Looking for a HOWTO [SEC=UNCLASSIFIED]

[Paul Murray <pm...@anbg.gov.au>]

Hi.

We have a whole swag of data, in RDF, which I would like to expose to the internet with a sparql server.

I'm having a bit of trouble untangling the various components - ARQ, Joseki, and all the rest, whether jena is a library, a standalone, product, part of a sprql server, a separate service that a sparql server needs access to, and so on.

Is there a howto document? All I want to do is 
* spark up a triple-store
* load some xml/rdf data into it.
* make it available read-only to the net
* on Solaris boxes

What bits of software do I download? Arq? Joseki? Jena? There are two backend stores: one backed by a RDBMS, the other sitting on a filesystem. I think the filesystem one is the way to go.

BTW: does jena store quadruples? Can I erase all triples "belonging" to some uri with a single command?



If you have received this transmission in error please notify us immediately by return e-mail and delete all copies. If this e-mail or any attachments have been sent to you in error, that error does not constitute waiver of any confidentiality, privilege or copyright in respect of information in the e-mail or attachments.

Please consider the environment before printing this email.

Re: Looking for a HOWTO [SEC=UNCLASSIFIED]

[Damian Steer <d....@bristol.ac.uk>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 02/09/11 11:48, Paolo Castagna wrote:
> 
> 
> Andy Seaborne wrote:
>>

>> It's easier to preload the database
>>
>> tdbloader --loc=DB ....data...
> 
> This requires users to get TDB (which is targeted at developers rather than
> end users), add TDB bin directory to their PATH:

True.

> If Fuseki's aim is to be as easy as possible to use, perhaps, we need to make
> easier for people to do an initial bulk loading. However, copying tdbloader
> and tdbloader2 scripts into Fuseki does not seem a great idea. Or, would that
> be ok?

When helping people I've used:

$ java -jar fuseki-sys.jar tdb.tdbloader ...

Not as fast as tdbloader2, but no download required. Having one mighty
jar is very handy.

Damian
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk5gt6oACgkQAyLCB+mTtykUqQCggXyn6f62UiHygsze9ku09L7S
S2kAoKU8dg+5lkRF4tH8+WzuZPbCmRHv
=XlFW
-----END PGP SIGNATURE-----

Re: Looking for a HOWTO [SEC=UNCLASSIFIED]

[Paolo Castagna <ca...@googlemail.com>]

Andy Seaborne wrote:
> 
> 
> On 02/09/11 04:45, Paul Murray wrote:
>> Hi.
>>
>> We have a whole swag of data, in RDF, which I would like to expose to
>> the internet with a sparql server.
>>
>> I'm having a bit of trouble untangling the various components - ARQ,
>> Joseki, and all the rest, whether jena is a library, a standalone,
>> product, part of a sprql server, a separate service that a sparql
>> server needs access to, and so on.
> 
> Just add one :-)
> 
> Fuseki
> http://openjena.org/wiki/Fuseki
> 
> It's the SPARQL server and contains all need to publish a SPARQL end
> point (Jena code library, query engine ARQ and database TDB).
> 
> * spark up a triple-store
> * load some xml/rdf data into it.
> * make it available read-only to the net
> * on Solaris boxes
> 
> It's easier to preload the database
> 
> tdbloader --loc=DB ....data...

This requires users to get TDB (which is targeted at developers rather than
end users), add TDB bin directory to their PATH:

cd /opt
wget http://www.openjena.org/repo/com/hp/hpl/jena/tdb/0.8.10/tdb-0.8.10.zip
unzip tdb-0.8.10.zip

or

cd /opt
sudo svn co https://svn.apache.org/repos/asf/incubator/jena/Jena2/TDB/trunk/ tdb
cd /opt/tdb
mvn package

Add the TDB bin directory to your PATH and verify typing:
tdbquery --version

If Fuseki's aim is to be as easy as possible to use, perhaps, we need to make
easier for people to do an initial bulk loading. However, copying tdbloader
and tdbloader2 scripts into Fuseki does not seem a great idea. Or, would that
be ok?

Paolo

> fuseki-server --loc=DB /Name
> 
> and the endpoint is http://?????/Name/sparql
> 
> It's pure Java6.
> 
>> What bits of software do I download? Arq? Joseki? Jena? There are two
>> backend stores: one backed by a RDBMS, the other sitting on a
>> filesystem. I think the filesystem one is the way to go.
>>
>> BTW: does jena store quadruples?
> 
> Yes.
> 
>> Can I erase all triples "belonging"
>> to some uri with a single command?
> 
> Yes - either via the server which has web forms for SPARQL Update
> commands or using the SPARQL Graph Store Protocol (scripes in the
> includes SOH package).
>     
>     Andy

Re: Looking for a HOWTO [SEC=UNCLASSIFIED]

[Andy Seaborne <an...@epimorphics.com>]

On 02/09/11 04:45, Paul Murray wrote:
> Hi.
>
> We have a whole swag of data, in RDF, which I would like to expose to
> the internet with a sparql server.
>
> I'm having a bit of trouble untangling the various components - ARQ,
> Joseki, and all the rest, whether jena is a library, a standalone,
> product, part of a sprql server, a separate service that a sparql
> server needs access to, and so on.

Just add one :-)

Fuseki
http://openjena.org/wiki/Fuseki

It's the SPARQL server and contains all need to publish a SPARQL end
point (Jena code library, query engine ARQ and database TDB).

* spark up a triple-store
* load some xml/rdf data into it.
* make it available read-only to the net
* on Solaris boxes

It's easier to preload the database

tdbloader --loc=DB ....data...
fuseki-server --loc=DB /Name

and the endpoint is http://?????/Name/sparql

It's pure Java6.

> What bits of software do I download? Arq? Joseki? Jena? There are two
> backend stores: one backed by a RDBMS, the other sitting on a
> filesystem. I think the filesystem one is the way to go.
>
> BTW: does jena store quadruples?

Yes.

 > Can I erase all triples "belonging"
> to some uri with a single command?

Yes - either via the server which has web forms for SPARQL Update 
commands or using the SPARQL Graph Store Protocol (scripes in the 
includes SOH package).
	
	Andy