Mercurial > public > madeira

annotate media/django/js/urlify.js @ 1:0dcfcdf50c62

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Initial import of Madeira project from the private repository.
author Brian Neal <bgneal@gmail.com>
date Mon, 06 Apr 2009 03:10:59 +0000
parents
children
rev   line source
bgneal@1 1 var LATIN_MAP = {
bgneal@1 2 'À': 'A', 'Á': 'A', 'Â': 'A', 'Ã': 'A', 'Ä': 'A', 'Å': 'A', 'Æ': 'AE', 'Ç':
bgneal@1 3 'C', 'È': 'E', 'É': 'E', 'Ê': 'E', 'Ë': 'E', 'Ì': 'I', 'Í': 'I', 'Î': 'I',
bgneal@1 4 'Ï': 'I', 'Ð': 'D', 'Ñ': 'N', 'Ò': 'O', 'Ó': 'O', 'Ô': 'O', 'Õ': 'O', 'Ö':
bgneal@1 5 'O', 'Ő': 'O', 'Ø': 'O', 'Ù': 'U', 'Ú': 'U', 'Û': 'U', 'Ü': 'U', 'Ű': 'U',
bgneal@1 6 'Ý': 'Y', 'Þ': 'TH', 'ß': 'ss', 'à':'a', 'á':'a', 'â': 'a', 'ã': 'a', 'ä':
bgneal@1 7 'a', 'å': 'a', 'æ': 'ae', 'ç': 'c', 'è': 'e', 'é': 'e', 'ê': 'e', 'ë': 'e',
bgneal@1 8 'ì': 'i', 'í': 'i', 'î': 'i', 'ï': 'i', 'ð': 'd', 'ñ': 'n', 'ò': 'o', 'ó':
bgneal@1 9 'o', 'ô': 'o', 'õ': 'o', 'ö': 'o', 'ő': 'o', 'ø': 'o', 'ù': 'u', 'ú': 'u',
bgneal@1 10 'û': 'u', 'ü': 'u', 'ű': 'u', 'ý': 'y', 'þ': 'th', 'ÿ': 'y'
bgneal@1 11 }
bgneal@1 12 var LATIN_SYMBOLS_MAP = {
bgneal@1 13 '©':'(c)'
bgneal@1 14 }
bgneal@1 15 var GREEK_MAP = {
bgneal@1 16 'α':'a', 'β':'b', 'γ':'g', 'δ':'d', 'ε':'e', 'ζ':'z', 'η':'h', 'θ':'8',
bgneal@1 17 'ι':'i', 'κ':'k', 'λ':'l', 'μ':'m', 'ν':'n', 'ξ':'3', 'ο':'o', 'π':'p',
bgneal@1 18 'ρ':'r', 'σ':'s', 'τ':'t', 'υ':'y', 'φ':'f', 'χ':'x', 'ψ':'ps', 'ω':'w',
bgneal@1 19 'ά':'a', 'έ':'e', 'ί':'i', 'ό':'o', 'ύ':'y', 'ή':'h', 'ώ':'w', 'ς':'s',
bgneal@1 20 'ϊ':'i', 'ΰ':'y', 'ϋ':'y', 'ΐ':'i',
bgneal@1 21 'Α':'A', 'Β':'B', 'Γ':'G', 'Δ':'D', 'Ε':'E', 'Ζ':'Z', 'Η':'H', 'Θ':'8',
bgneal@1 22 'Ι':'I', 'Κ':'K', 'Λ':'L', 'Μ':'M', 'Ν':'N', 'Ξ':'3', 'Ο':'O', 'Π':'P',
bgneal@1 23 'Ρ':'R', 'Σ':'S', 'Τ':'T', 'Υ':'Y', 'Φ':'F', 'Χ':'X', 'Ψ':'PS', 'Ω':'W',
bgneal@1 24 'Ά':'A', 'Έ':'E', 'Ί':'I', 'Ό':'O', 'Ύ':'Y', 'Ή':'H', 'Ώ':'W', 'Ϊ':'I',
bgneal@1 25 'Ϋ':'Y'
bgneal@1 26 }
bgneal@1 27 var TURKISH_MAP = {
bgneal@1 28 'ş':'s', 'Ş':'S', 'ı':'i', 'İ':'I', 'ç':'c', 'Ç':'C', 'ü':'u', 'Ü':'U',
bgneal@1 29 'ö':'o', 'Ö':'O', 'ğ':'g', 'Ğ':'G'
bgneal@1 30 }
bgneal@1 31 var RUSSIAN_MAP = {
bgneal@1 32 'а':'a', 'б':'b', 'в':'v', 'г':'g', 'д':'d', 'е':'e', 'ё':'yo', 'ж':'zh',
