Re: [qvtd-dev] Recent commits

[Ed Willink <ed@xxxxxxxxxxxxx>]

Hi

I find the merges very irritating aesthetically. Visual clutter, 
redundant commit lines.

But more seriously history is broken since as with a multiple 
inheritance, there is no longer a primary ancestor. Instead there is a 
fix-up commit that is hard to view.

So where rebase can be done at modest cost, it's worth it. Where there 
is genuine long running concurrent development, rebase is probably very 
difficult and probably misleading too.

In this case it took about ten minutes to discover that 10 premature 
files needed deleting to allow the rebase to just work.

    Regards

        Ed

On 01/05/2015 11:15, Adolfo Sanchez-Barbudo Herrera wrote:
> Hi Ed,
>
> The result looks ok to me.
>
> Some comments in-lined below.
>
> Regards,
> Adolfo.
> --
> On 01/05/2015 10:26, Ed Willink wrote:
> > Hi
> >
> > I successfully rebased asanchez/mtc then rebased Horacios's and some of
> > my recent work.
>
> I'm curious about:
> - the procedure of the rebase, providing there were already merges and 
> that I got issues with moved files in my first rebase.
> - apart from the "straight" line in history for anyone uncomfortable 
> with not seeing that straightness, what's the "advantage" of spending 
> time on a rabase rather than advancing the master to the already 
> merged asanchez/mtc2 ?
>
> > There was one conflict on OCL2QVTiBroker, whereby Horacio's code was
> > switching from Flat to Nested scheduling.
> >
> > Flat scheduling has no test failures. Nested scheduling has one, so I
> > pushed the no-failures to master.
> >
> > At least QVTs-pToQVTi.etl appears to have multiple copies separate
> > updates. I removed the second copy of rule MiddlePropertyAssignments.
> > Please review lost changes and see whether there is some opportunity for
> > sharing a common *.etl and only refining the necessary changes.
>
> I presume one of them is obsolete, but that's something that Horacio 
> should better respond.
> > I think that there is little that has not now been promoted to master so
> > please delete dead branches. Please ping for anything I've overlooked.
> >
> >      Regards
> >
> >          Ed
> > _______________________________________________
> > qvtd-dev mailing list
> > qvtd-dev@xxxxxxxxxxx
> > To change your delivery options, retrieve your password, or unsubscribe
> > from this list, visit
> > https://dev.eclipse.org/mailman/listinfo/qvtd-dev
> _______________________________________________
> qvtd-dev mailing list
> qvtd-dev@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or 
> unsubscribe from this list, visit
> https://dev.eclipse.org/mailman/listinfo/qvtd-dev
>
>
> -----
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 2015.0.5941 / Virus Database: 4339/9671 - Release Date: 05/01/15