Upgrade to ExtJS 4.0.0 - Released 04/26/2011
[extjs.git] / src / util / Inflector.js
diff --git a/src/util/Inflector.js b/src/util/Inflector.js
new file mode 100644 (file)
index 0000000..7f401e0
--- /dev/null
@@ -0,0 +1,311 @@
+/**
+ * @class Ext.util.Inflector
+ * @extends Object
+ * <p>General purpose inflector class that {@link #pluralize pluralizes}, {@link #singularize singularizes} and 
+ * {@link #ordinalize ordinalizes} words. Sample usage:</p>
+ * 
+<pre><code>
+//turning singular words into plurals
+Ext.util.Inflector.pluralize('word'); //'words'
+Ext.util.Inflector.pluralize('person'); //'people'
+Ext.util.Inflector.pluralize('sheep'); //'sheep'
+
+//turning plurals into singulars
+Ext.util.Inflector.singularize('words'); //'word'
+Ext.util.Inflector.singularize('people'); //'person'
+Ext.util.Inflector.singularize('sheep'); //'sheep'
+
+//ordinalizing numbers
+Ext.util.Inflector.ordinalize(11); //"11th"
+Ext.util.Inflector.ordinalize(21); //"21th"
+Ext.util.Inflector.ordinalize(1043); //"1043rd"
+</code></pre>
+ * 
+ * <p><u>Customization</u></p>
+ * 
+ * <p>The Inflector comes with a default set of US English pluralization rules. These can be augmented with additional
+ * rules if the default rules do not meet your application's requirements, or swapped out entirely for other languages.
+ * Here is how we might add a rule that pluralizes "ox" to "oxen":</p>
+ * 
+<pre><code>
+Ext.util.Inflector.plural(/^(ox)$/i, "$1en");
+</code></pre>
+ * 
+ * <p>Each rule consists of two items - a regular expression that matches one or more rules, and a replacement string.
+ * In this case, the regular expression will only match the string "ox", and will replace that match with "oxen". 
+ * Here's how we could add the inverse rule:</p>
+ * 
+<pre><code>
+Ext.util.Inflector.singular(/^(ox)en$/i, "$1");
+</code></pre>
+ * 
+ * <p>Note that the ox/oxen rules are present by default.</p>
+ * 
+ * @singleton
+ */
+
+Ext.define('Ext.util.Inflector', {
+
+    /* Begin Definitions */
+
+    singleton: true,
+
+    /* End Definitions */
+
+    /**
+     * @private
+     * The registered plural tuples. Each item in the array should contain two items - the first must be a regular
+     * expression that matchers the singular form of a word, the second must be a String that replaces the matched
+     * part of the regular expression. This is managed by the {@link #plural} method.
+     * @property plurals
+     * @type Array
+     */
+    plurals: [
+        [(/(quiz)$/i),                "$1zes"  ],
+        [(/^(ox)$/i),                 "$1en"   ],
+        [(/([m|l])ouse$/i),           "$1ice"  ],
+        [(/(matr|vert|ind)ix|ex$/i),  "$1ices" ],
+        [(/(x|ch|ss|sh)$/i),          "$1es"   ],
+        [(/([^aeiouy]|qu)y$/i),       "$1ies"  ],
+        [(/(hive)$/i),                "$1s"    ],
+        [(/(?:([^f])fe|([lr])f)$/i),  "$1$2ves"],
+        [(/sis$/i),                   "ses"    ],
+        [(/([ti])um$/i),              "$1a"    ],
+        [(/(buffal|tomat|potat)o$/i), "$1oes"  ],
+        [(/(bu)s$/i),                 "$1ses"  ],
+        [(/(alias|status|sex)$/i),    "$1es"   ],
+        [(/(octop|vir)us$/i),         "$1i"    ],
+        [(/(ax|test)is$/i),           "$1es"   ],
+        [(/^person$/),                "people" ],
+        [(/^man$/),                   "men"    ],
+        [(/^(child)$/),               "$1ren"  ],
+        [(/s$/i),                     "s"      ],
+        [(/$/),                       "s"      ]
+    ],
+    
+    /**
+     * @private
+     * The set of registered singular matchers. Each item in the array should contain two items - the first must be a 
+     * regular expression that matches the plural form of a word, the second must be a String that replaces the 
+     * matched part of the regular expression. This is managed by the {@link #singular} method.
