Sorting languages by script.

* Clarified naming: separated regions from region groups.
* New utility functions for sorting and grouping languages.
* JSHint fixes.
* Added tests.

Change-Id: I8153a2b89fbc7b9a7d6b7deeb87208efc9f4d021

resources/ext.uls.core.js

@@ -13,6 +13,7 @@
 
 	ULS.prototype = {
 		constructor: ULS,
+
 		show: function() {
 			var pos = $.extend( {}, this.$element.offset(), {
 				height: this.$element[0].offsetHeight
@@ -29,20 +30,24 @@
 			$( 'input#languagefilter' ).focus();
 			return this;
 		},
+
 		hide: function() {
 			this.$menu.hide();
 			this.shown = false;
 			return this;
 		},
+
 		render: function() {
 			// Rendering stuff here
 		},
+
 		setLang: function( langCode ) {
 			// TODO: dependency on MediaWiki
 			var uri = new mw.Uri( window.location.href );
 			uri.extend( { setlang: langCode } );
 			window.location.href = uri.toString();
 		},
+
 		listen: function() {
 			var that = this;
 			// Register all event listeners to the ULS here.
@@ -73,6 +78,7 @@
 			// trigger a search for all languages.
 			$( "#languagefilter" ).languagefilter( "search" );
 		},
+
 		keyup: function( e ) {
 			switch( e.keyCode ) {
 				case 27: // escape
@@ -84,6 +90,7 @@
 			e.stopPropagation();
 			e.preventDefault();
 		},
+
 		keypress: function( e ) {
 			if ( !this.shown ) {
 				return;
@@ -96,6 +103,7 @@
 			}
 			e.stopPropagation();
 		},
+
 		click: function( e ) {
 			e.stopPropagation();
 			e.preventDefault();

resources/ext.uls.data.utils.js

@@ -1,11 +1,11 @@
-/*
+/**
  * @author Amir E. Aharoni
- * Utilities for querying language data.
+ * Utility functions for querying language data.
  */
 (function ( $ ) {
 	"use strict";
 
-	/*
+	/**
 	 * Returns the script of the language.
 	 * @param string language code
 	 * @return string
@@ -14,7 +14,7 @@
 		return $.uls.data.languages[language][0];
 	};
 
-	/*
+	/**
 	 * Returns the regions in which a language is spoken.
 	 * @param string language code
 	 * @return array of strings
@@ -23,7 +23,7 @@
 		return $.uls.data.languages[language][1];
 	};
 
-	/*
+	/**
 	 * Returns the autonym of the language.
 	 * @param string language code
 	 * @return string
@@ -32,7 +32,7 @@
 		return $.uls.data.autonyms[language];
 	};
 
-	/*
+	/**
 	 * Returns all languages written in script.
 	 * @param script string
 	 * @return array of strings (languages codes)
@@ -41,7 +41,7 @@
 		return $.uls.data.languagesInScripts( [ script ] );
 	};
 
-	/*
+	/**
 	 * Returns all languages written in the given scripts.
 	 * @param scripts array of strings
 	 * @return array of strings (languages codes)
@@ -61,7 +61,7 @@
 		return languagesInScripts;
 	};
 
-	/*
+	/**
 	 * Returns all languages in a given region.
 	 * @param region string
 	 * @return array of strings (languages codes)
@@ -70,7 +70,7 @@
 		return $.uls.data.languagesInRegions( [ region ] );
 	};
 
-	/*
+	/**
 	 * Returns all languages in given regions.
 	 * @param region array of strings.
 	 * @return array of strings (languages codes)
@@ -90,7 +90,16 @@
 		return languagesInRegions;
 	};
 
-	/*
+	/**
+	 * Returns all languages in a region group.
+	 * @param groupNum number.
+	 * @return array of strings (languages codes)
+	 */
+	$.uls.data.languagesInRegionGroup = function( groupNum ) {
+		return $.uls.data.languagesInRegions( $.uls.data.regionsInGroup( groupNum ) );
+	};
+
+	/**
 	 * Returns an associative array of languages in a region,
 	 * grouped by script.
 	 * @param string region code
@@ -112,7 +121,84 @@
 		return languagesByScriptInRegion;
 	};
 