bgneal@1 33 'з':'z', 'и':'i', 'й':'j', 'к':'k', 'л':'l', 'м':'m', 'н':'n', 'о':'o',
bgneal@1 34 'п':'p', 'р':'r', 'с':'s', 'т':'t', 'у':'u', 'ф':'f', 'х':'h', 'ц':'c',
bgneal@1 35 'ч':'ch', 'ш':'sh', 'щ':'sh', 'ъ':'', 'ы':'y', 'ь':'', 'э':'e', 'ю':'yu',
bgneal@1 36 'я':'ya',
bgneal@1 37 'А':'A', 'Б':'B', 'В':'V', 'Г':'G', 'Д':'D', 'Е':'E', 'Ё':'Yo', 'Ж':'Zh',
bgneal@1 38 'З':'Z', 'И':'I', 'Й':'J', 'К':'K', 'Л':'L', 'М':'M', 'Н':'N', 'О':'O',
bgneal@1 39 'П':'P', 'Р':'R', 'С':'S', 'Т':'T', 'У':'U', 'Ф':'F', 'Х':'H', 'Ц':'C',
bgneal@1 40 'Ч':'Ch', 'Ш':'Sh', 'Щ':'Sh', 'Ъ':'', 'Ы':'Y', 'Ь':'', 'Э':'E', 'Ю':'Yu',
bgneal@1 41 'Я':'Ya'
bgneal@1 42 }
bgneal@1 43 var UKRAINIAN_MAP = {
bgneal@1 44 'Є':'Ye', 'І':'I', 'Ї':'Yi', 'Ґ':'G', 'є':'ye', 'і':'i', 'ї':'yi', 'ґ':'g'
bgneal@1 45 }
bgneal@1 46 var CZECH_MAP = {
bgneal@1 47 'č':'c', 'ď':'d', 'ě':'e', 'ň': 'n', 'ř':'r', 'š':'s', 'ť':'t', 'ů':'u',
bgneal@1 48 'ž':'z', 'Č':'C', 'Ď':'D', 'Ě':'E', 'Ň': 'N', 'Ř':'R', 'Š':'S', 'Ť':'T',
bgneal@1 49 'Ů':'U', 'Ž':'Z'
bgneal@1 50 }
bgneal@1 51
bgneal@1 52 var POLISH_MAP = {
bgneal@1 53 'ą':'a', 'ć':'c', 'ę':'e', 'ł':'l', 'ń':'n', 'ó':'o', 'ś':'s', 'ź':'z',
bgneal@1 54 'ż':'z', 'Ą':'A', 'Ć':'C', 'Ę':'e', 'Ł':'L', 'Ń':'N', 'Ó':'o', 'Ś':'S',
bgneal@1 55 'Ź':'Z', 'Ż':'Z'
bgneal@1 56 }
bgneal@1 57
bgneal@1 58 var LATVIAN_MAP = {
bgneal@1 59 'ā':'a', 'č':'c', 'ē':'e', 'ģ':'g', 'ī':'i', 'ķ':'k', 'ļ':'l', 'ņ':'n',
bgneal@1 60 'š':'s', 'ū':'u', 'ž':'z', 'Ā':'A', 'Č':'C', 'Ē':'E', 'Ģ':'G', 'Ī':'i',
bgneal@1 61 'Ķ':'k', 'Ļ':'L', 'Ņ':'N', 'Š':'S', 'Ū':'u', 'Ž':'Z'
bgneal@1 62 }
bgneal@1 63
bgneal@1 64 var ALL_DOWNCODE_MAPS=new Array()
bgneal@1 65 ALL_DOWNCODE_MAPS[0]=LATIN_MAP
bgneal@1 66 ALL_DOWNCODE_MAPS[1]=LATIN_SYMBOLS_MAP
bgneal@1 67 ALL_DOWNCODE_MAPS[2]=GREEK_MAP
bgneal@1 68 ALL_DOWNCODE_MAPS[3]=TURKISH_MAP
bgneal@1 69 ALL_DOWNCODE_MAPS[4]=RUSSIAN_MAP
bgneal@1 70 ALL_DOWNCODE_MAPS[5]=UKRAINIAN_MAP
bgneal@1 71 ALL_DOWNCODE_MAPS[6]=CZECH_MAP
bgneal@1 72 ALL_DOWNCODE_MAPS[7]=POLISH_MAP
bgneal@1 73 ALL_DOWNCODE_MAPS[8]=LATVIAN_MAP
bgneal@1 74
bgneal@1 75 var Downcoder = new Object();
bgneal@1 76 Downcoder.Initialize = function()
bgneal@1 77 {
bgneal@1 78 if (Downcoder.map) // already made
