urlify.js 5.5 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140
  1. var LATIN_MAP = {
  2. 'À': 'A', 'Á': 'A', 'Â': 'A', 'Ã': 'A', 'Ä': 'A', 'Å': 'A', 'Æ': 'AE', 'Ç':
  3. 'C', 'È': 'E', 'É': 'E', 'Ê': 'E', 'Ë': 'E', 'Ì': 'I', 'Í': 'I', 'Î': 'I',
  4. 'Ï': 'I', 'Ð': 'D', 'Ñ': 'N', 'Ò': 'O', 'Ó': 'O', 'Ô': 'O', 'Õ': 'O', 'Ö':
  5. 'O', 'Ő': 'O', 'Ø': 'O', 'Ù': 'U', 'Ú': 'U', 'Û': 'U', 'Ü': 'U', 'Ű': 'U',
  6. 'Ý': 'Y', 'Þ': 'TH', 'ß': 'ss', 'à':'a', 'á':'a', 'â': 'a', 'ã': 'a', 'ä':
  7. 'a', 'å': 'a', 'æ': 'ae', 'ç': 'c', 'è': 'e', 'é': 'e', 'ê': 'e', 'ë': 'e',
  8. 'ì': 'i', 'í': 'i', 'î': 'i', 'ï': 'i', 'ð': 'd', 'ñ': 'n', 'ò': 'o', 'ó':
  9. 'o', 'ô': 'o', 'õ': 'o', 'ö': 'o', 'ő': 'o', 'ø': 'o', 'ù': 'u', 'ú': 'u',
  10. 'û': 'u', 'ü': 'u', 'ű': 'u', 'ý': 'y', 'þ': 'th', 'ÿ': 'y'
  11. }
  12. var LATIN_SYMBOLS_MAP = {
  13. '©':'(c)'
  14. }
  15. var GREEK_MAP = {
  16. 'α':'a', 'β':'b', 'γ':'g', 'δ':'d', 'ε':'e', 'ζ':'z', 'η':'h', 'θ':'8',
  17. 'ι':'i', 'κ':'k', 'λ':'l', 'μ':'m', 'ν':'n', 'ξ':'3', 'ο':'o', 'π':'p',
  18. 'ρ':'r', 'σ':'s', 'τ':'t', 'υ':'y', 'φ':'f', 'χ':'x', 'ψ':'ps', 'ω':'w',
  19. 'ά':'a', 'έ':'e', 'ί':'i', 'ό':'o', 'ύ':'y', 'ή':'h', 'ώ':'w', 'ς':'s',
  20. 'ϊ':'i', 'ΰ':'y', 'ϋ':'y', 'ΐ':'i',
  21. 'Α':'A', 'Β':'B', 'Γ':'G', 'Δ':'D', 'Ε':'E', 'Ζ':'Z', 'Η':'H', 'Θ':'8',
  22. 'Ι':'I', 'Κ':'K', 'Λ':'L', 'Μ':'M', 'Ν':'N', 'Ξ':'3', 'Ο':'O', 'Π':'P',
  23. 'Ρ':'R', 'Σ':'S', 'Τ':'T', 'Υ':'Y', 'Φ':'F', 'Χ':'X', 'Ψ':'PS', 'Ω':'W',
  24. 'Ά':'A', 'Έ':'E', 'Ί':'I', 'Ό':'O', 'Ύ':'Y', 'Ή':'H', 'Ώ':'W', 'Ϊ':'I',
  25. 'Ϋ':'Y'
  26. }
  27. var TURKISH_MAP = {
  28. 'ş':'s', 'Ş':'S', 'ı':'i', 'İ':'I', 'ç':'c', 'Ç':'C', 'ü':'u', 'Ü':'U',
  29. 'ö':'o', 'Ö':'O', 'ğ':'g', 'Ğ':'G'
  30. }
  31. var RUSSIAN_MAP = {
  32. 'а':'a', 'б':'b', 'в':'v', 'г':'g', 'д':'d', 'е':'e', 'ё':'yo', 'ж':'zh',
  33. 'з':'z', 'и':'i', 'й':'j', 'к':'k', 'л':'l', 'м':'m', 'н':'n', 'о':'o',
  34. 'п':'p', 'р':'r', 'с':'s', 'т':'t', 'у':'u', 'ф':'f', 'х':'h', 'ц':'c',
  35. 'ч':'ch', 'ш':'sh', 'щ':'sh', 'ъ':'', 'ы':'y', 'ь':'', 'э':'e', 'ю':'yu',
  36. 'я':'ya',
  37. 'А':'A', 'Б':'B', 'В':'V', 'Г':'G', 'Д':'D', 'Е':'E', 'Ё':'Yo', 'Ж':'Zh',
  38. 'З':'Z', 'И':'I', 'Й':'J', 'К':'K', 'Л':'L', 'М':'M', 'Н':'N', 'О':'O',
  39. 'П':'P', 'Р':'R', 'С':'S', 'Т':'T', 'У':'U', 'Ф':'F', 'Х':'H', 'Ц':'C',
  40. 'Ч':'Ch', 'Ш':'Sh', 'Щ':'Sh', 'Ъ':'', 'Ы':'Y', 'Ь':'', 'Э':'E', 'Ю':'Yu',
  41. 'Я':'Ya'
  42. }
  43. var UKRAINIAN_MAP = {
  44. 'Є':'Ye', 'І':'I', 'Ї':'Yi', 'Ґ':'G', 'є':'ye', 'і':'i', 'ї':'yi', 'ґ':'g'
  45. }
  46. var CZECH_MAP = {
  47. 'č':'c', 'ď':'d', 'ě':'e', 'ň': 'n', 'ř':'r', 'š':'s', 'ť':'t', 'ů':'u',
  48. 'ž':'z', 'Č':'C', 'Ď':'D', 'Ě':'E', 'Ň': 'N', 'Ř':'R', 'Š':'S', 'Ť':'T',
  49. 'Ů':'U', 'Ž':'Z'
  50. }
  51. var POLISH_MAP = {
