More good news! email@example.com pointed out that XUL templates that build from datasources referenced via http:// url stopped working at some point -- the templates are built empty. These loads are being done asynchronously (as confirmed by checking the nsRDFXMLDataSource:5 log), as opposed to the blocking loads used by rdfcat (bug 89446) that have been broken for some time. I confirmed this on linux; it's probably on other platforms as well. This is sort of funny, actually -- I've tried on several occasions to rewrite rdfcat using non-blocking Refresh() but haven't been successful. My most recent attempt (a month or more ago) worked fine for file urls but failed for http urls. I never understood why; it was probably because of this bug. Presumably the magic goo in Refresh + associated code is no longer holding things together.
It looks like this was fixed recently
really? fascinating. the tree i tried it with was a bit stale; i'll try it with a fresher one. perhaps all is right in the world after all.
Ok, the problem I was seeing was just that the web server was serving the rdf as text/plain instead of text/xml (*sigh*). So is there actually a bug here? Or should I just WFM it?
Apparently I hallucinated this whole thing while in some sort of frenzy. Not only does this work for me, I've also finally succeeded in making the async rewrite of rdfcat work, which is further proof that there's no problem. WFM.
Status: NEW → RESOLVED
Last Resolved: 17 years ago
Resolution: --- → WORKSFORME
tever is not RDF QA anymore
QA Contact: tever → nobody
You need to log in before you can comment on or make changes to this bug.