MediaWiki:ReferrerWikiUselang.js

//<source lang="javascript">

/**
* Makes a best guess at the user language.
* 
* Refreshes the page to the best guess language
* Makes the uselang parameter persistent among page links to commons
*/
var persistentUselang ={
	// The supported list of projects that include language codes at the start of their url
	'projectList' : [ 'wikipedia' , 'wiktionary' , 'wikinews' , 'wikibooks' , 'wikisource' , 'wikiversity' , 'wikiquote' ],


	// The supported list of language codes, We list out the languages here to avoid an extra api request.
	// Api request to get this list: 
	// http://commons.wikimedia.org/w/api.php?action=query&meta=siteinfo&siprop=languages&format=jsonfm

	'rewriteLanguageList' : ["aa", "ab", "ace", "af", "ak", "aln", "als", "am", "an", "ang", "ar", "arc", 
 "arn", "arz", "as", "ast", "av", "avk", "ay", "az", "ba", "bar", "bat-smg", "bcc", "bcl", "be", 
 "be-tarask", "be-x-old", "bg", "bh", "bi", "bm", "bn", "bo", "bpy", "bqi", "br", "bs", "bug", "bxr",
 "ca", "cbk-zam", "cdo", "ce", "ceb", "ch", "cho", "chr", "chy", "ckb", "ckb-latn", "ckb-arab", "co", 
 "cr", "crh", "crh-latn", "crh-cyrl", "cs", "csb", "cu", "cv", "cy", "da", "de", "de-at", "de-ch", 
 "de-formal", "diq", "dk", "dsb", "dv", "dz", "ee", "el", "eml", "en", "en-gb", "eo", "es", "et", 
 "eu", "ext", "fa", "ff", "fi", "fiu-vro", "fj", "fo", "fr", "frc", "frp", "fur", "fy", "ga", "gag", 
 "gan", "gan-hans", "gan-hant", "gd", "gl", "glk", "gn", "got", "grc", "gsw", "gu", "gv", "ha", 
 "hak", "haw", "he", "hi", "hif", "hif-deva", "hif-latn", "hil", "ho", "hr", "hsb", "ht", "hu", 
 "hy", "hz", "ia", "id", "ie", "ig", "ii", "ik", "ike-cans", "ike-latn", "ilo", "inh", "io", "is", 
 "it", "iu", "ja", "jbo", "jut", "jv", "ka", "kaa", "kab", "kg", "ki", "kiu", "kj", "kk", "kk-arab",
 "kk-cyrl", "kk-latn", "kk-cn", "kk-kz", "kk-tr", "kl", "km", "kn", "ko", "ko-kp", "kr", "kri", 
 "krj", "ks", "ksh", "ku", "ku-latn", "ku-arab", "kv", "kw", "ky", "la", "lad", "lb", "lbe", "lez", 
 "lfn", "lg", "li", "lij", "lmo", "ln", "lo", "loz", "lt", "lv", "lzh", "mai", "map-bms", "mdf", 
 "mg", "mh", "mhr", "mi", "mk", "ml", "mn", "mo", "mr", "ms", "mt", "mus", "mwl", "my", "myv", 
 "mzn", "na", "nah", "nan", "nap", "nb", "nds", "nds-nl", "ne", "new", "ng", "niu", "nl", "nn", 
 "no", "nov", "nrm", "nso", "nv", "ny", "oc", "om", "or", "os", "pa", "pag", "pam", "pap", "pcd",
 "pdc", "pdt", "pfl", "pi", "pih", "pl", "pms", "pnb", "pnt", "ps", "pt", "pt-br", "qu", "rif",
 "rm", "rmy", "rn", "ro", "roa-rup", "roa-tara", "ru", "ruq", "ruq-cyrl", "ruq-latn", "rw", "sa",
 "sah", "sc", "scn", "sco", "sd", "sdc", "se", "sei", "sg", "sh", "shi", "si", "simple", "sk", 
 "sl", "sli", "sm", "sma", "sn", "so", "sq", "sr", "sr-ec", "sr-el", "srn", "ss", "st", "stq", 
 "su", "sv", "sw", "szl", "ta", "tcy", "te", "tet", "tg", "tg-cyrl", "tg-latn", "th", "ti", "tk", 
 "tl", "tn", "to", "tokipona", "tp", "tpi", "tr", "ts", "tt", "tt-cyrl", "tt-latn", "tum", "tw",
 "ty", "tyv", "udm", "ug", "ug-arab", "ug-latn", "uk", "ur", "uz", "ve", "vec", "vep", "vi", "vls",
 "vo", "vro", "wa", "war", "wo", "wuu", "xal", "xh", "xmf", "yi", "yo", "yue", "za", "zea", "zh", 
 "zh-classical", "zh-cn", "zh-hans", "zh-hant", "zh-hk", "zh-min-nan", "zh-mo", "zh-my", "zh-sg", 
 "zh-tw", "zh-yue", "zu" ],