-	/*
+	/**
+	 * Returns an associative array of languages in a region,
+	 * grouped by script group.
+	 * @param string region code
+	 * @return associative array
+	 */
+	$.uls.data.languagesByScriptGroupInRegion = function( region ) {
+		return $.uls.data.languagesByScriptGroupInRegions( [ region ] );
+	};
+
+	/**
+	 * Returns an associative array of languages in several regions,
+	 * grouped by script group.
+	 * @param array of strings - region codes
+	 * @return associative array
+	 */
+	$.uls.data.languagesByScriptGroupInRegions = function( regions ) {
+		var languagesByScriptGroupInRegions = {};
+
+		for ( var language in $.uls.data.languages ) {
+			for ( var i = 0; i < regions.length; i++ ) {
+				if ( $.inArray( regions[i], $.uls.data.regions( language ) ) !== -1 ) {
+					var scriptGroup = $.uls.data.scriptGroupOfLanguage( language );
+					if ( languagesByScriptGroupInRegions[scriptGroup] === undefined ) {
+						languagesByScriptGroupInRegions[scriptGroup] = [];
+					}
+					languagesByScriptGroupInRegions[scriptGroup].push( language );
+					break;
+				}
+			}
+		}
+
+		return languagesByScriptGroupInRegions;
+	};
+
+	/**
+	 * Returns an array of languages grouped by region group,
+	 * region, script group and script.
+	 * @return associative array
+	 */
+	$.uls.data.allLanguagesByRegionAndScript = function() {
+		var allLanguagesByRegionAndScript = {},
+			region,
+			regionGroup;
+
+		for ( region in $.uls.data.regiongroups ) {
+			regionGroup = $.uls.data.regiongroups[region];
+			if ( allLanguagesByRegionAndScript[regionGroup] === undefined ) {
+				allLanguagesByRegionAndScript[regionGroup] = {};
+			}
+			allLanguagesByRegionAndScript[regionGroup][region] = {};
+		}
+
+		for ( var language in $.uls.data.languages ) {
+			var script = $.uls.data.script( language );
+			var scriptGroup = $.uls.data.groupOfScript( script );
+			var regions = $.uls.data.regions( language );
+
+			for ( var regionNum = 0; regionNum < regions.length; regionNum++ ) {
+				region = regions[regionNum];
+				regionGroup = $.uls.data.regiongroups[region];
+
+				if ( allLanguagesByRegionAndScript[regionGroup][region][scriptGroup] === undefined ) {
+					allLanguagesByRegionAndScript[regionGroup][region][scriptGroup] = {};
+				}
+
+				if ( allLanguagesByRegionAndScript[regionGroup][region][scriptGroup][script] === undefined ) {
+					allLanguagesByRegionAndScript[regionGroup][region][scriptGroup][script] = [];
+				}
+
+				allLanguagesByRegionAndScript[regionGroup][region][scriptGroup][script].push( language );
+			}
+		}
+
+		return allLanguagesByRegionAndScript;
+	};
+
+	/**
 	 * Returns all regions in a region group.
 	 * @param number groupNum
 	 * @return array of strings
@@ -129,7 +215,7 @@
 		return regionsInGroup;
 	};
 
-	/*
+	/**
 	 * Returns the script group of a script or 'Other' if it doesn't
 	 * belong to any group.
 	 * @param string script code
@@ -145,9 +231,13 @@
 		return 'Other';
 	};
 
-	$.uls.data.sortByScriptGroup = function( languages ) {
+	/**
-		// FIXME sort it.
+	 * Returns the script group of a language.
-		return languages;
+	 * @param string language code
+	 * @return string script group name
+	 */
+	$.uls.data.scriptGroupOfLanguage = function( language ) {
+		return $.uls.data.groupOfScript( $.uls.data.script( language ) );
 	};
 
 } )( jQuery );

resources/ext.uls.languagefilter.js

@@ -21,8 +21,8 @@
 		listen: function() {
 			this.$element.on( 'keyup', $.proxy( this.keyup, this ));
 			if ( $.browser.webkit || $.browser.msie ) {
-				this.$element.on( 'keydown', $.proxy( this.keyup, this ) )
+				this.$element.on( 'keydown', $.proxy( this.keyup, this ) );
-			};
+			}
 		},
 
 		keyup: function( e ) {
@@ -34,23 +34,24 @@
 			var that = this;
 			var languages = this.options.languages;
 			var query = this.$element.val();
-			$.each( languages, function ( code, name ) {
+
+			$.each( languages, function ( langCode, name ) {
 				if ( query === "" ) {
-					that.render(code);
+					that.render( langCode );
 				}
-				else if ( that.filter( code, query ) ) {
+				else if ( that.filter( langCode, query ) ) {
-					that.render(code);
+					that.render( langCode );
 				}
 			} ) ;
 		},
 
-		render : function( code ) {
+		render: function( langCode ) {
 			var that = this;
 			var $target = this.options.$target;
 			if ( !$target ) {
 				return;
 			}
-			$target.append(code);
+			$target.append( langCode );
 		},
 
 		escapeRegex: function( value ) {
@@ -65,13 +66,14 @@
 		 * c) ISO 639 code match with search string.
 		 * d) ISO 15924 code for the script match the search string.
 		 */
-		filter: function( code, searchTerm ) {
+		filter: function( langCode, searchTerm ) {
 			// FIXME script is ISO 15924 code. We might need actual name of script.
 			var matcher = new RegExp( this.escapeRegex( searchTerm ), 'i' );
-			return matcher.test( this.options.languages[code] ) ||
+
-				matcher.test( $.uls.data.autonym( code ) ) ||
+			return matcher.test( this.options.languages[langCode] ) ||
-				matcher.test( code ) ||
+				matcher.test( $.uls.data.autonym( langCode ) ) ||
-				matcher.test( $.uls.data.script( code ) );
+				matcher.test( langCode ) ||
+				matcher.test( $.uls.data.script( langCode ) );
 		}
 
 	};
@@ -80,7 +82,7 @@
 		return this.each( function() {
 			var $this = $( this ),
 				data = $this.data( 'languagefilter' ),
-				options = typeof option == 'object' && option;
+				options = typeof option === 'object' && option;
 			if ( !data ) {
 				$this.data( 'languagefilter', ( data = new LanguageFilter( this, options ) ) );
 			}
@@ -91,17 +93,16 @@
 	};
 
 	$.fn.languagefilter.defaults = {
-			$target: null, // where to append the results
+		$target: null, // Where to append the results
-			languages: null, // languages as code:name format. default values is from data-languages
+		languages: null, // Languages as code:name format. Default values come from data.languages.
-			clickhandler: null,
+		clickhandler: null
 	};
 
 	$.fn.languagefilter.Constructor = LanguageFilter;
 
-
 	/* RegionSelector Plugin Definition */
 
-	/*
+	/**
 	 * Region selector is a language selector based on regions.
 	 * Usage: $( 'jqueryselector' ).regionselector( options );
 	 * The attached element should have data-regiongroup attribute
@@ -117,11 +118,11 @@
 
 	RegionSelector.prototype = {
 		constructor: RegionSelector,
+
 		test: function( langCode ) {
-			var that = this;
+			var that = this,
-			var languages = $.uls.data.languages;
+				regionGroups = $.uls.data.regiongroups,
-			var regionGroups = $.uls.data.regiongroups;
+				regions = $.uls.data.regions( langCode );
-			var regions = languages[langCode][1];
 			// 1. loop over all regiongroups - like {EU: 2, AF: 2, AS: 3 ...}
 			// 2. check that the region matches the active region group
 			// 3. if this language is included in that region, show it
@@ -133,11 +134,9 @@
 				}
 			} );
 		},
-		show: function() {
-			var that = this,
-				languages = $.uls.data.languages;
 
