Hi,

> Well your timing is very fortunate, ss of the end of last week the
> query results can be serialized as both JSON and SPARQL XML and the
> query/result APIs have changed (hopefully this will the final method
> signature) see:
>
> http://copia.ogbuji.net/blog/2006-06-23/rdf_query_apis_for_rdflib
>

Thanks! It works great :) I had some minor problems though. One thing
I got a bit caught up on is that the results are in unicode and have
to be translated into str() before passed on the wire. Also, bnode's
are not formatted correctly:

in rdflib/sparql/QueryResult.py (row 8):
-        return '<bnode>%s</bnode' % val
+        return '<bnode>%s</bnode>' % val

Querying on subgraphs does not work as I expected. A query is
dispatched to the conjunctive graph (which is stated in the docs) and
not only over the graph on which the query was executed. My plan was
to query only a subgraph or a context to avoid the overhead of the
entire store.

> With regards to graph provenance tracking and garbage collection you
> might be interested in Michel Pelletier's development branch (not yet
> merged) which includes a framework for registering listeners to graph
> modfication:
>
> http://rdflib.net/pipermail/dev/2006-May/000010.html

Ok, thanks, I will look closer into that, it definitely sounds interesting.

Cheers,
Mikael
_______________________________________________
Dev mailing list
[email protected]
http://rdflib.net/mailman/listinfo/dev

Reply via email to