+     * @property singulars
+     * @type Array
+     */
+    singulars: [
+      [(/(quiz)zes$/i),                                                    "$1"     ],
+      [(/(matr)ices$/i),                                                   "$1ix"   ],
+      [(/(vert|ind)ices$/i),                                               "$1ex"   ],
+      [(/^(ox)en/i),                                                       "$1"     ],
+      [(/(alias|status)es$/i),                                             "$1"     ],
+      [(/(octop|vir)i$/i),                                                 "$1us"   ],
+      [(/(cris|ax|test)es$/i),                                             "$1is"   ],
+      [(/(shoe)s$/i),                                                      "$1"     ],
+      [(/(o)es$/i),                                                        "$1"     ],
+      [(/(bus)es$/i),                                                      "$1"     ],
+      [(/([m|l])ice$/i),                                                   "$1ouse" ],
+      [(/(x|ch|ss|sh)es$/i),                                               "$1"     ],
+      [(/(m)ovies$/i),                                                     "$1ovie" ],
+      [(/(s)eries$/i),                                                     "$1eries"],
+      [(/([^aeiouy]|qu)ies$/i),                                            "$1y"    ],
+      [(/([lr])ves$/i),                                                    "$1f"    ],
+      [(/(tive)s$/i),                                                      "$1"     ],
+      [(/(hive)s$/i),                                                      "$1"     ],
+      [(/([^f])ves$/i),                                                    "$1fe"   ],
+      [(/(^analy)ses$/i),                                                  "$1sis"  ],
+      [(/((a)naly|(b)a|(d)iagno|(p)arenthe|(p)rogno|(s)ynop|(t)he)ses$/i), "$1$2sis"],
+      [(/([ti])a$/i),                                                      "$1um"   ],
+      [(/(n)ews$/i),                                                       "$1ews"  ],
+      [(/people$/i),                                                       "person" ],
+      [(/s$/i),                                                            ""       ]
+    ],
+    
+    /**
+     * @private
+     * The registered uncountable words
+     * @property uncountable
+     * @type Array
+     */
+     uncountable: [
+        "sheep",
+        "fish",
+        "series",
+        "species",
+        "money",
+        "rice",
+        "information",
+        "equipment",
+        "grass",
+        "mud",
+        "offspring",
+        "deer",
+        "means"
+    ],
+    
+    /**
+     * Adds a new singularization rule to the Inflector. See the intro docs for more information
+     * @param {RegExp} matcher The matcher regex
+     * @param {String} replacer The replacement string, which can reference matches from the matcher argument
+     */
+    singular: function(matcher, replacer) {
+        this.singulars.unshift([matcher, replacer]);
+    },
+    
+    /**
+     * Adds a new pluralization rule to the Inflector. See the intro docs for more information
+     * @param {RegExp} matcher The matcher regex
+     * @param {String} replacer The replacement string, which can reference matches from the matcher argument
+     */
+    plural: function(matcher, replacer) {
+        this.plurals.unshift([matcher, replacer]);
+    },
+    
+    /**
+     * Removes all registered singularization rules
+     */
+    clearSingulars: function() {
+        this.singulars = [];
+    },
+    
+    /**
+     * Removes all registered pluralization rules
+     */
+    clearPlurals: function() {
+        this.plurals = [];
+    },
+    
+    /**
+     * Returns true if the given word is transnumeral (the word is its own singular and plural form - e.g. sheep, fish)
+     * @param {String} word The word to test
+     * @return {Boolean} True if the word is transnumeral
+     */
+    isTransnumeral: function(word) {
+        return Ext.Array.indexOf(this.uncountable, word) != -1;
+    },
+
+    /**
