SPARQL Query to other SPARQL Endpoints

[WaltonG]

So far we can sucessfully send a SPARQL query to wikidata. Currently am working on sending a SPARQL query to any SPARQL enpoint. I was thinking of programmatically changing the below line with the endpoint url provided by the user.

SparqlExtension/module/scripts/index/wikidata/Sparql.js

Line 8 in b14a6a6

var SPARQL_SERVICE_URI = 'https://query.wikidata.org/bigdata/namespace/wdq/sparql',

The challenge am encountering is status code 307-internal redirect. I think its caused by

SparqlExtension/module/scripts/index/wikidata/Sparql.js

Lines 95 to 96 in b14a6a6

	query = encodeURI( 'prefix schema: <http://schema.org/> '
	+ 'SELECT * WHERE {<http://www.wikidata.org> schema:dateModified ?y}' ),

Am not sure how to edit the line so that it can accomodate any SPARQL endpoint

[Abbe98]

I don't think there is anything in the SPARQL standard(s) allowing one to check when the data was updated. Maybe there is an extension to the SPARQL 1.1 Service Description or the SPARQL 1.1 Protocol standards which expose such data but there is no way to use the SPARQL 1.1 Query Language for such a check.

That said, isn't update lag a very Wikibase(or even Wikidata) specific issue?

[WaltonG]

Thanks for the heads up. Am getting somewhere as now am receiving status code 200

[ostephens]

What does the nocache parameter do? How does it compare to using a cache-control: no-cache header on the http request?

[WaltonG]

I think its used to avoid browser caching. However i have managed to query two enpoints https://lod.openlinksw.com/sparql and http://dbpedia.org/sparql by editing the below line

SparqlExtension/module/scripts/index/wikidata/Sparql.js

Line 8 in b14a6a6

var SPARQL_SERVICE_URI = 'https://query.wikidata.org/bigdata/namespace/wdq/sparql',

Users need to provide the correct SPARQL endpoint URI for the query to be successful.

As @Abbe98 stated i think the below code is not needed for a query to be successful

SparqlExtension/module/scripts/index/wikidata/Sparql.js

Lines 92 to 127 in b14a6a6

	SELF.prototype.queryDataUpdatedTime = function() {
	// Cache the update time only for a minute
	var deferred = $.Deferred(),
	query = encodeURI( 'prefix schema: <http://schema.org/> '
	+ 'SELECT * WHERE {<http://www.wikidata.org> schema:dateModified ?y}' ),
	url = this._serviceUri + '?query=' + query + '&nocache='
	+ Math.floor( Date.now() / 60000 ),
	settings = {
	headers: {
	Accept: 'application/sparql-results+json'
	}
	};
	$.ajax( url, settings ).done( function( data, textStatus, jqXHR ) {
	if ( !data.results.bindings[0] ) {
	deferred.reject();
	return;
	}
	var updateDate = new Date( data.results.bindings[0][data.head.vars[0]].value ),
	dateText = updateDate.toLocaleTimeString( navigator.language, {
	timeZoneName: 'short'
	} ) + ', ' + updateDate.toLocaleDateString( navigator.language, {
	month: 'short',
	day: 'numeric',
	year: 'numeric'
	} ),
	differenceInSeconds = Math.round( ( new Date() - updateDate ) / 1000 );
	deferred.resolve( dateText, differenceInSeconds );
	} ).fail( function() {
	deferred.reject();
	} );
	return deferred;
	};

[Abbe98]

Another thing that might be worth considering would be that if SPARQL queries are done from JavaScript/the client like above, endpoints not supporting CORS won't work and there will likely be SSL-related issues as well.

[WaltonG]

Sure ill have to do some enhancement and utilize the Openrefine HTTPClient