[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#685605: porting the RDF generation to tracker.debian.org



Hi Oliver,

On Tue, Apr 07, 2015 at 09:08:53AM +0200, Olivier Berger wrote:
> > My goal is essentially to port the existing functionality from the
> > packages.qa.debian.org to tracker.debian.org, but in a way that it could be
> > extended later.

Ok. The vision has expanded here. I've got a service set up at
http://rdf.debian.net (not necessarily stable currently, still in
development) and the code running it is in collab-maint/semantic-debian.git.

There are two parts to this. The first is the triples generator that uses
UDD as a backend. This produces a Turtle dump. The second part is a Python
Flask application that allows for URIs to be dereferenced and provides
content negotiation.

You may be interested to know that the current dataset contains 456676
triples generated from UDD data. I plan to add more. The linking of
upstreams to CPAN looked interesting, although the URL given appears broken
currently.

> The differences aren't big, but some choices are probably better. But
> for a start, porting in an "exact" way may be the safer bet, until some
> consumer complains about the lack of certain details.

I will read up on the changes later, for now I'm just getting the framework
in place. It's probably not so important we follow the standards as much as
we follow what others have done to allow for easier linking.

> I think the order of priority could be Turtle, the JSON-LD and finally
> RDF/XML if need be, nowadays. But we lack statistics on the popularity.

Ok. Currently I've got Turtle, then RDF/XML, then HTML+RDFa. It stops at the
first one it finds in the Accept header. JSON-LD support currently is not
included in rdflib and I don't want to add third-party libraries into the
mix until I've got the basics working.

> That sounds like a perfect way to go : I'm quite busy at the moment
> (recent parent of twins ;), so I can't dedicate much effort on that for
> the coming months, but I'm glad to help.

I haven't got stuck yet, so no effort required on your part so far. (:

I'll keep this bug updated as progress happens.

Thanks,
Iain.

-- 
e: irl@fsfe.org            w: iain.learmonth.me
x: irl@jabber.fsfe.org     t: EPVPN 2105
c: 2M0STB                  g: IO87we
p: 1F72 607C 5FF2 CCD5 3F01 600D 56FF 9EA4 E984 6C49

Attachment: pgp_otPJL6Fle.pgp
Description: PGP signature


Reply to: