On 28/02/12 17:08, Robert Vesse wrote:
Hi Rafal
Just a word of warning, timeouts for remote queries don't work quite the same
as timeouts for local queries.
For remote queries you have two timeouts - connect and read. Connect is the
maximum time allowed to make the connection to the remote server, read is the
maximum time allowed for data to start being received after the connection has
been made. If neither timeout is hit then ARQ will try to read the entire
result - not necessarily at once depending on the type of query and the result
format e.g. most SELECT results are read in a streaming fashion - so if you
need timeouts at that stage that is up to you.
As Andy says you can get and use the snapshots via Maven
Rob
good point - and there is more than one way to have HTTP query timeouts.
You can configure the query engine itself to timeout queries - fixed
timeout though.
Andy
On Feb 28, 2012, at 5:09 AM, Andy Seaborne wrote:
On 28/02/12 12:02, Ozga, Rafal wrote:
Hi,
It seems that timeouts in QueryEngineHTTP were only implemented in 2.9.1
version of ARQ
( https://issues.apache.org/jira/browse/JENA/fixforversion/12319291,
https://issues.apache.org/jira/browse/JENA-56,
https://issues.apache.org/jira/browse/JENA-210), and actually quite recently
13 February.
From what I can see that version hasn¹t been released officially yet the
last available one in the maven repository is 2.9.0-incubating:
https://repository.apache.org/content/repositories/releases/org/apache/jena/
jena-arq/
So the obvious question: what is the release date for the version
2.9.1-incubating ?
Now, provided that this is a critical issue for us is it better for us to
wait for that release or we should rather clone the current repository
snapshot and build the desired version ourselves ?
Kind regards,
Rafal
There isn't a confirmed date.
You don't have to built it yourself - there are snapshots builds done every
night:
https://repository.apache.org/content/groups/snapshots/org/apache/jena/jena-arq/2.9.1-incubating-SNAPSHOT/
You can check build status at:
https://builds.apache.org/view/G-L/view/Jena/
Generally, we only have working code in the trunk. Some build failures are due
to the Apache Jenkins setup, which is large and complex to cope with all the
projects using it. The build slaves do sometimes get out of sync with the
master and sometimes there are interal comms (which, I guess, are due to system
load).
Andy
