Just a quick chase up from myself as well with regards the following, particularly finding that the lack of serialization support in sparql results is limiting things a bit (can't expose a public sparql endpoint):
any word / update / eta? Regards! Nathan wrote: > Hi Ivan, > > I think I explained badly in the first place.. when one uploads XML > contained in a CDATA section (via PUT to a rdf_sink folder), the XML > contained is converted to (and stored as) a string literal, rather than > an XMLLiteral. > > Need to do a bit of explaining here, as I've noted a few issues. > > one can save snippets of XML in rdf triples in the following ways: > > > Option 1: RDF+XML w/ CDATA > <content:encoded><![CDATA[<p>some <strong>html</strong> in > here.</p>]]></content:encoded> > > > Option 2: rdf:XMLLiteral in turtle/n3 > <http://something.com/anyoluri> > <content:encoded> """<p xmlns="http://www.w3.org/1999/xhtml";>some > <strong xmlns="http://www.w3.org/1999/xhtml";>html</strong> in > here.</p>"""^^rdf:XMLLiteral . > > > Option 3: XML+RDF w/ namespace and embedded XML > <?xml version="1.0" encoding="UTF-8"?> > <rdf:RDF > xmlns="http://www.w3.org/1999/xhtml"; > xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"; > xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"; > xmlns:content="http://purl.org/rss/1.0/modules/content/"; > xmlns:sioc="http://rdfs.org/sioc/ns#";> > > <sioc:Post rdf:about="http://something.com/anyoluri";> > <content:encoded rdf:parseType="Literal"> > <p>some <strong>html</strong> here.</p> > </content:encoded> > </sioc:Post> > </rdf:RDF> > > > Details on Option 1: > When storing xml using (1) the xml get's converted to a double-encoded > string literal. > > Details on Option 2: > haven't tested, sure it'll be the same as (3) though. > > Details on Option 3: > The XML snippets gets stored correctly as an XMLLiteral (great), however: > > -a: can't get any of the bif:(x|xpath_|xquery_)contains methods to work > correctly > > -b: differing errors occur when output in sparql results (as follows): > > > SELECT w/ content type "application/rdf+xml": > error: > <res:valuerdf:parseType="Literal"> > note: > the lack of a space between res:value and rdf:parseType > > > SELECT w/ content type "text/rdf+n3": > error: > res:binding [ res:variable "o" ; res:value "some html here." ] > note: > converted to only text, all xml stripped > > > SELECT w/ content type "text/plain" (turtle) > error: > "some html > here."^^<http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral> > note: > XMLLiteral is correct, but xml is stripped again > > > SELECT w/ application/sparql-results+xml, > application/sparql-results+json, > application/javascript, > text/html > note: > all fine! > > > CONSTRUCT w/ application/rdf+json > error: > "value" : "some html here." , "datatype" : > "http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral"; > note: > all xml stripped but XMLLiteral correct > > > CONSTRUCT w/ text/rdf+n3 > error: > content:encoded some html here. . > note: > everything stripped, no type quotes anything > > > CONSTRUCT w/ application/rdf+xml > note: > perfect! > > > DESCRIBE w/ text/rdf+n3 > error: > content:encoded "some html here." . > note: > quoted, but no type & xml stripped > > DESCRIBE w/ application/rdf+json > error: > "value" : "some html here." , "datatype" : > "http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral"; > note: > all xml stripped but XMLLiteral correct > > > DESCRIBE w/ application/rdf+xml > note: > perfect! > > > As always I'm using the 6.0.1 snapshot; to replicate any of this and > test for yourself the RDF in option 3 is probably ideal to replicate. > And to replicate the initial content:encoded CDATA issue simply drop in > RDF/RSS 1.0 with html in a content:encoded cdata section. > > Many Regards, > > Nathan > > > Ivan Mikhailov wrote: >> Hello Nathan, >> >> We have a support for CDATA output but first of all, the XML serializer >> should know that an element should be encoded with CDATA. For an XSLT >> output, this can be controlled by options of the stylesheet. How was the >> XML in question generated? What's missing in that path? >> >> Best Regards, >> >> Ivan Mikhailov >> OpenLink Software >> http://virtuoso.openlinksw.com >> >> On Mon, 2009-11-30 at 17:25 +0000, Nathan wrote: >>> Hi All, >>> >>> Slight question about content:encoded (I know it's a bit of a special case). >>> >>> assumption(s): >>> 1 - the following is valid content:encoded >>> <![CDATA[<p>some <strong>html</strong> in here.</p>]]> >>> >>> 2 - the following is invalid content:encoded >>> <p>some <strong>html</strong>.</p> >>> >>> question / problem >>> virtuoso consumes [1] perfectly, yet returns http://bit.ly/5zp6aQ >>> (without cdata); so following the above assumptions isn't it invalid >>> considering the element has a type of content:encoded(?) >>> >>> Apologies in advance, it's nothing urgent and I can easily work around >>> it on the client-side. >>> >>> Many Regards, >>> >>> Nathan >>> >>> ------------------------------------------------------------------------------ >>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day >>> trial. Simplify your report design, integration and deployment - and focus >>> on >>> what you do best, core application coding. Discover what's new with >>> Crystal Reports now. http://p.sf.net/sfu/bobj-july >>> _______________________________________________ >>> Virtuoso-users mailing list >>> [email protected] >>> https://lists.sourceforge.net/lists/listinfo/virtuoso-users >> >> > >