-			languages = $.uls.data.sortByScriptGroup( languages );
+		show: function() {
+			var that = this;
 
 			// Make the selected region (and it only) active
 			$( '.regionselector' ).removeClass( 'active' );
@@ -145,13 +144,20 @@
 
 			// Repopulate the list of languages
 			that.options.$target.empty();
-			$.each( languages, function( langCode, langDef ) {
+
-				that.test( langCode );
+			var regions = $.uls.data.regionsInGroup( that.regionGroup );
-			} );
+			var languagesInRegion = $.uls.data.languagesByScriptGroupInRegions( regions );
+			for ( var scriptGroup in languagesInRegion ) {
+				for ( var langNum = 0; langNum < languagesInRegion[scriptGroup].length; langNum++ ) {
+					that.test( languagesInRegion[scriptGroup][langNum] );
+				}
+			}
+
 			if ( that.options.callback ) {
 				that.options.callback.call();
 			}
 		},
+
 		render: function( langCode, region ) {
 			var $target = this.options.$target;
 			if ( !$target ) {
@@ -159,15 +165,18 @@
 			}
 			$target.append( langCode, region );
 		},
+
 		listen: function() {
 			this.$element.on( 'click', $.proxy( this.click, this ) );
 		},
+
 		click: function( e ) {
 			e.stopPropagation();
 			e.preventDefault();
 			this.show();
 		}
 	};
+
 	/* RegionSelector Plugin Definition */
 
 	$.fn.regionselector = function( option ) {
@@ -175,6 +184,7 @@
 			var $this = $( this ),
 				data = $this.data( 'regionselector' ),
 				options = typeof option === 'object' && option;
+
 			if ( !data ) {
 				$this.data( 'regionselector', ( data = new RegionSelector( this, options ) ) );
 			}
@@ -191,5 +201,4 @@
 
 	$.fn.regionselector.Constructor = RegionSelector;
 
-
 } )( jQuery );

resources/ext.uls.lcd.js

@@ -16,6 +16,7 @@
 			var that = this;
 			this.addToRegion( langCode, regionCode );
 		},
+
 		/**
 		 * Add the language to a region.
 		 * If the region parameter is given , add to that region alone
@@ -27,6 +28,10 @@
 			var that = this,
 				language = $.uls.data.languages[langCode];
 
+			// TODO: This shouldn't happen.
+			// This currently happens with some special cases, such as qqq and
+			// some codes that are incorrectly used in Wikimedia projects for
+			// historical reasons.
 			if ( !language ) {
 				console.log( "Definition for " + langCode + " was not found in the language database." );
 				return;
@@ -49,7 +54,10 @@
 					.append(
 						$( '<a>' ).prop( 'href', '#' ).html( langName )
 					);
-				that.getColumn( regionCode ).append( $li );
+
+				// Append the element to the column in the list
+				var column = that.getColumn( regionCode );
+				column.append( $li );
 
 				if ( that.options.clickhandler ) {
 					$li.click( function() {
@@ -60,12 +68,16 @@
 		},
 
 		getColumn: function( regionCode ) {
-			var $ul = $( "div#" + regionCode ).find( 'ul:last' );
+			var $divRegionCode = $( 'div#' + regionCode );
+
+			var $ul = $divRegionCode.find( 'ul:last' );
 			if ( $ul.length === 0 || $ul.find( 'li' ).length >= 10 ) {
 				$ul = $( '<ul>' );
-				$( 'div#' + regionCode ).append( $ul );
+				$divRegionCode.append( $ul );
 			}
-			$( 'div#' + regionCode ).show();
+
+			$divRegionCode.show();
+
 			return $ul;
 		},
 

tests/qunit/ext.uls.tests.js

@@ -3,10 +3,12 @@
  *
  * @file
  * @author Santhosh Thottingal
- * @copyright Copyright © 2012 Santhosh Thottingal
+ * @author Amir E. Aharoni
+ * @copyright Copyright © 2012 Santhosh Thottingal, Amir E. Aharoni
  * @license http://www.gnu.org/copyleft/gpl.html GNU General Public License 2.0 or later
  */
 ( function () {
+"use strict";
 
 module( "ext.uls", QUnit.newMwEnvironment() );
 