	/**
	 *  Run the page link rewrites and or refresh the page
	 */
	'setPageLang': function(){	
		// Check if we need to refresh the page 
		if( this.checkLanguageRefresh() ){
			// page is being refreshed ( end all script execution )
			return true;
		}
		// try to rewrite all the links in the page to keep uselang persistent
		return this.rewritePageLinks();
	},
	/**
	 * Checks the rewrite page link mode and calls the doRewritePageLinks() link
	 */
	'rewritePageLinks': function(){	
		var _this = this;
		var useLang = this.getUseLang();				
		// Check if we have a uselang to rewrite links for:
		if( !useLang ){
			// @@TODO We could do an ajax call to the geoIP service here		
			/*$j.getScript('http://geoiplookup.wikimedia.org/', function(){
				// NOTE geoIP should guess language from location. 
				_this.doRewritePageLinks( useLang );
			})
			*/
			return false;
		}
		this.doRewritePageLinks( useLang );
	},
	/**
	 * Rewrites all the links in the page appending uselang argument
	 * @param {String} useLang
	 * 		The language code to rewrite links for
	 */
	'doRewritePageLinks': function( useLang ){
		if( ! this.isSupportedRewriteLanguage( useLang ) ){
			return false;
		}
		// Wait until the document is ready before checking links in the page: 
		$j(document).ready(function(){
			// Note uselang hacks should not be affected because you have to be logged in to the upload files and hit uselang="ownwork"
			$j('a').each(function( inx, node ){
				var linkUrl = $j( node ).attr('href');
				// Only rewrite commons and relative urls:
				if( linkUrl && linkUrl.match( /^http\:\/\/commons.wikimedia.org\/|^\// ) ){
					// Only rewrite urls that don't already include uselang 
					if( linkUrl.indexOf( 'uselang=' == -1 ) ) {
						// Question marks should be escaped in urls except for the arguments separator
						// check if question mark is in the url: 
						var paramType = ( linkUrl.indexOf( '?' ) == -1 )? '?' : '&';
						$j( node ).attr('href',  linkUrl + paramType + 'uselang=' + useLang );
					};
				};
			});
		});
	},
	/**
	 * Checks if the language is supported
	 * @param {string} uselang
	 * 		The language code to check
	 * @return {boolean}
	 * 		true if language is supported
	 * 		false if the language is not supported
	 */
	'isSupportedRewriteLanguage': function( useLang ){
		// Check that the language is supported:
		if( $j.inArray( useLang, this.rewriteLanguageList ) == -1 ){
			return false;
		}
		// Just in case someone did not see documentation above ( don't rewrite en )
		if( useLang == 'en' ){
			return false;
		}
		return true;
	},
	
	/**
	 * Checks if the page should be refreshed to include the uselang param
	 */
	'checkLanguageRefresh': function(){
		var useLang = this.getUseLang();
		// Check if we need to refresh the browser url ( browser language or document.referrer ) 
		if( useLang && !getParamValue( 'uselang' ) && this.isSupportedRewriteLanguage( useLang ) ){	
			var paramType = ( document.URL.indexOf( '?' ) == -1 ) ? '?' : '&';
			window.location.href = document.URL + paramType + 'uselang=' + useLang;
			return true;
		}
		return false;
	},
	/**
	 * Master function to get the user language code. 
	 */
	'getUseLang': function(){		
		// Don't do anything for logged in users ( respect the user preference ) 
		if( wgUserName ){
			this.useLang = false;
		}
		
		// Check if the page already has a uselang parameter in the url use that.
		if( getParamValue( 'uselang' ) ){
			return getParamValue( 'uselang' )
		}
		// Check the referrer url to set the use lang.
		if( this.getReferreLang() ){
			return this.getReferreLang();
			
		}
		// Check for the user browser language
		if( this.getBrowserLang() ){
			return this.getBrowserLang();
		}
		return false;
	},
	/**
	 * Get the language code from wikimedia site the user came from
	 * @return {String}
	 * 		the referrer language code
	 */
	'getReferreLang': function(){
		if( this.referrerLang ){
			return this.referrerLang;
		}		 
		if( document.referrer ){
			// We should use mw.parseUri ( for now regEx )
			var matches = document.referrer.match(/^http\:\/\/([^\.]*)\.([^\.]*)/);
			if( matches && jQuery.inArray( matches[2], this.projectList) != -1 ){
				// FIXME add in refresh url to add uselang on links from other wikipedias
				this.referrerLang = matches[1];
			}
		}
		return this.referrerLang;
	},
	/**
	 * Get the user Browser language code
	 * @return {String}
	 * 		the browser language code
	 */
	'getBrowserLang': function(){
		if( this.browserLang ){
			return this.browserLang;
		}
		if( window.navigator.language ){
			// fix en-US -> en while keeping all lower case language keys zh-hans
			// NOTE: this assumes country codes are always upper case while language codes are all lower case .
			var browserLang = window.navigator.language;
			var matches = browserLang.match( /([a-z]+)(\-[a-z]+)?/ );
			//return window.navigator.language;
			this.browserLang = matches[1] + ( matches[2] )? matches[2] : '';		
		}
		return this.browserLang;
	}
}
// Run the persistentUselang set page lang function
persistentUselang.setPageLang();

// </source>