Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [mdt-ocl.dev] Building branches on Hudson

Doesn't exactly sound as if you're in favor of having a specific hudson job for a bug branch. Anyway, I'd find it great if you could provide a template job for us to clone in case we'd like to have one. This template job would ideally have a config that, as you previously suggested, would not build javadoc, would not sign and would not upload any artifacts to anywhere --- just plain build and test execution.

Thanks and best,
-- Axel

On 7/15/2011 4:02 PM, Adolfo Sánchez-Barbudo Herrera wrote:
Axel,

As far as I had understood the idea was doing integration builds to
verify correct execution of Junits prior to merge into the master ... a
punctual task which I think we may coordinate. During the development of
a bug resolution or various (of them) the usual task should be executing
them in local, it should take less time, it shouldn't consume Eclipse
server's resources ( I don't usually see this practice in Eclipse hudson
server, although if it is available... I don't see inconveniences to use
it) ... Similarly, as far as I'm able to, I try to tests releng stuff in
local, even though knowing that the configuration in Eclipse servers
(enviroment variables, servers sanity, etc) may change and vary the
expected results.

Anyway, as far you are allowed feel free to clone jobs, as I commented
you would still need to configure the job to use the proper branch/tag ...

Cheers,
Adolfo.

El 15/07/2011 13:26, Axel Uhl escribió:
I don't follow... copying from a template would still require to
configure the new created job to use the desired branch... so why do we
need this job creation ?....

Because one or more people may work on more than one bug branch at one
time, all of which may desire continuous build/test support.

Having a simple (unique) job designed to execute tests so that it were
configurable in somehow to use the desired branch/tag (via job parameter
or configuring the job it self) should suffice...

Unless more or us compete for the same job. Theoretically, you could
have one job per committer. But then, if one committer works on
different bugs simultaneously, the committer would have to change the
job configuration each time a context switch happens. I'd still prefer
the job per branch approach.

Best,
-- Axel
_______________________________________________
mdt-ocl.dev mailing list
mdt-ocl.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt-ocl.dev


--
Open Canarias, S.L.
	*Adolfo Sánchez-Barbudo Herrera*
adolfosbh(at)opencanarias(dot)com
<mailto:adolfosbh%28at%29opencanarias%28dot%29com>
C/Elías Ramos González, 4, ofc. 304
38001 SANTA CRUZ DE TENERIFE
Tel.: +34 922 240231



_______________________________________________
mdt-ocl.dev mailing list
mdt-ocl.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt-ocl.dev



Back to the top