[openbiblio-dev] installation problems

[rich hardy]

Thanks again for the update!

I'm following the instructions in INSTALL.txt so I'm using Virtuoso.

So I put back rdflib3.2 and I included the following lines in
controllers/graph.py (because it already had some rdflib code there):
import rdflib

rdflib.plugin.register('sparql', rdflib.query.Processor,
                       'rdfextras.sparql.processor', 'Processor')
rdflib.plugin.register('sparql', rdflib.query.Result,
                       'rdfextras.sparql.query', 'SPARQLQueryResult')

But I now get the following error when I try to access
http://localhost:5000/entry/GB5006595:
RuntimeError: maximum recursion depth exceeded

Trace:
File '~/openbiblio/openbiblio/controllers/graph.py', line 121 in html_view
  c.graph = self._get_or_infer_graph(uri)
File '~/openbiblio/openbiblio/controllers/graph.py', line 49 in
_get_or_infer_graph
  [graph.add(x) for x in cg.triples((None, None, None))]
File 'build/bdist.linux-i686/egg/rdflib/query.py', line 139 in __getattr__
File 'build/bdist.linux-i686/egg/rdflib/query.py', line 139 in __getattr__
File 'build/bdist.linux-i686/egg/rdflib/query.py', line 139 in __getattr__
**cut**
File 'build/bdist.linux-i686/egg/rdflib/query.py', line 139 in __getattr__
RuntimeError: maximum recursion depth exceeded

I'm not sure what to do here, and I can't actually find the openbiblio
'handler' package?

Thank you again!

On Thu, Jan 19, 2012 at 6:03 PM, William Waites <ww at styx.org> wrote:
> On Thu, 19 Jan 2012 16:12:40 -0300, rich hardy <ontouser at gmail.com> said:
>
>    > Hi, *formatting edit* I'm still having a couple of issues.  Does
>    > openbiblio use rdflib 3.2 or 2.4?
>
> As I recall, this went back and forth a little bit. ORDF is able to
> use either, and this is because we needed to keep support for
> inferencing and things like that that wanted to use FuXi and an old
> branch of 2.4.1 called layercake. But openbiblio is indeed intended to
> use the 3.X rdflib, but it hasn't been tested with 3.2 (or possibly
> even 3.1) but only 3.0 I believe.
>
> However, I think we only really used this with a foreign back-end like
> virtuoso or 4store. So when handler.query gets called, which is in
> ORDF, it passes that along to the remote SPARQL endpoint, which means
> it doesn't touch rdflib's SPARQL.
>
> So if you want to use rdflib's own SPARQL with a native storage you'll
> need to do the magic incantation to register the plugins from extras,
> as outlined here: https://code.google.com/p/rdflib/wiki/IntroSparql
> somewhere convenient in the call stack. Then it *should* work, but
> it's not well tested.
>
> Cheers,
> -w