ld join
us. One wouldn't need to be well-versed in Perl as long as one would be able
to make a substantial contribution to writing this module.
Best,
Kjetil
--
Kjetil Kjernsmo
Ph.d Research Fellow, Semantic Web
kje...@kjernsmo.net
http://www.kjetil.kjernsmo.net/
s, but I'm not sufficiently familiar with the code
to fix it.
Med vennlig hilsen
Kjetil Kjernsmo
--
Senior Kunnskapsingeniør / Redaktør for SPARQL F&R
Mobil: 986 48 234
Epost: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| COMPUTAS SETTER KUNNSKAP I SYSTEM |
Com
lly quite urgent...
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer / SPARQL F&R Editor
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
rator?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer / SPARQL F&R Editor
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
On Wednesday 27 May 2009 13:58:46 Frederick Giasson wrote:
> Hi Kjetil,
>
>
> Try one of these PL functions:
>
> DB.DBA.TTLP_MT()
> DB.DBA.RDF_LOAD_RDFXML_MT()
Yeah, I am aware of them, but how can we use them from the SPARQL endpoint?
Kind regards
Kjetil Kjernsmo
--
Seni
to insert data into an arbitrary graph using HTTP?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer / SPARQL F&R Editor
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysak
s-1-8-t3"."O")
AND "s-1-8-t3"."O" = __i2idn (
__bft( 'http://www.computas.com/mediasone-ontologi#Aktoer' , 1))
AND "s-1-6-t4"."G" = __i2idn (
__bft( 'http://msone.computas.no/graphs' , 1))
AND
All,
Just a tiny follow-up. It seems like the trace_on setting disappears after a
server restart. Can I set it persistently somewhere?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com
class", $122 "callret")
10: $46 "ctor-1" := := artm $123 "callret"
14: BReturn 0
Subquery Select(TOP ) ($46 "ctor-1")
}
After code:
0: if ($133 "user_aggr_notfirst" 1(=) ) then 15 else 4
unkn 4
4: $133 "user_a
of Ubuntu, this is the hard deadline. To be
confident, it should be uploaded well in advanced before that date.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE
On Tuesday 21 April 2009 11:21:32 Kjetil Kjernsmo wrote:
> øl and Øl yields different results and given the "i" operator, we would
> expect it to give the same result.
Actually, this turns out to have become worse in 5.0.11. Now
prefix dct: <http://purl.org/dc/terms/>
7 "in_iter"
over ($29 "callret", $30 "callret", $31 "callret", $32 "callret")
from DB.DBA.RDF_QUAD by RDF_QUAD_OGPS 0.87 rows
Key RDF_QUAD_OGPS ASC ($210 "s-1-28-t9.S")
, ,
row specs: >
Current of: <$212 "" sp
FILTER regex(?title, "øl", "i") } ORDER BY
When this is run http://msoneutv.computas.no:8890/sparql which has a patched
5.0.10, it seems to have some case sensitivity problems with the UTF-8
character, in the final filter clause.
øl and Øl yields different results and giv
r_env")
7: $47 "callret" := Call DB.DBA.SPARQL_DESC_DICT ($46 "callret",
$28 "callret", , , ,
$26 "callret")
14: $48 "callret-0" := Call DB.DBA.RDF_FORMAT_TRIPLE_DICT_AS_TTL
($47 "callret")
21: BReturn 0
Select (TOP
that these queries are so simple, the corresponding SQL isn't the
interesting part here?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
t0"."S" = __i2idn (
__bft(
'http://rabbit.computas.int:8180/sublima-webapp-1.0-SNAPSHOT/topic/topic000747'
,
1))
AND isiri_id ( "s-1-1-t0"."O")
AND ( "s-1-1-t1"."S" < min_bnode_iri_id ())
AND "s-1-1-t1".
elation' )),
vector (
1,
2,
3,
DB.DBA.RDF_MAKE_IID_OF_QNAME (
UNAME'http://www.w3.org/2000/01/rdf-schema#label' ),
1,
3),
vector (
1,
ink policy documents, that I don't know in
detail would be pretty clear about this? :-)
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
n UTF-8, whereas the latter in ISO-8859-1. This
works, but seems a fragile hack. It seems to do the job for now, but if those
two things would happen to be in the same query, different parts of the
string would need to be encoded differently, it seems, which would be
problematic.
Is there a
s not return any results, whereas a FILTER clause with
FILTER regex(str(?label), "^Ø", "i")
does. Indeed, there are only uppercase Ø's in beginning of words in this
index, but the "i" modifier should fix that for us, shouldn't it?
For ASCII letters, this works
making?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
oint, and it seems the performance
penalty isn't disastrous, so I guess I can live with it for now.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS
-graph-uri " + defaultGraph
+ " \n" +
req.getParameterValues("query")[0]);
but that seems to cause Jena to croak and say the query is invalid.
