yade-dev team mailing list archive

[Robert Caulk <rcaulk@xxxxxxxxxxxx>]

Hello Anton,

Thank you, Anton. I agree that yade should be compilable and operational
with any combination of packages. I managed to fix the issue with my recent
commits, based on my testing it should now pass the buildbot.

But my question remains, the LINSOLV package is not compiled or tested by
the buildbot? In other words, I could make a detrimental change to LINSOLV,
but the buildbot would report success?


Best,

Robert

On Tue, May 2, 2017 at 1:38 PM, Anton Gladky <gladky.anton@xxxxxxxxx> wrote:

> Hi Robert,
>
> the Yade should work in any combination of enabled features and should
> not fail to compile or to crash. I did not look into your changes, but I
> think
> you need to add a couple of #ifdef guards to make it compilable.
>
> Best regards
>
> Anton
>
>
> 2017-05-02 21:45 GMT+02:00 Robert Caulk <rcaulk@xxxxxxxxxxxx>:
> > Ok so after looking into this more I noticed that the buildbot is not
> using
> > the default cmake build. In this case, LINSOLV is disabled on the
> buildbot
> > machine, which is what led to the occurrence of the error on the buildbot
> > but not my desktop.
> >
> > Shouldn't we have the buildbot also testing out compilation of features
> that
> > are considered "default" in addition to a stripped down build? In this
> case,
> > the entire LINSOLV package is not compiled and tested by the buildbot
> after
> > every commit.
> >
> > Best,
> >
> > Robert
> >
> > On Tue, May 2, 2017 at 10:55 AM, Robert Caulk <rcaulk@xxxxxxxxxxxx>
> wrote:
> >>
> >> Hey guys,
> >>
> >> So I have an issue I am hoping someone might be able to help me with.
> >> These changes compile perfectly on my desktop [1]. But the buildbot
> catches
> >> an issue that my desktop did not catch [2]. It appears to be an issue
> with
> >> the use of templates inheritance in the flow engine regime, which I
> have now
> >> fixed.  But again, it compiles perfectly fine on my desktop either way.
> So
> >> the only way for me to find out if this is going to pass the buildbot
> is to
> >> push more changes to the trunk? And let's be honest, the likelihood of
> the
> >> buildbot getting upset about something else is high and the last thing
> we
> >> need are pushes to the trunk in the form of compilation
> trial-and-errors.
> >>
> >> So what should I do? Has anyone else encountered this issue where the
> >> trunk compiles fine on their own computer but the buildbot catches a
> >> problem?
> >>
> >> Cheers,
> >>
> >> Robert
> >>
> >>
> >>
> >> [1]https://github.com/yade/trunk/commit/f6970362d9e6e866e8ad
> bc8cbea18e54f677f785
> >>
> >> [2]https://yade-dem.org/buildbot/builders/yade-full/builds/
> 3979/steps/compile/logs/stdio
> >>
> >> On Tue, May 2, 2017 at 9:58 AM, Robert Caulk <rcaulk@xxxxxxxxxxxx>
> wrote:
> >>>
> >>> Very sorry, taking care of this now.
> >>>
> >>> On Tue, May 2, 2017 at 10:51 AM, <buildbot@xxxxxxxxxxxx> wrote:
> >>>>
> >>>> The Buildbot has detected a failed build on builder yade-full while
> >>>> building yade.
> >>>> Full details are available at:
> >>>>  https://yade-dem.org/buildbot/builders/yade-full/builds/3979
> >>>>
> >>>> Buildbot URL: https://yade-dem.org/buildbot/
> >>>>
> >>>> Buildslave for this Build: r0calcul9
> >>>>
> >>>> Build Reason: scheduler
> >>>> Build Source Stamp: [branch master]
> >>>> f6970362d9e6e866e8adbc8cbea18e54f677f785
> >>>> Blamelist: robcaulk <rob.caulk@xxxxxxxxx>
> >>>>
> >>>> BUILD FAILED: failed compile
> >>>>
> >>>> sincerely,
> >>>>  -The Buildbot
> >>>>
> >>>>
> >>>>
> >>>> _______________________________________________
> >>>> Mailing list: https://launchpad.net/~yade-dev
> >>>> Post to     : yade-dev@xxxxxxxxxxxxxxxxxxx
> >>>> Unsubscribe : https://launchpad.net/~yade-dev
> >>>> More help   : https://help.launchpad.net/ListHelp
> >>>
> >>>
> >>
> >
> >
> > _______________________________________________
> > Mailing list: https://launchpad.net/~yade-dev
> > Post to     : yade-dev@xxxxxxxxxxxxxxxxxxx
> > Unsubscribe : https://launchpad.net/~yade-dev
> > More help   : https://help.launchpad.net/ListHelp
> >
>