Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [tycho-dev] Replacing BeanShell by Groovy

I think we are currently violating the Eclipse IP policy by using the maven-invoker-plugin and not having a build/test CQ for it. AFAIK, we have this problem in the current state of Tycho, and it is not getting worse by migrating to Groovy.

Therefore, I'll first look into the contribution, and then work towards an automated report of our IP-relevant dependencies.

Regards
Tobias


> -----Original Message-----
> From: tycho-dev-bounces@xxxxxxxxxxx [mailto:tycho-dev-bounces@xxxxxxxxxxx]
> On Behalf Of Igor Fedorenko
> Sent: Freitag, 30. November 2012 04:05
> To: tycho-dev@xxxxxxxxxxx
> Subject: Re: [tycho-dev] Replacing BeanShell by Groovy
> 
> Not sure what comments you expect. I already told several times that I
> will neither veto to endorse move to groovy, but will help review and
> merge the fix otherwise.
> 
> I also have no plans to file any CQs regarding tycho test code (this was
> never a major concern for me in this case), but other tycho developers
> may choose to do so.
> 
> --
> Regards,
> Igor
> 
> On 12-11-29 9:33 AM, Max Rydahl Andersen wrote:
> >
> > Any comments on the three problems/challenges ?
> >
> > On 22 Nov 2012, at 08:59, Max Rydahl Andersen <max.andersen@xxxxxxxxxx>
> wrote:
> >
> >> Hey guys,
> >>
> >> Lets just take a step back and split these issues up and solve them
> individually.
> >>
> >> I see three problems here:
> >>
> >> 1) source feature names are blank when published on updatesites - makes
> install and review of features really ugly for us and users
> >>
> >> 2) There is a missing CQ for the maven-invoker-plugin for the current
> Tycho usage of Beanshell (independent of Groovy usage)
> >>
> >> 3) There is a request on moving to groovy shell instead of Bean Shell
> which would not require any (known) code changes and would make testing
> easier in the future.
> >>
> >> Is that a basic summary of the issue?
> >>
> >> #1 is getting urgent we'll work on rewriting the test to use beanshell
> once we get through our own product current release. If anyone else is up
> for rewriting the test in meantime that would be appreciated. ETA for us
> on this is ~in December
> >>
> >> #2 is this true or not ? If true, is there any committer who want to
> help to fix this and what can we do to help ? If no - then Groovy should
> be fine to use as well as Beanshell?
> >>
> >> #3 Any objections against moving to groovy if #2 is fixed ?
> >>
> >> Thank you,
> >> Max
> >>
> >> On 20 Nov 2012, at 15:41, Mickael Istria <mistria@xxxxxxxxxx> wrote:
> >>
> >>> On 11/20/2012 01:59 PM, Igor Fedorenko wrote:
> >>>> Instead of trying to force us do the extra work needed to switch to
> >>>> groovy
> >>> Have you estimated how much "extra work" it would be. I'd say 5
> minutes to open CQ. Once accepted, a few second to merge both
> contributions (move from BeanShell to Groovy, fix source feature title).
> Not sure such a little time needs such a debate.
> >>>
> >>>> why not rewrite the test in beanshell as I originally suggested?
> >>> Groovy is a way better technology for the use-cases that are in Tycho
> than BeanShell. Parsing XML in Groovy is trivial. There are a some TODOs
> in current BeanShell scripts that are to be filled with XML Parsing and
> verification. With Groovy, those TODOs would never have existed, and there
> would always have been XML parsing.
> >>>
> >>>> The patch would have been accepted long time ago if you did not
> insist
> >>>> on groovy. /me shrugs
> >>> I agreed it would be better to keep only one language for validation.
> There is a Gerrit contribution pending to move all tests to Groovy, so
> it's easier to write better quality validation after maven-invoker-plugin.
> It's just renaming files. So I still don't understand why you are not
> enthusiast in using Groovy.
> >>>
> >>> Using BeanShell or Groovy requires a CQ for maven-invoker-plugin,
> which was not requested yet. BeanShell and Groovy would get both accepted
> as part of the same CQ as dependencies of maven-invoker-plugin. So in any
> case, you HAVE TO do this extra work to align with IP cleanness for Tycho.
> I described how to request this CQ previously.
> >>> So I also don't understand why does it seem so difficult to open a CQ.
> >>> --
> >>> Mickael Istria
> >>> Eclipse developer at JBoss, by Red Hat
> >>> My blog - My Tweets
> >>> _______________________________________________
> >>> tycho-dev mailing list
> >>> tycho-dev@xxxxxxxxxxx
> >>> https://dev.eclipse.org/mailman/listinfo/tycho-dev
> >>
> >> _______________________________________________
> >> tycho-dev mailing list
> >> tycho-dev@xxxxxxxxxxx
> >> https://dev.eclipse.org/mailman/listinfo/tycho-dev
> >
> > _______________________________________________
> > tycho-dev mailing list
> > tycho-dev@xxxxxxxxxxx
> > https://dev.eclipse.org/mailman/listinfo/tycho-dev
> >
> _______________________________________________
> tycho-dev mailing list
> tycho-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/tycho-dev


Back to the top