Then, the question is of course whether using JDBC gives me much? I could just
use the endpoint and then parse and reserialise the result, but that s
://musicbrainz.org/mm/mm-2.1#Track> ,
>
>Oh, cool! But I can't remember seeing that in the SPARQL spec, is that a
>Virtuoso extension?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.compu
ness happens...
> In addition, you may wish to abbreviate the query as follows:
> FILTER (?type in (
> <http://musicbrainz.org/mm/mm-2.1#Track> ,
Oh, cool! But I can't remember seeing that in the SPARQL spec, is that a
Virtuoso extension?
Kind regards
t; ) .
?resource dct:title ?title .
} UNION {
?resource a ?type .
FILTER (?type = <http://www.computas.com/mediasone#Text> ) .
?resource dct:title ?title .
}
FILTER regex(?title, "turi", "i")
}
ORDER BY ?title LIMIT 4 OFFSET 0
The query
Just to follow up on this issue:
On Thursday 29 January 2009 17:09:10 Kjetil Kjernsmo wrote:
> So, in spite of the data being loaded many times into the graph named
> http://msone.computas.no/graphs it doesn't show up. And suddenly, it does,
> with no changes to what I do, as far
ch, so it doesn't sound technically unfeasible from a naive viewpoint.
Yup, it doesn't sound too unfeasible.
Another thing we've done a bit is to construct URIs from literals where a base
URI and e.g. a dct:title is concatenated.
I think we may raise this with the DAWG II.
/rdf-sparql-query/#FunctionMapping
it appears that you cannot cast from string to URI, only from URI to string.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS P
ILL, usually I have to.
>Note you can also use the following param in the [SPARQL] section to
>control result set sizes:
>
> ResultSetMaxRows = 10
Yes, that's in the default setup, so I have that allready.
Kind regards
Kjetil Kjernsmo
--
Senior Knowl
?conductoruri foaf:name ?conductorname .
}
OPTIONAL {
<http://www.wizardis.com.au/2005/12/MAVIS#TitleWorkkey41402>
sub:extinfo ?dbpediatext .
}
}
It is submitted by an HTTP POST operation.
We have now rewritte
dct:title we have a dumb:title, but this
would lead to data duplication and is certainly less elegant than modifying
the indexer to do the job.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
we are looking for a better solution to this
problem than we have ourselves. Additionally, our own solution requires Jena,
so we would prefer a Virtuoso-only solution.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: h
x and an error if it didn't conform?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
computas.com/mediasone-maven/trunk/modelling/src/resources/mediasone-default-graph.n3
So, in spite of the data being loaded many times into the graph named
http://msone.computas.no/graphs it doesn't show up. And suddenly, it does,
with no changes to what I do, as far as I can tell.
Weird, h
t;oslo central station" (the former
being a common abbreviation of the latter). It appears from the BNF that this
is not possible, correct?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
|
SYS_SPARQL_HOST.
It appears not to return the query the whole merged graph, but I don't
understand which subset it returns. It appears to return data from a few
graphs.
Perhaps I'm just blind, but I hope someone will help me see :-)
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge En
the logs, but
that's not what I want, I need those that ran against the endpoint.
How do I do that?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
All,
I've been looking for an option to turn on logging of the queries that are
sent to the server (at the endpoint would also do). This would be an
extremely useful aid in debugging our application (which generates SPARQL
queries).
Is there such an option in Virtuoso?
Kind regards
K
hat's why, but it
shouldn't be relevant to the problem.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
hem I'd like. Also, the CONSTRUCT is arguably more
complex (for the programmer) than the DESCRIBE query.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
the same) and the you run the script on
the directory where the files are. So, I'm not just testing with a
browser :-)
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
t the actual data from a DESCRIBE,
I have to explicitly name the graph. Which creates problems on the client
side... I have to admit that I am a bit confused about graph naming.
Does this problem have a more trivial solution?
The endpoint where all this runs is a public test endpoint at
http:/
rygghetslarm_005975
So, from my perspective, it seems rather wrong that the resource
http://www.sjukvardsradgivningen.se/artikel.asp?CategoryID=22952
is returned when it does not have a dct:identifier property with the object
http://rabbit.computas.int:8180/sublima-webapp-1.0-SNAPSHOT
be found in (a work in progress)
https://submarine.computas.com/mediasone-maven/trunk/backend/
I'm really sorry I couldn't find something that's easier to debug, but I
wanted to share the status with you nevertheless. I'll continue to work on
this problem.
Kind regards
Kjeti
e dct:title ?title .
?title bif:contains "'oslo s'" .
}
LIMIT 10
It would have to be rewritten a bit, but it appears that the critical thing is
the literal that has a space but no asterisk.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Emai
On Monday 17 November 2008 17:19:57 Kjetil Kjernsmo wrote:
> All,
>
> I have another problem with my SPARQL Endpoint servlet, which is just
> supposed to generate RDF/XML-ABBREV from a query.