bgneal@1 79 return ;
bgneal@1 80 Downcoder.map ={}
bgneal@1 81 Downcoder.chars = '' ;
bgneal@1 82 for(var i in ALL_DOWNCODE_MAPS)
bgneal@1 83 {
bgneal@1 84 var lookup = ALL_DOWNCODE_MAPS[i]
bgneal@1 85 for (var c in lookup)
bgneal@1 86 {
bgneal@1 87 Downcoder.map[c] = lookup[c] ;
bgneal@1 88 Downcoder.chars += c ;
bgneal@1 89 }
bgneal@1 90 }
bgneal@1 91 Downcoder.regex = new RegExp('[' + Downcoder.chars + ']|[^' + Downcoder.chars + ']+','g') ;
bgneal@1 92 }
bgneal@1 93
bgneal@1 94 downcode= function( slug )
bgneal@1 95 {
bgneal@1 96 Downcoder.Initialize() ;
bgneal@1 97 var downcoded =""
bgneal@1 98 var pieces = slug.match(Downcoder.regex);
bgneal@1 99 if(pieces)
bgneal@1 100 {
bgneal@1 101 for (var i = 0 ; i < pieces.length ; i++)
bgneal@1 102 {
bgneal@1 103 if (pieces[i].length == 1)
bgneal@1 104 {
bgneal@1 105 var mapped = Downcoder.map[pieces[i]] ;
bgneal@1 106 if (mapped != null)
bgneal@1 107 {
bgneal@1 108 downcoded+=mapped;
bgneal@1 109 continue ;
bgneal@1 110 }
bgneal@1 111 }
bgneal@1 112 downcoded+=pieces[i];
bgneal@1 113 }
bgneal@1 114 }
bgneal@1 115 else
bgneal@1 116 {
bgneal@1 117 downcoded = slug;
bgneal@1 118 }
bgneal@1 119 return downcoded;
bgneal@1 120 }
bgneal@1 121
bgneal@1 122
bgneal@1 123 function URLify(s, num_chars) {
bgneal@1 124 // changes, e.g., "Petty theft" to "petty_theft"
bgneal@1 125 // remove all these words from the string before urlifying
bgneal@1 126 s = downcode(s);
bgneal@1 127 removelist = ["a", "an", "as", "at", "before", "but", "by", "for", "from",
bgneal@1 128 "is", "in", "into", "like", "of", "off", "on", "onto", "per",
bgneal@1 129 "since", "than", "the", "this", "that", "to", "up", "via",
bgneal@1 130 "with"];
bgneal@1 131 r = new RegExp('\\b(' + removelist.join('|') + ')\\b', 'gi');
bgneal@1 132 s = s.replace(r, '');
bgneal@1 133 // if downcode doesn't hit, the char will be stripped here
bgneal@1 134 s = s.replace(/[^-\w\s]/g, ''); // remove unneeded chars
bgneal@1 135 s = s.replace(/^\s+|\s+$/g, ''); // trim leading/trailing spaces
bgneal@1 136 s = s.replace(/[-\s]+/g, '-'); // convert spaces to hyphens
bgneal@1 137 s = s.toLowerCase(); // convert to lowercase
bgneal@1 138 return s.substring(0, num_chars);// trim to first num_chars chars
bgneal@1 139 }
bgneal@1 140