SPARQL engine unsupported (ActiveRdfError)

On 28.08.2007, at 18:14, Laurent Ostiz wrote:

> Hi !
>
> Just installed ActiveRDF with support for SPARQL and
Redland.  
> Example file from the wiki works well with redland, but
fails with  
> SPARQL :
>
>
/usr/local/lib/ruby/gems/1.8/gems/activerdf_sparql-1.3.2/lib
/ 
> activerdf_sparql/sparql.rb:28:in `initialize': SPARQL
engine  
> unsupported (ActiveRdfError)
>         from
/usr/local/lib/ruby/gems/1.8/gems/activerdf-1.6.3/lib/ 
> active_rdf/federation/connection_pool.rb:123:in `new'
>         from
/usr/local/lib/ruby/gems/1.8/gems/activerdf-1.6.3/lib/ 
> active_rdf/federation/connection_pool.rb:123:in `send'
>         from
/usr/local/lib/ruby/gems/1.8/gems/activerdf-1.6.3/lib/ 
> active_rdf/federation/connection_pool.rb:123:in
`create_adapter'
>         from
/usr/local/lib/ruby/gems/1.8/gems/activerdf-1.6.3/lib/ 
> active_rdf/federation/connection_pool.rb:72:in
`add_data_source'
>         from activerdf_example_sparql.rb:4



It seems like something was changed in the sparql adapter
which is  
not yet reflected in the tutorial examples.

can you try replacing the line:
ConnectionPool.add_data_source :type => :sparql, :url
=> "http:// 
m3pe.org:8080/repositories/test-people", :results =>
:sparql_xml

with the line:
ConnectionPool.add_data_source :type => :sparql, :url
=> "http:// 
m3pe.org:8080/repositories/test-people", :results  
=> :sparql_xml, :engine => :joseki

That should usually work.

Now the other problem is that the test sparql server seems
to be down.

If you have any other type of sparql server available you
try with that.

Available sparql endpoint engine types are:
:yars2, :sesame2, :joseki, :virtuoso

I will update the getting started guide, and write Eyal a
mail,  
reminding him of restarting the sparql endpoint. 
_______________________________________________
ActiveRDF mailing list
ActiveRDFlists.deri.org
http
://lists.deri.org/mailman/listinfo/activerdf

I just got into ActiveRDF and encountered the same problem.
After 
looking through the source I came across that list of
supported engines 
and added :engine => :sesame2 to the end of my data
source. Now, 
everything works fine as far as I can tell but I'm curious
how you're 
suppose to know which engine to use. Are we suppose to guess
which 
software the SPARQL endpoint is running on?

Shawn

Benjamin Heitmann wrote:
>
> On 28.08.2007, at 18:14, Laurent Ostiz wrote:
>
>> Hi !
>>
>> Just installed ActiveRDF with support for SPARQL
and Redland. Example 
>> file from the wiki works well with redland, but
fails with SPARQL :
>>
>>
/usr/local/lib/ruby/gems/1.8/gems/activerdf_sparql-1.3.2/lib
/activerdf_sparql/sparql.rb:28:in 
>> `initialize': SPARQL engine unsupported
(ActiveRdfError)
>>         from 
>>
/usr/local/lib/ruby/gems/1.8/gems/activerdf-1.6.3/lib/active
_rdf/federation/connection_pool.rb:123:in 
>> `new'
>>         from 
>>
/usr/local/lib/ruby/gems/1.8/gems/activerdf-1.6.3/lib/active
_rdf/federation/connection_pool.rb:123:in 
>> `send'
>>         from 
>>
/usr/local/lib/ruby/gems/1.8/gems/activerdf-1.6.3/lib/active
_rdf/federation/connection_pool.rb:123:in 
>> `create_adapter'
>>         from 
>>
/usr/local/lib/ruby/gems/1.8/gems/activerdf-1.6.3/lib/active
_rdf/federation/connection_pool.rb:72:in 
>> `add_data_source'
>>         from activerdf_example_sparql.rb:4
>
>
>
> It seems like something was changed in the sparql
adapter which is not 
> yet reflected in the tutorial examples.
>
> can you try replacing the line:
> ConnectionPool.add_data_source :type => :sparql,
:url => 
> "h
ttp://m3pe.org:8080/repositories/test-people",
:results => :sparql_xml
>
> with the line:
> ConnectionPool.add_data_source :type => :sparql,
:url => 
> "h
ttp://m3pe.org:8080/repositories/test-people",
:results => 
> :sparql_xml, :engine => :joseki
>
> That should usually work.
>
> Now the other problem is that the test sparql server
seems to be down.
>
> If you have any other type of sparql server available
you try with that.
>
> Available sparql endpoint engine types are:
> :yars2, :sesame2, :joseki, :virtuoso
>
> I will update the getting started guide, and write Eyal
a mail, 
> reminding him of restarting the sparql 
>
endpoint._______________________________________________
> ActiveRDF mailing list
> ActiveRDFlists.deri.org
> http
://lists.deri.org/mailman/listinfo/activerdf
>
_______________________________________________
ActiveRDF mailing list
ActiveRDFlists.deri.org
http
://lists.deri.org/mailman/listinfo/activerdf

On 08/28/07/08/07 15:42 -0400, Shawn Simister wrote:
> I just got into ActiveRDF and encountered the same
problem. After looking 
> through the source I came across that list of supported
engines and added 
> :engine => :sesame2 to the end of my data source.
Now, everything works 
> fine as far as I can tell but I'm curious how you're
suppose to know which 
> engine to use. Are we suppose to guess which software
the SPARQL endpoint 
> is running on?
>
Shawn, you're making a very good point here: it is something
that the 
SPARQL spec indeed does not cover. If all endpoints would
behave the same 
there wouldn't be a problem, but unfortunately they don't.
They all 
implement bits and pieces of the spec, and so we have to
tune our queries 
to them. That's why we need to know what engine is running,
and 
unfortunately, there's no automatic way to figure that out
(no http headers 
or something like that).  I hope that if we just wait a
couple of 
months/years, the endpoints will all behave in a more
standard manner.

With respect to testing on m3pe.org endpoint: you can better
use the 
DBpedia.org endpoint, since it (a) is managed more
professionally than the 
one on m3pe.org and (b) it actually contains interesting
data. See some 
earlier posts in the archive for querying dbpedia (you have
to use the 
'virtuoso' engine).

  -eyal
_______________________________________________
ActiveRDF mailing list
ActiveRDFlists.deri.org
http
://lists.deri.org/mailman/listinfo/activerdf