[Geoserver-devel] Guide to run WMS 1.3 cite tests?

Hi,
the cite testing guide has not been updated to include wms 1.3 testing.
http://docs.geoserver.org/latest/en/developer/cite-test-guide/index.html

I see the data directory, but what are the instructions to get the wms 1.3
cite scripts? I guess r2740 might not be the right one for wms 1.3

Cheers
Andrea


Ing. Andrea Aime
Senior Software Engineer

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy

phone: +39 0584962313
fax: +39 0584962313

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf


Indeed. Part of the reason that the docs are not updated is that the cite testing continues to evolve into a mess. With wms 1.3.0 you can’t actually run the tests without hacking them first. I have been meaning to try and get the fixes into cite svn.

I wonder if we should update the devel guide and just make the cite testing procedures simply point one to hudson and run the jobs there? Where there is now a job for wms 1.3.0.

It is a pretty painful process running cite locally as many have found out. And now with all the hacked versions of the tests coupled with the cite groups decision to go ahead and continue to change the tests forgoing backwards compatibility makes the process even more convoluted.

On Wed, Dec 22, 2010 at 12:07 PM, Andrea Aime <andrea.aime@anonymised.com> wrote:

Hi,
the cite testing guide has not been updated to include wms 1.3 testing.
http://docs.geoserver.org/latest/en/developer/cite-test-guide/index.html

I see the data directory, but what are the instructions to get the wms 1.3
cite scripts? I guess r2740 might not be the right one for wms 1.3

Cheers
Andrea


Ing. Andrea Aime
Senior Software Engineer

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy

phone: +39 0584962313
fax: +39 0584962313

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf



Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their database environment, and,
should the need arise, upgrade to a full multi-node Oracle RAC database
without downtime or disruption
http://p.sf.net/sfu/oracle-sfdevnl


Geoserver-devel mailing list
Geoserver-devel@anonymised.comsts.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel


Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.

On Tue, Jan 4, 2011 at 7:05 PM, Justin Deoliveira <jdeolive@anonymised.com> wrote:

Indeed. Part of the reason that the docs are not updated is that the cite
testing continues to evolve into a mess. With wms 1.3.0 you can't actually
run the tests without hacking them first. I have been meaning to try and get
the fixes into cite svn.
I wonder if we should update the devel guide and just make the cite testing
procedures simply point one to hudson and run the jobs there? Where there is
now a job for wms 1.3.0.

Errr... not sure I got you there. Would that be a configurable thing?
Or a CITE engine that you stand up somewhere and that can be used by
the various developers to run the tests?

Cheers
Andrea

-----------------------------------------------------
Ing. Andrea Aime
Senior Software Engineer

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy

phone: +39 0584962313
fax: +39 0584962313

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf

-----------------------------------------------------

On Tue, Jan 4, 2011 at 11:31 AM, Andrea Aime <andrea.aime@anonymised.com> wrote:

On Tue, Jan 4, 2011 at 7:05 PM, Justin Deoliveira <jdeolive@anonymised.com> wrote:

Indeed. Part of the reason that the docs are not updated is that the cite
testing continues to evolve into a mess. With wms 1.3.0 you can’t actually
run the tests without hacking them first. I have been meaning to try and get
the fixes into cite svn.
I wonder if we should update the devel guide and just make the cite testing
procedures simply point one to hudson and run the jobs there? Where there is
now a job for wms 1.3.0.

Errr… not sure I got you there. Would that be a configurable thing?
Or a CITE engine that you stand up somewhere and that can be used by
the various developers to run the tests?

No. I was suggesting that the release step be just run the hudson job as is. Not configurable. Which would run against the latest nightly build. Obviously not very flexible but would work in terms of simply running the tests for a release.

Making the geoserver being tested configurable is doable but requires a bit of scripting work. We could also stand up a dedicated cite testing server somewhere which has all the hacks of the one currently on gridlock. That would work too but again a bit of work and requires a dedicated server.

Cheers
Andrea


Ing. Andrea Aime
Senior Software Engineer

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy

phone: +39 0584962313
fax: +39 0584962313

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf



Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.

