en/rebase-backend (was Re: "rebase -ri" (was Re: Problems with ra/rebase-i-more-options - should we revert it?))

en/rebase-backend (was Re: "rebase -ri" (was Re: Problems with ra/rebase-i-more-options - should we revert it?))

[Elijah Newren]

Hi Junio,

On Mon, Jan 13, 2020 at 2:07 PM Junio C Hamano <gitster@pobox.com> wrote:
>
> Junio C Hamano <gitster@pobox.com> writes:
>
> > Junio C Hamano <gitster@pobox.com> writes:
> >
> >> I will push out what I wish to be able to tag as the final [*1*]
> >> shortly but without actually tagging, so that it can get a bit wider
> >> exposure than just the usual "Gitster tested locally and then did
> >> let Travis try them" testing.
> >
> > I haven't heard from any failure report so (taking no news as good
> > news) I'll cut the final today based on what is already on the
> > public repositories everywhere.
>
> By the way, as one of the methods to double check that my result of
> reverting the merge made sense, I ran "git rebase -ri v2.24.0 pu"
> and excised the merge and the problematic topic out of the todo
> list.  With the rerere database populated beforehand, it was more or
> less a painless exercise (except for one topic, en/rebase-backend,
> which is one of the topics that was queued forking 'master' after
> the topic got merged *and* actually depended on what the topic did)
> and after about 1700+ steps (which did not take more than 20
> minutes, including the time spent for the manual rebasing of
> en/rebase-backend topic) I got the same tree for 'pu' I pushed out
> last night.

I wonder if I should have been the one fixing up the en/rebase-backend topic...

Also, with the new release and the review comments Phillip posted on
the en/rebase-backend series, would you rather see me address those as
additional patches on top of en/rebase-backend, or should we kick that
topic out of next and have me send a full re-roll?  I'm not sure
what'd be better and I don't mind going either direction...

Re: en/rebase-backend (was Re: "rebase -ri" (was Re: Problems with ra/rebase-i-more-options - should we revert it?))

[Emily Shaffer]

On Tue, Jan 14, 2020 at 08:52:57AM -0800, Elijah Newren wrote:
> Hi Junio,
> 
> On Mon, Jan 13, 2020 at 2:07 PM Junio C Hamano <gitster@pobox.com> wrote:
> >
> > Junio C Hamano <gitster@pobox.com> writes:
> >
> > > Junio C Hamano <gitster@pobox.com> writes:
> > >
> > >> I will push out what I wish to be able to tag as the final [*1*]
> > >> shortly but without actually tagging, so that it can get a bit wider
> > >> exposure than just the usual "Gitster tested locally and then did
> > >> let Travis try them" testing.
> > >
> > > I haven't heard from any failure report so (taking no news as good
> > > news) I'll cut the final today based on what is already on the
> > > public repositories everywhere.
> >
> > By the way, as one of the methods to double check that my result of
> > reverting the merge made sense, I ran "git rebase -ri v2.24.0 pu"
> > and excised the merge and the problematic topic out of the todo
> > list.  With the rerere database populated beforehand, it was more or
> > less a painless exercise (except for one topic, en/rebase-backend,
> > which is one of the topics that was queued forking 'master' after
> > the topic got merged *and* actually depended on what the topic did)
> > and after about 1700+ steps (which did not take more than 20
> > minutes, including the time spent for the manual rebasing of
> > en/rebase-backend topic) I got the same tree for 'pu' I pushed out
> > last night.
> 
> I wonder if I should have been the one fixing up the en/rebase-backend topic...
> 
> Also, with the new release and the review comments Phillip posted on
> the en/rebase-backend series, would you rather see me address those as
> additional patches on top of en/rebase-backend, or should we kick that
> topic out of next and have me send a full re-roll?  I'm not sure
> what'd be better and I don't mind going either direction...


For what it's worth, I had started to look into one of the bugs Jonathan
Nieder reported[1] as a patch on top of en/rebase-backend topic. I'll
keep an eye on this thread; if you're interested in rerolling the whole
topic then it might be less time for you to identify the right place to
call "post-commit" than it would be for me, plus less conflict
resolution time.

[1]: https://lore.kernel.org/git/20200110231436.GA24315@google.com/
(bullet point 1)

 - Emily

Re: en/rebase-backend (was Re: "rebase -ri" (was Re: Problems with ra/rebase-i-more-options - should we revert it?))

[Elijah Newren]

Hi Emily,

On Tue, Jan 14, 2020 at 12:59 PM Emily Shaffer <emilyshaffer@google.com> wrote:
>
> On Tue, Jan 14, 2020 at 08:52:57AM -0800, Elijah Newren wrote:
> > Hi Junio,
> >
> > On Mon, Jan 13, 2020 at 2:07 PM Junio C Hamano <gitster@pobox.com> wrote:
> > >
> > > Junio C Hamano <gitster@pobox.com> writes:
> > >
> > > > Junio C Hamano <gitster@pobox.com> writes:
> > > >
> > > >> I will push out what I wish to be able to tag as the final [*1*]
> > > >> shortly but without actually tagging, so that it can get a bit wider
> > > >> exposure than just the usual "Gitster tested locally and then did
> > > >> let Travis try them" testing.
> > > >
> > > > I haven't heard from any failure report so (taking no news as good
> > > > news) I'll cut the final today based on what is already on the
> > > > public repositories everywhere.
> > >
> > > By the way, as one of the methods to double check that my result of
> > > reverting the merge made sense, I ran "git rebase -ri v2.24.0 pu"
> > > and excised the merge and the problematic topic out of the todo
> > > list.  With the rerere database populated beforehand, it was more or
> > > less a painless exercise (except for one topic, en/rebase-backend,
> > > which is one of the topics that was queued forking 'master' after
> > > the topic got merged *and* actually depended on what the topic did)
> > > and after about 1700+ steps (which did not take more than 20
> > > minutes, including the time spent for the manual rebasing of
> > > en/rebase-backend topic) I got the same tree for 'pu' I pushed out
> > > last night.
> >
> > I wonder if I should have been the one fixing up the en/rebase-backend topic...
> >
> > Also, with the new release and the review comments Phillip posted on
> > the en/rebase-backend series, would you rather see me address those as
> > additional patches on top of en/rebase-backend, or should we kick that
> > topic out of next and have me send a full re-roll?  I'm not sure
> > what'd be better and I don't mind going either direction...
>
>
> For what it's worth, I had started to look into one of the bugs Jonathan
> Nieder reported[1] as a patch on top of en/rebase-backend topic. I'll
> keep an eye on this thread; if you're interested in rerolling the whole
> topic then it might be less time for you to identify the right place to
> call "post-commit" than it would be for me, plus less conflict
> resolution time.
>
> [1]: https://lore.kernel.org/git/20200110231436.GA24315@google.com/
> (bullet point 1)

Ooh, that'd be great if you could fix the post-commit piece,
especially since you're more familiar with the hooks in general and
probably have access to the actual usecases where people are using
post-commit hooks.

I am planning on rerolling the topic, but I'd be happy to insert a
patch authored by you into the series.  Or maybe it'd be even better
to just merge them as totally independent series -- I strongly doubt
there would be any conflicts.  I didn't touch the am codepaths in
en/rebase-backend (any am codepaths I touched were submitted
separately and already made it into v2.25.0), and since the resolution
of the post-commit hook sounded like it should be modifying 'am' to
behave like the merge/interactive backends (as per Junio's
comments[2]), the fix should be orthogonal to the rest of
en/rebase-backend.

[2] https://lore.kernel.org/git/xmqqa76sl67u.fsf@gitster-ct.c.googlers.com/