On 21/06/11 16:21, Dalkandura Arachchige Gunaratna wrote:
Thnx Andy. I have observed that sometimes when I try for several times with the
same timeout manually using the DBpedia sparql endpoint, it works. A better
strategy would be to increase the timeout from the client side for some
iterations before stop querying. But if there is not way to set the timeout for
the client side then anyways this would not work.
The people running dbpedia may disagree with you!
I think I can't set server timeout (DBpedia sparql endpoint) when I am using a
java program to query the endpoint. Is there a way that I can set the DBpedia
sparql endpoint timeout using a program? Thank you.
Kalpa Gunaratna
There's some kind of protocol request parameter (non-standard) to set
the timeout of the request.
You can set it with QueryEngineHTTP.addParam.
Andy
----- Original Message -----
From: Andy Seaborne<[email protected]>
Date: Tuesday, June 21, 2011 3:38 am
Subject: Re: Jena ARQ API - how to set timeout in Java
To: [email protected]
On 20/06/11 17:32, Dalkandura Arachchige Gunaratna wrote:
Hi, I am using Jena ARQ API to query DBpedia dataset using a sparql
endpoint. I want to set the timeout limit for querying using a java
program so that the program will not crash if sufficient timeout
is set.
My intention is to have a simple loop to handle so that if an
exceptionhappens, the query will be re-done.
I have come across a code sample to set timeout in the
internet and
I
am not sure whether this works correctly. Is this correct or
there is
any other way of doing this?
QueryEngineHTTP qexec = (QueryEngineHTTP)
QueryExecutionFactory.sparqlService(> strSparqlEndpoint,
qExecuteQuery);
qexec.addParam("timeout","10000");>
Kalpa Gunaratna
QueryEngineHTTP itself does not support timeouts:
https://issues.apache.org/jira/browse/JENA-56
This would be timeout at the local end anyway - in the
dbpedia.org (if
that is what you are querying) then the timeout can be set so
the server
times the query out. See the documentation for dbpedia and
Virtuoso.
In either case, retrying a query is unlikely to be a good
strategy - if
the query timeouted once, why won't it timeout the next time as well?
Andy
