3 Copyright 2012 Yahoo! Inc. All rights reserved.
4 Licensed under the BSD License.
5 http://yuilibrary.com/license/
7 YUI.add('text-accentfold', function(Y) {
17 * Provides a basic accent folding implementation that converts common accented
18 * letters (like "á") to their non-accented forms (like "a").
21 * @submodule text-accentfold
26 * Provides a basic accent folding implementation that converts common accented
27 * letters (like "á") to their non-accented forms (like "a").
31 * This implementation is not comprehensive, and should only be used as a last
32 * resort when accent folding can't be done on the server. A comprehensive
33 * accent folding implementation would require much more character data to be
34 * sent to the browser, resulting in a significant performance penalty. This
35 * implementation strives for a compromise between usefulness and performance.
39 * Accent folding is a destructive operation that can't be reversed, and may
40 * change or destroy the actual meaning of the text depending on the language.
41 * It should not be used on strings that will later be displayed to a user,
42 * unless this is done with the understanding that linguistic meaning may be
43 * lost and that you may in fact confuse or insult the user by doing so.
47 * When used for matching, accent folding is likely to produce erroneous matches
48 * for languages in which characters with diacritics are considered different
49 * from their base characters, or where correct folding would map to other
50 * character sequences than just stripped characters. For example, in German
51 * "ü" is a character that's clearly different from "u" and should match "ue"
52 * instead. The word "betrügen" means "to defraud", while "betrugen" is the past
53 * tense of "to behave". The name "Müller" is expected to match "Mueller", but
54 * not "Muller". On the other hand, accent folding falls short for languages
55 * where different base characters are expected to match. In Japanese, for
56 * example, hiragana and katakana characters with the same pronunciation ("あ"
57 * and "ア") are commonly treated as equivalent for lookups, but accent folding
58 * treats them as different.
61 * @class Text.AccentFold
67 FoldData = Text.Data.AccentFold,
70 // -- Public Static Methods ------------------------------------------------
73 * Returns <code>true</code> if the specified string contains one or more
74 * characters that can be folded, <code>false</code> otherwise.
77 * @param {String} string String to test.
81 canFold: function (string) {
84 for (letter in FoldData) {
85 if (FoldData.hasOwnProperty(letter) &&
86 string.search(FoldData[letter]) !== -1) {
95 * Compares the accent-folded versions of two strings and returns
96 * <code>true</code> if they're the same, <code>false</code> otherwise. If
97 * a custom comparison function is supplied, the accent-folded strings will
98 * be passed to that function for comparison.
101 * @param {String} a First string to compare.
102 * @param {String} b Second string to compare.
103 * @param {Function} func (optional) Custom comparison function. Should
104 * return a truthy or falsy value.
105 * @return {Boolean} Results of the comparison.
108 compare: function (a, b, func) {
109 var aFolded = AccentFold.fold(a),
110 bFolded = AccentFold.fold(b);
112 return func ? !!func(aFolded, bFolded) : aFolded === bFolded;
117 * Returns a copy of <em>haystack</em> containing only the strings for which
118 * the supplied function returns <code>true</code>.
122 * While comparisons will be made using accent-folded strings, the returned
123 * array of matches will contain the original strings that were passed in.
127 * @param {Array} haystack Array of strings to filter.
128 * @param {Function} func Comparison function. Will receive an accent-folded
129 * haystack string as an argument, and should return a truthy or falsy
131 * @return {Array} Filtered copy of <em>haystack</em>.
134 filter: function (haystack, func) {
135 return YArray.filter(haystack, function (item) {
136 return func(AccentFold.fold(item));
141 * Accent-folds the specified string or array of strings and returns a copy
142 * in which common accented letters have been converted to their closest
143 * non-accented, lowercase forms.
146 * @param {String|Array} input String or array of strings to be folded.
147 * @return {String|Array} Folded string or array of strings.
150 fold: function (input) {
151 if (Y.Lang.isArray(input)) {
152 return YArray.map(input, AccentFold.fold);
155 input = input.toLowerCase();
157 Y.Object.each(FoldData, function (regex, letter) {
158 input = input.replace(regex, letter);
165 Text.AccentFold = AccentFold;
168 }, '3.5.1' ,{requires:['array-extras', 'text-data-accentfold']});