@@ -16,7 +18,7 @@ module( "ext.uls", QUnit.newMwEnvironment() );
  */
 var orphanScript = function () {
 	for ( var language in $.uls.data.languages ) {
-		var script = $.uls.data.languages[language][0];
+		var script = $.uls.data.script( language );
 		if ( $.uls.data.groupOfScript( script ) === 'Other' ) {
 			return script;
 		}
@@ -31,16 +33,28 @@ test( "-- Initial check", function() {
 } );
 
 test( "-- $.uls.data testing", function() {
-	expect( 12 );
+	expect( 15 );
 
 	// This test assumes that we don't want any scripts to be in the 'Other'
 	// group. Actually, this may become wrong some day.
 	strictEqual( orphanScript(), '', 'No orphan scripts found.' );
-	strictEqual( $.uls.data.groupOfScript( 'Beng' ), 'SouthAsian', 'Bengali script belongs to the SouthAsian group.' );
+	strictEqual(
+		$.uls.data.groupOfScript( 'Beng' ),
+		'SouthAsian',
+		'Bengali script belongs to the SouthAsian group.'
+	);
+	strictEqual(
+		$.uls.data.scriptGroupOfLanguage( 'iu' ),
+		'NativeAmerican',
+		'The script of the Inupiaq language belongs to the NativeAmerican group.'
+	);
 
 	strictEqual( $.uls.data.script( 'ii' ), 'Yiii', 'Correct script of the Yi language was selected' );
 	deepEqual( $.uls.data.regions( 'lzz' ), [ 'EU', 'ME' ], 'Correct regions of the Laz language were selected' );
 
+	var allLanguagesByRegionAndScript = $.uls.data.allLanguagesByRegionAndScript();
+	deepEqual( allLanguagesByRegionAndScript['3']['AS']['SouthEastAsian']['Bugi'], ['bug'], 'All languages in the Buginese script in Asia were selected' );
+
 	deepEqual( $.uls.data.languagesInRegion( 'AU' ), ["en", "en-gb", "hif", "hif-latn", "mi", "na"], "languages of region AU are selected correctly" );
 	deepEqual( $.uls.data.languagesInRegions( ['NA', 'WW'] ),
 		[
@@ -62,10 +76,13 @@ test( "-- $.uls.data testing", function() {
 	deepEqual( $.uls.data.regionsInGroup( 1 ), ["NA", "LA", "SA"], "regions in group 1 are selected correctly" );
 	deepEqual( $.uls.data.regionsInGroup( 4 ), ["WW"], "regions in group 4 are selected correctly" );
 
-	languagesByScriptInNA = $.uls.data.languagesByScriptInRegion( 'NA' );
+	var languagesByScriptInNA = $.uls.data.languagesByScriptInRegion( 'NA' );
 	deepEqual( languagesByScriptInNA['Cans'], ["cr", "ike-cans", "iu"], "correct languages in Cans in NA selected" );
 
 	strictEqual( $.uls.data.autonym( 'pa' ), 'ਪੰਜਾਬੀ', 'Correct autonym of the Punjabi language was selected' );
+
+	var languagesByScriptGroupInEMEA = $.uls.data.languagesByScriptGroupInRegions( $.uls.data.regionsInGroup( 2 ) );
+	deepEqual( languagesByScriptGroupInEMEA['WestCaucasian'], ['hy', 'ka', 'xmf'], 'Correct languages in WestCaucasian script group selected' );
 } );
 
 }() );