On Tue, Jan 4, 2011 at 7:44 PM, Justin Deoliveira <jdeolive@anonymised.com> wrote:

On Tue, Jan 4, 2011 at 11:31 AM, Andrea Aime <andrea.aime@anonymised.com>
wrote:

On Tue, Jan 4, 2011 at 7:05 PM, Justin Deoliveira <jdeolive@anonymised.com>
wrote:
> Indeed. Part of the reason that the docs are not updated is that the
> cite
> testing continues to evolve into a mess. With wms 1.3.0 you can't
> actually
> run the tests without hacking them first. I have been meaning to try and
> get
> the fixes into cite svn.
> I wonder if we should update the devel guide and just make the cite
> testing
> procedures simply point one to hudson and run the jobs there? Where
> there is
> now a job for wms 1.3.0.

Errr... not sure I got you there. Would that be a configurable thing?
Or a CITE engine that you stand up somewhere and that can be used by
the various developers to run the tests?

No. I was suggesting that the release step be just run the hudson job as is.
Not configurable. Which would run against the latest nightly build.
Obviously not very flexible but would work in terms of simply running the
tests for a release.

That works fine normally, having the CITE tests run nightly is a big plus
for me and I guess for anyone developing against GeoServer.

However, picture a situation in which I make a commit that breaks the
WMS 1.3 tests. How am I supposed to figure out how to fix it if I cannot
reproduce it locally? Make blind changes hoping the remote hudson
will bless me with a working build? :slight_smile:

Can the set of hacks to make the WMS 1.3 tests be turned into a patch?

What about the official OGC cite tests, are they still online and usable?
Would GeoServer pass those tests?

Cheers
Andrea

-----------------------------------------------------
Ing. Andrea Aime
Senior Software Engineer

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy

phone: +39 0584962313
fax: +39 0584962313

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf

-----------------------------------------------------

On Tue, Jan 4, 2011 at 12:10 PM, Andrea Aime <andrea.aime@anonymised.com> wrote:

On Tue, Jan 4, 2011 at 7:44 PM, Justin Deoliveira <jdeolive@anonymised.com> wrote:

On Tue, Jan 4, 2011 at 11:31 AM, Andrea Aime <andrea.aime@anonymised.com>
wrote:

On Tue, Jan 4, 2011 at 7:05 PM, Justin Deoliveira <jdeolive@anonymised.com>
wrote:

Indeed. Part of the reason that the docs are not updated is that the
cite
testing continues to evolve into a mess. With wms 1.3.0 you can’t
actually
run the tests without hacking them first. I have been meaning to try and
get
the fixes into cite svn.
I wonder if we should update the devel guide and just make the cite
testing
procedures simply point one to hudson and run the jobs there? Where
there is
now a job for wms 1.3.0.

Errr… not sure I got you there. Would that be a configurable thing?
Or a CITE engine that you stand up somewhere and that can be used by
the various developers to run the tests?

No. I was suggesting that the release step be just run the hudson job as is.
Not configurable. Which would run against the latest nightly build.
Obviously not very flexible but would work in terms of simply running the
tests for a release.

That works fine normally, having the CITE tests run nightly is a big plus
for me and I guess for anyone developing against GeoServer.

However, picture a situation in which I make a commit that breaks the
WMS 1.3 tests. How am I supposed to figure out how to fix it if I cannot
reproduce it locally? Make blind changes hoping the remote hudson
will bless me with a working build? :slight_smile:

I agree that the situation is not perfect but there is only so much time to spend on this. I tried to make the process easier for the developer but the cite group is making that hard. At this point quite frankly all I can do is set up some infrastructure to run the tests and try to reason with the cite group to get our changes in. Hopefully one day we will reach a point in which the fixes are in or the issues have been resolved someone, in which it will (a) be easier to run the tests locally and (b) probably a good online test engine we can use. At this point i am not very optimistic through.

Can the set of hacks to make the WMS 1.3 tests be turned into a patch?

Some can, some can’t since I brought up the issues and they were rejected saying that we will fix them later and just release now. The cite process is also very opaque. Closed bug tracker, etc… Although there seems to be some progress there.

What about the official OGC cite tests, are they still online and usable?

Would GeoServer pass those tests?

There is a version online but its only the latest, which we don’t pass for wfs 1.1. For wms 1.3 i am not so sure. The bugs i found in the trunk version are pretty obvious, the test suites don’t even compile. Which leads me to believe that noone is currently using them.

Cheers
Andrea


Ing. Andrea Aime
Senior Software Engineer

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy

phone: +39 0584962313
fax: +39 0584962313

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf



Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.