",
  126. }, {
  127. base: 'U',
  128. chars: "\u24CA\uFF35\xD9\xDA\xDB\u0168\u1E78\u016A\u1E7A\u016C\xDC\u01DB\u01D7\u01D5\u01D9\u1EE6\u016E\u0170\u01D3\u0214\u0216\u01AF\u1EEA\u1EE8\u1EEE\u1EEC\u1EF0\u1EE4\u1E72\u0172\u1E76\u1E74\u0244",
  129. }, {
  130. base: 'V',
  131. chars: "\u24CB\uFF36\u1E7C\u1E7E\u01B2\uA75E\u0245",
  132. }, {
  133. base: 'VY',
  134. chars: "\uA760",
  135. }, {
  136. base: 'W',
  137. chars: "\u24CC\uFF37\u1E80\u1E82\u0174\u1E86\u1E84\u1E88\u2C72",
  138. }, {
  139. base: 'X',
  140. chars: "\u24CD\uFF38\u1E8A\u1E8C",
  141. }, {
  142. base: 'Y',
  143. chars: "\u24CE\uFF39\u1EF2\xDD\u0176\u1EF8\u0232\u1E8E\u0178\u1EF6\u1EF4\u01B3\u024E\u1EFE",
  144. }, {
  145. base: 'Z',
  146. chars: "\u24CF\uFF3A\u0179\u1E90\u017B\u017D\u1E92\u1E94\u01B5\u0224\u2C7F\u2C6B\uA762",
  147. }, {
  148. base: 'a',
  149. chars: "\u24D0\uFF41\u1E9A\u00E0\u00E1\u00E2\u1EA7\u1EA5\u1EAB\u1EA9\u00E3\u0101\u0103\u1EB1\u1EAF\u1EB5\u1EB3\u0227\u01E1\u00E4\u01DF\u1EA3\u00E5\u01FB\u01CE\u0201\u0203\u1EA1\u1EAD\u1EB7\u1E01\u0105\u2C65\u0250\u0251",
  150. }, {
  151. base: 'aa',
  152. chars: "\uA733",
  153. }, {
  154. base: 'ae',
  155. chars: "\u00E6\u01FD\u01E3",
  156. }, {
  157. base: 'ao',
  158. chars: "\uA735",
  159. }, {
  160. base: 'au',
  161. chars: "\uA737",
  162. }, {
  163. base: 'av',
  164. chars: "\uA739\uA73B",
  165. }, {
  166. base: 'ay',
  167. chars: "\uA73D",
  168. }, {
  169. base: 'b',
  170. chars: "\u24D1\uFF42\u1E03\u1E05\u1E07\u0180\u0183\u0253\u0182",
  171. }, {
  172. base: 'c',
  173. chars: "\uFF43\u24D2\u0107\u0109\u010B\u010D\u00E7\u1E09\u0188\u023C\uA73F\u2184",
  174. }, {
  175. base: 'd',
  176. chars: "\u24D3\uFF44\u1E0B\u010F\u1E0D\u1E11\u1E13\u1E0F\u0111\u018C\u0256\u0257\u018B\u13E7\u0501\uA7AA",
  177. }, {
  178. base: 'dh',
  179. chars: "\u00F0",
  180. }, {
  181. base: 'dz',
  182. chars: "\u01F3\u01C6",
  183. }, {
  184. base: 'e',
  185. chars: "\u24D4\uFF45\u00E8\u00E9\u00EA\u1EC1\u1EBF\u1EC5\u1EC3\u1EBD\u0113\u1E15\u1E17\u0115\u0117\u00EB\u1EBB\u011B\u0205\u0207\u1EB9\u1EC7\u0229\u1E1D\u0119\u1E19\u1E1B\u0247\u01DD",
  186. }, {
  187. base: 'f',
  188. chars: "\u24D5\uFF46\u1E1F\u0192",
  189. }, {
  190. base: 'ff',
  191. chars: "\uFB00",
  192. }, {
  193. base: 'fi',
  194. chars: "\uFB01",
  195. }, {
  196. base: 'fl',
  197. chars: "\uFB02",
  198. }, {
  199. base: 'ffi',
  200. chars: "\uFB03",
  201. }, {
  202. base: 'ffl',
  203. chars: "\uFB04",
  204. }, {
  205. base: 'g',
  206. chars: "\u24D6\uFF47\u01F5\u011D\u1E21\u011F\u0121\u01E7\u0123\u01E5\u0260\uA7A1\uA77F\u1D79",
  207. }, {
  208. base: 'h',
  209. chars: "\u24D7\uFF48\u0125\u1E23\u1E27\u021F\u1E25\u1E29\u1E2B\u1E96\u0127\u2C68\u2C76\u0265",
  210. }, {
  211. base: 'hv',
  212. chars: "\u0195",
  213. }, {
  214. base: 'i',
  215. chars: "\u24D8\uFF49\xEC\xED\xEE\u0129\u012B\u012D\xEF\u1E2F\u1EC9\u01D0\u0209\u020B\u1ECB\u012F\u1E2D\u0268\u0131",
  216. }, {
  217. base: 'j',
  218. chars: "\u24D9\uFF4A\u0135\u01F0\u0249",
  219. }, {
  220. base: 'k',
  221. chars: "\u24DA\uFF4B\u1E31\u01E9\u1E33\u0137\u1E35\u0199\u2C6A\uA741\uA743\uA745\uA7A3",
  222. }, {