  52. 'ą':'a', 'ć':'c', 'ę':'e', 'ł':'l', 'ń':'n', 'ó':'o', 'ś':'s', 'ź':'z',
  53. 'ż':'z', 'Ą':'A', 'Ć':'C', 'Ę':'e', 'Ł':'L', 'Ń':'N', 'Ó':'o', 'Ś':'S',
  54. 'Ź':'Z', 'Ż':'Z'
  55. }
  56. var LATVIAN_MAP = {
  57. 'ā':'a', 'č':'c', 'ē':'e', 'ģ':'g', 'ī':'i', 'ķ':'k', 'ļ':'l', 'ņ':'n',
  58. 'š':'s', 'ū':'u', 'ž':'z', 'Ā':'A', 'Č':'C', 'Ē':'E', 'Ģ':'G', 'Ī':'i',
  59. 'Ķ':'k', 'Ļ':'L', 'Ņ':'N', 'Š':'S', 'Ū':'u', 'Ž':'Z'
  60. }
  61. var ALL_DOWNCODE_MAPS=new Array()
  62. ALL_DOWNCODE_MAPS[0]=LATIN_MAP
  63. ALL_DOWNCODE_MAPS[1]=LATIN_SYMBOLS_MAP
  64. ALL_DOWNCODE_MAPS[2]=GREEK_MAP
  65. ALL_DOWNCODE_MAPS[3]=TURKISH_MAP
  66. ALL_DOWNCODE_MAPS[4]=RUSSIAN_MAP
  67. ALL_DOWNCODE_MAPS[5]=UKRAINIAN_MAP
  68. ALL_DOWNCODE_MAPS[6]=CZECH_MAP
  69. ALL_DOWNCODE_MAPS[7]=POLISH_MAP
  70. ALL_DOWNCODE_MAPS[8]=LATVIAN_MAP
  71. var Downcoder = new Object();
  72. Downcoder.Initialize = function()
  73. {
  74. if (Downcoder.map) // already made
  75. return ;
  76. Downcoder.map ={}
  77. Downcoder.chars = '' ;
  78. for(var i in ALL_DOWNCODE_MAPS)
  79. {
  80. var lookup = ALL_DOWNCODE_MAPS[i]
  81. for (var c in lookup)
  82. {
  83. Downcoder.map[c] = lookup[c] ;
  84. Downcoder.chars += c ;
  85. }
  86. }
  87. Downcoder.regex = new RegExp('[' + Downcoder.chars + ']|[^' + Downcoder.chars + ']+','g') ;
  88. }
  89. downcode= function( slug )
  90. {
  91. Downcoder.Initialize() ;
  92. var downcoded =""
  93. var pieces = slug.match(Downcoder.regex);
  94. if(pieces)
  95. {
  96. for (var i = 0 ; i < pieces.length ; i++)
  97. {
  98. if (pieces[i].length == 1)
  99. {
  100. var mapped = Downcoder.map[pieces[i]] ;
  101. if (mapped != null)
  102. {
  103. downcoded+=mapped;
  104. continue ;
  105. }
  106. }
  107. downcoded+=pieces[i];
  108. }
  109. }
  110. else
  111. {
  112. downcoded = slug;
  113. }
  114. return downcoded;
  115. }
  116. function URLify(s, num_chars) {
  117. // changes, e.g., "Petty theft" to "petty_theft"
  118. // remove all these words from the string before urlifying
  119. s = downcode(s);
  120. removelist = ["a", "an", "as", "at", "before", "but", "by", "for", "from",
  121. "is", "in", "into", "like", "of", "off", "on", "onto", "per",
  122. "since", "than", "the", "this", "that", "to", "up", "via",
  123. "with"];
  124. r = new RegExp('\\b(' + removelist.join('|') + ')\\b', 'gi');
  125. s = s.replace(r, '');
  126. // if downcode doesn't hit, the char will be stripped here
  127. s = s.replace(/[^-\w\s]/g, ''); // remove unneeded chars
  128. s = s.replace(/^\s+|\s+$/g, ''); // trim leading/trailing spaces
  129. s = s.replace(/[-\s]+/g, '-'); // convert spaces to hyphens
  130. s = s.toLowerCase(); // convert to lowercase
  131. return s.substring(0, num_chars);// trim to first num_chars chars
  132. }