Re: git-rebase produces incorrect output

From: Elijah Newren <newren@gmail.com>
Date: Nov 30, 2019, 16:37 UTC
Message-ID: <CABPp-BHPkwnbLR0xNaY3SphCpmR6QRA1j953c2hU7ZFpL38Nig@mail.gmail.com>
In-Reply-To: <CABPp-BGiu2nVMQY_t-rnFR5GQUz_ipyEE8oDocKeO+h+t4Mn4A@mail.gmail.com>

On Fri, Nov 29, 2019 at 8:22 PM Elijah Newren <newren@gmail.com> wrote:

> On Fri, Nov 29, 2019 at 12:24 AM Pavel Roskin <plroskin@gmail.com> wrote:

Show 20 quoted lines

> So, to summarize here:
>   * you have a case where the default 3 lines of context mess stuff
> up; but rebase --merge works great
>   * am doesn't have a -U option, and ignores the diff.context setting,
> making it impossible to force the am backend to work on your case
>   * rebase doesn't have an option to use the merge/interactive backend
> by default (nor an --am option to override it)
>
> Also:
>   * The performance of the merge/interactive backend is slightly
> better than the am-backend
> (https://public-inbox.org/git/CABPp-BF=ev03WgODk6TMQmuNoatg2kiEe5DR__gJ0OTVqHSnfQ@mail.gmail.com/)
> and will continue to get better
>   * The merge/interactive backend supports many more options than the
> am-backend, though the am one still has a few the merge backend
> doesn't.  Once the ra/rebase-i-more-options topic merges, --whitespace
> will be the only consequential option that the am-backend supports
> that the merge/interactive-backend doesn't.  (There's also -C, but as
> noted above, the merge/interactive backend already have access to the
> full file).

In case it wasn't clear above, the merge/interactive backend could just accept the -C option and ignore it and do nothing, since it already has access to the full file (thus why I consider the -C option to not be consequential).

Also, I remembered and dug out a few more items about the default
rebase backend that might be worth including in this summary:
  * The am backend operates with incomplete tree information as well,
limiting what the merge/resolve/whatever can do and what information
can be provided to the user (see
https://public-inbox.org/git/xmqqh8jeh1id.fsf@gitster-ct.c.googlers.com/)
  * The interactive backend, although slightly faster than the
am-backend (on p3400 at least), is slightly slower with split-index
which hasn't yet been investigated (see
https://public-inbox.org/git/nycvar.QRO.7.76.6.1901312310280.41@tvgsbejvaqbjf.bet/)

> Maybe we should just switch the default, for everyone?  (And provide
> an --am option to override it and a config setting to get the old
> default?)

CC'ing a few folks for opinions on switching the default backend of rebase from --am to --merge. Johannes already agreed it was the right path eventually[1], and Junio suggested the am backend should be deprecated[2] and eventually replaced, so I was going to push on this after some merge performance work but perhaps it's a good time to query if it's time to switch the default sooner.

[1] See the end of https://public-inbox.org/git/nycvar.QRO.7.76.6.1808311158540.71@tvgsbejvaqbjf.bet/, also linked above. [2] https://public-inbox.org/git/xmqqh8jeh1id.fsf@gitster-ct.c.googlers.com/, also linked above.