  223. base: 'l',
  224. chars: "\u24DB\uFF4C\u0140\u013A\u013E\u1E37\u1E39\u013C\u1E3D\u1E3B\u017F\u0142\u019A\u026B\u2C61\uA749\uA781\uA747\u026D",
  225. }, {
  226. base: 'lj',
  227. chars: "\u01C9",
  228. }, {
  229. base: 'm',
  230. chars: "\u24DC\uFF4D\u1E3F\u1E41\u1E43\u0271\u026F",
  231. }, {
  232. base: 'n',
  233. chars: "\u24DD\uFF4E\u01F9\u0144\xF1\u1E45\u0148\u1E47\u0146\u1E4B\u1E49\u019E\u0272\u0149\uA791\uA7A5\u043B\u0509",
  234. }, {
  235. base: 'nj',
  236. chars: "\u01CC",
  237. }, {
  238. base: 'o',
  239. chars: "\u24DE\uFF4F\xF2\xF3\xF4\u1ED3\u1ED1\u1ED7\u1ED5\xF5\u1E4D\u022D\u1E4F\u014D\u1E51\u1E53\u014F\u022F\u0231\xF6\u022B\u1ECF\u0151\u01D2\u020D\u020F\u01A1\u1EDD\u1EDB\u1EE1\u1EDF\u1EE3\u1ECD\u1ED9\u01EB\u01ED\xF8\u01FF\uA74B\uA74D\u0275\u0254\u1D11",
  240. }, {
  241. base: 'oe',
  242. chars: "\u0153",
  243. }, {
  244. base: 'oi',
  245. chars: "\u01A3",
  246. }, {
  247. base: 'oo',
  248. chars: "\uA74F",
  249. }, {
  250. base: 'ou',
  251. chars: "\u0223",
  252. }, {
  253. base: 'p',
  254. chars: "\u24DF\uFF50\u1E55\u1E57\u01A5\u1D7D\uA751\uA753\uA755\u03C1",
  255. }, {
  256. base: 'q',
  257. chars: "\u24E0\uFF51\u024B\uA757\uA759",
  258. }, {
  259. base: 'r',
  260. chars: "\u24E1\uFF52\u0155\u1E59\u0159\u0211\u0213\u1E5B\u1E5D\u0157\u1E5F\u024D\u027D\uA75B\uA7A7\uA783",
  261. }, {
  262. base: 's',
  263. chars: "\u24E2\uFF53\u015B\u1E65\u015D\u1E61\u0161\u1E67\u1E63\u1E69\u0219\u015F\u023F\uA7A9\uA785\u1E9B\u0282",
  264. }, {
  265. base: 'ss',
  266. chars: "\xDF",
  267. }, {
  268. base: 't',
  269. chars: "\u24E3\uFF54\u1E6B\u1E97\u0165\u1E6D\u021B\u0163\u1E71\u1E6F\u0167\u01AD\u0288\u2C66\uA787",
  270. }, {
  271. base: 'th',
  272. chars: "\u00FE",
  273. }, {
  274. base: 'tz',
  275. chars: "\uA729",
  276. }, {
  277. base: 'u',
  278. chars: "\u24E4\uFF55\xF9\xFA\xFB\u0169\u1E79\u016B\u1E7B\u016D\xFC\u01DC\u01D8\u01D6\u01DA\u1EE7\u016F\u0171\u01D4\u0215\u0217\u01B0\u1EEB\u1EE9\u1EEF\u1EED\u1EF1\u1EE5\u1E73\u0173\u1E77\u1E75\u0289",
  279. }, {
  280. base: 'v',
  281. chars: "\u24E5\uFF56\u1E7D\u1E7F\u028B\uA75F\u028C",
  282. }, {
  283. base: 'vy',
  284. chars: "\uA761",
  285. }, {
  286. base: 'w',
  287. chars: "\u24E6\uFF57\u1E81\u1E83\u0175\u1E87\u1E85\u1E98\u1E89\u2C73",
  288. }, {
  289. base: 'x',
  290. chars: "\u24E7\uFF58\u1E8B\u1E8D",
  291. }, {
  292. base: 'y',
  293. chars: "\u24E8\uFF59\u1EF3\xFD\u0177\u1EF9\u0233\u1E8F\xFF\u1EF7\u1E99\u1EF5\u01B4\u024F\u1EFF",
  294. }, {
  295. base: 'z',
  296. chars: "\u24E9\uFF5A\u017A\u1E91\u017C\u017E\u1E93\u1E95\u01B6\u0225\u0240\u2C6C\uA763",
  297. }
  298. ];
  299. var diacriticsMap = {};
  300. for (var i = 0; i < replacementList.length; i += 1) {
  301. var chars = replacementList[i].chars;
  302. for (var j = 0; j < chars.length; j += 1) {
  303. diacriticsMap[chars[j]] = replacementList[i].base;
  304. }
  305. }
  306. function removeDiacritics(str) {
  307. return str.replace(/[^\u0000-\u007e]/g, function(c) {
  308. return diacriticsMap[c] || c;
  309. });
  310. }
  311. exports.replacementList = replacementList;
  312. exports.diacriticsMap = diacriticsMap;