git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [RFC] Git rerere and non-conflicting changes during conflict resolution

From
Junio C Hamano <gitster@pobox.com>
Date
Jul 26, 2017, 08:06 UTC
Message-ID
<xmqq60eftyfb.fsf@gitster.mtv.corp.google.com>
In-Reply-To
<20170725205843.bi6kyqjlzyodmxuq@sigill.intra.peff.net>
Jeff King <peff@peff.net> writes:
Show 5 quoted lines
> Hrm. That doesn't quite work, though. Because if your <A,B> are the
> merge, then merging a topic to next will get an "A" that is a merge
> commit from next. But that commit will never end up in master. What's
> causing the conflict is really some "A" that is in the history between
> the merge base and "A" (but we don't know which).

There may be a misunderstanding. When I said the key <A,B> is a pair of branch names, I didn't mean 'A' to be the name of an integration branch (e.g. 'pu') and 'B' to be the name of a topic. Rather, both 'A' and 'B' are the names of topic branches.

IOW, instead of having refs/merge-fix/sd/branch-copy that says "I know when I merge sd/branch-copy to pu or jch, there is a semantic conflict with some unnamed topic that is likely to be already in there", i.e. keying with only a single topic name, the ideal I presented would say 'sd/branch-copy and mh/packed-ref-store topics are both by themselves OK, but when merged together, the end result of textual merge needs to be further fixed up by cherry-picking this change', by keying a change with a pair of topic names, sd/branch-copy (which introduces a new method in the ref backend vtable) and mh/packed-ref-store (which adds a new ref backend). The latter does not know the need for the new method, and the former does not know the need to implement its new method in a new backend, so a merge needs a trivial implementation of the new method added to the new backend, which is what refs/merge-fix/sd/branch-copy does.

And better yet, instead of A=sd/branch-copy B=mh/packed-ref-store, we could point at the exact commit on each of these branches that introduce the semantic conflict. I would probably pick these two

  A=52d59cc6 ("branch: add a --copy (-c) option to go with --move (-m)", 2017-06-18)
  B=67be7c5a ("packed-backend: new module for handling packed references", 2017-06-23)

so when we are on commit X that has A but not B, and are trying to merge branch Y that has B but not A, we want the merge-fix to kick in. Walking "rev-list --left-right X...Y" and noticing A and B in the output would be a way to notice it.

[footnote]
*1* https://github.com/gitster/git/ should mirror these refs in the
    refs/merge-fix/ hierarchymentioned in the body of this article.
Previous: Junio C Hamano
Message 10 of 10 in “[RFC] Git rerere and non-conflicting changes during conflict resolution”
  1. Raman GuptaJul 25, 2017
  2. Jeff KingJul 25, 2017
  3. Raman GuptaJul 25, 2017
  4. Jeff KingJul 25, 2017
  5. Junio C HamanoJul 25, 2017
  6. Junio C HamanoJul 25, 2017
  7. Jeff KingJul 25, 2017
  8. Junio C HamanoJul 26, 2017
  9. Junio C HamanoJul 27, 2017
  10. Junio C HamanoJul 26, 2017

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.