+     * Returns the pluralized form of a word (e.g. Ext.util.Inflector.pluralize('word') returns 'words')
+     * @param {String} word The word to pluralize
+     * @return {String} The pluralized form of the word
+     */
+    pluralize: function(word) {
+        if (this.isTransnumeral(word)) {
+            return word;
+        }
+
+        var plurals = this.plurals,
+            length  = plurals.length,
+            tuple, regex, i;
+        
+        for (i = 0; i < length; i++) {
+            tuple = plurals[i];
+            regex = tuple[0];
+            
+            if (regex == word || (regex.test && regex.test(word))) {
+                return word.replace(regex, tuple[1]);
+            }
+        }
+        
+        return word;
+    },
+    
+    /**
+     * Returns the singularized form of a word (e.g. Ext.util.Inflector.singularize('words') returns 'word')
+     * @param {String} word The word to singularize
+     * @return {String} The singularized form of the word
+     */
+    singularize: function(word) {
+        if (this.isTransnumeral(word)) {
+            return word;
+        }
+
+        var singulars = this.singulars,
+            length    = singulars.length,
+            tuple, regex, i;
+        
+        for (i = 0; i < length; i++) {
+            tuple = singulars[i];
+            regex = tuple[0];
+            
+            if (regex == word || (regex.test && regex.test(word))) {
+                return word.replace(regex, tuple[1]);
+            }
+        }
+        
+        return word;
+    },
+    
+    /**
+     * Returns the correct {@link Ext.data.Model Model} name for a given string. Mostly used internally by the data 
+     * package
+     * @param {String} word The word to classify
+     * @return {String} The classified version of the word
+     */
+    classify: function(word) {
+        return Ext.String.capitalize(this.singularize(word));
+    },
+    
+    /**
+     * Ordinalizes a given number by adding a prefix such as 'st', 'nd', 'rd' or 'th' based on the last digit of the 
+     * number. 21 -> 21st, 22 -> 22nd, 23 -> 23rd, 24 -> 24th etc
+     * @param {Number} number The number to ordinalize
+     * @return {String} The ordinalized number
+     */
+    ordinalize: function(number) {
+        var parsed = parseInt(number, 10),
+            mod10  = parsed % 10,
+            mod100 = parsed % 100;
+        
+        //11 through 13 are a special case
+        if (11 <= mod100 && mod100 <= 13) {
+            return number + "th";
+        } else {
+            switch(mod10) {
+                case 1 : return number + "st";
+                case 2 : return number + "nd";
+                case 3 : return number + "rd";
+                default: return number + "th";
+            }
+        }
+    }
+}, function() {
+    //aside from the rules above, there are a number of words that have irregular pluralization so we add them here
+    var irregulars = {
+            alumnus: 'alumni',
+            cactus : 'cacti',
+            focus  : 'foci',
+            nucleus: 'nuclei',
+            radius: 'radii',
+            stimulus: 'stimuli',
+            ellipsis: 'ellipses',
+            paralysis: 'paralyses',
+            oasis: 'oases',
+            appendix: 'appendices',
+            index: 'indexes',
+            beau: 'beaux',
+            bureau: 'bureaux',
+            tableau: 'tableaux',
+            woman: 'women',
+            child: 'children',
+            man: 'men',
+            corpus:    'corpora',
+            criterion: 'criteria',
+            curriculum:        'curricula',
+            genus: 'genera',
+            memorandum:        'memoranda',
+            phenomenon:        'phenomena',
+            foot: 'feet',
+            goose: 'geese',
+            tooth: 'teeth',
+            antenna: 'antennae',
+            formula: 'formulae',
+            nebula: 'nebulae',
+            vertebra: 'vertebrae',
+            vita: 'vitae'
+        },
+        singular;
+    
+    for (singular in irregulars) {
+        this.plural(singular, irregulars[singular]);
+        this.singular(irregulars[singular], singular);
+    }
+});
\ No newline at end of file