Thanks to Hugh, I now run a 5.0.10 prerelease, which fixes a related bug, but
it didn'
aber", "i")
}
LIMIT 10
Our development endpoint running 5.0.9 is here:
http://msone.computas.no:8890/sparql
This doesn't return any solutions, but
FILTER regex(str(?title), "^ Aber", "i")
does. So, it appears that something is wrong with the cas
exits like mysql. So,
my feature request is that isql should just stop the current query, not exit
when hitting ctrl+c.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR
On Thursday 20 November 2008 12:47:45 Kjetil Kjernsmo wrote:
> The problem has a pretty clear symptom: When running against an endpoint
> based on a Jena Model, I get the following date:
>
> rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime";>2003-07-11
> 09
n the endpoint code. Of course, it could be in some Jena
code, but we've never seen this when using a Postgres backend, so it seems it
is most likely in the driver.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
res.setContentType("application/rdf+xml");
> PrintWriter out = res.getWriter();
> model.write(out, "RDF/XML-ABBREV");
> out.close();
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
BC(TM) Version 3.x [Build 3.25]
Oh, mine is 3.19...
I just redownloaded it from
http://virtuoso.openlinksw.com/dataspace/dav/wiki/Main/VirtJenaProvider#Required%20Files
and that too is 3.19... Where are the new ones?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +4
;RDF/XML-ABBREV");
out.close();
The exception message doesn't mean much to me, can anybody give more details
about what this means?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
ds to running a
forward chaining reasoner on the top of Virtuoso, both in terms of
performance and in terms of ease of implementation.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE
mall Debian repository for Ubuntu
8.04 (which is a Long Term Support release), 8.10 (which is current) and
Debian Lenny. It is just a static web server
http://www.debian.org/doc/manuals/repository-howto/repository-howto.en.html
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47
Tim,
Great stuff!
But I just ran cvs up and it didn't seem to give me a debian/ directory.
Didn't it make it in?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR
a running
number to the %U.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
bit hard to read at times. In
particular, I wouldn't think that most SPARQL users would need to understand
the details of how the quads are stored in the database, which is now the
first thing you encounter. Perhaps these belong in different chapters or
documents?
Kind regards
Kjetil
d
it be possible to reuse this code on the endpoint?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
looks like both are optional.
Likewise
dct:language : ;
which is quite nonsensical, but it passes the other parsers, and
:rdf:type owl:Ontology ;
Well, you get the idea? Perhaps this is something that entered the spec with
the Team Submission?
Kind regards
Kjetil Kj
On Wednesday 08 October 2008 21:10:11 Tim Haynes wrote:
> Kjetil Kjernsmo wrote:
> > On Wednesday 08 October 2008 17:15:50 Tim Haynes wrote:
> >> I'm wondering how you've gone about it; I saw the mention of
> >> http://debianpaket.de/svn-buildpackage/index.ht
; around).
This should perhaps be done when first configuring virtuoso in the install
process?
> It might also help to specify ./configure --with-layout=debian so
> it uses /var/lib/virtuoso/ as the database directory.
That's a nice feature! It can be added to debian/rules but pe
On Wednesday 08 October 2008 16:01:38 Sergio Fernández wrote:
> On Tue, 2008-10-07 at 15:07 +0200, Kjetil Kjernsmo wrote:
> > I've been packaging Virtuoso for Debian (...)
>
> Great work Kjetil, I'd the same idea some weeks ago: I contacted with
> some colleagues that
until packaging is done for real.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 | Fax:+47 6783
1001
xtensive free
QA. Also, in my experience, sysadmins are generally your friend if you ask
them to install something that's allready in their distro, but can be a real
bottleneck if not. From my own perspective as a developer, it eases setup of
development systems a lot.
Kind regards
Kjetil Kje
in the standards process.
It seems to me that a solution to this problem would be to allow matching
graphs that are not part of the result set, e.g.
DESCRIBE ?foo FROM WHERE GRAPH { ?foo dct:title "Foo" . } or
something like that...
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge
e literals seem to cause us trouble. We are
working on several fronts with this problem, one solution involves Virtuoso,
some does not.
We are using Jena as a SPARQL Endpoint, but we are also using SPARQL Update.
With the current architecture, it should be straightforward to switch back
and forth,
solved by simply using Virtuoso, is that the cost of serializing and
transferring the result set is too high. In that case, I was wondering if I
could somehow exclude the sub:literals predicate from the DESCRIBE result?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Direct: +
s linked to the subject
I'm interested in here, but also to literals linked to subjects that is an
object of this resource... Therefore, it would be nice to have the
possibility of defining a pattern for this index.
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Direct: +47 678
70 matches
Mail list logo