Hi,
I don't know about you, but build times are starting to
worry me. The last successfull build of GeoServer 1.6.x
on my PC took 5:30 minutes! And this is after fixing the
double catalog loading issue reported in http://jira.codehaus.org/browse/GEOS-1546
and working to remove all log out of tests (I have both
working now, finally I can see a list of tests and
eventual failures instead of a long list of layers loading).
This is bad, because time is getting up very fast as we
add unit tests, especially wfs ones (the wfs module alone
takes 2:30 to build on my pc). Something must be done
to locate and kill whatever source or slowness there
is around.
I think a significant one may be the XML schema cache
being purged at each unit test start... what do you think?
Is there anything we can do about it?
Also, for read only tests we could avoid re-creating
the data dir over and over as Gabriel suggested...
Cheers
Andrea
I agree. And its on my todo list to fix the unit testing setup /
teardown as per Gabriels suggestion. If someone wants to go ahead and do
that I am fine with it, I just have bigger fish to fry at the moment.
As for the schema cache... that is a hard one to fix on 1.6.x. It should
be fixed on trunk now because schemas are actual singletons now.
-Justin
Andrea Aime wrote:
Hi,
I don't know about you, but build times are starting to
worry me. The last successfull build of GeoServer 1.6.x
on my PC took 5:30 minutes! And this is after fixing the
double catalog loading issue reported in
http://jira.codehaus.org/browse/GEOS-1546
and working to remove all log out of tests (I have both
working now, finally I can see a list of tests and
eventual failures instead of a long list of layers loading).
This is bad, because time is getting up very fast as we
add unit tests, especially wfs ones (the wfs module alone
takes 2:30 to build on my pc). Something must be done
to locate and kill whatever source or slowness there
is around.
I think a significant one may be the XML schema cache
being purged at each unit test start... what do you think?
Is there anything we can do about it?
Also, for read only tests we could avoid re-creating
the data dir over and over as Gabriel suggested...
Cheers
Andrea
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel
!DSPAM:4007,4746b7b4296061849620573!
--
Justin Deoliveira
The Open Planning Project
http://topp.openplans.org