[Discussion] cherry-picking a merge

Earlier "git cherry-pick" learned the "-m <parent-number>" option to allow cherry-picking a merge commit.

When you have this history:

  ---o---o---C---A---M
      \             /
       o---o-------B

You can replay the change between A and M (in other words, the effect of merging B into A) on top of C to create a new commit, with:

        $ git cherry-pick -m 1 M

In the current implementation, the resulting commit has a single parent C. This is quite similar to a squash merge of B into C.

When you think about it, as long as the topological relationship between A and B is very similar to that of C and B (iow, "merge-base A B" and "merge-base C B" are the same), the effect should be the same as a real merge between B and C, shouldn't it?

  ---o---o---C---A---M
      \       \     /
       o---o---\---B
                \   \
                 `---X

I am wondering if it makes sense to record the result of "cherry-pick -m" as a real merge between the current HEAD and all the other parents of the cherry-picked merge except the one that is named with the <parent-number>.

Junio C Hamano <gitster@pobox.com> wrote:

Yes.

Then `rebase -i` might be able to learn how to "pick" merge commits. And then my coworkers can stop bothering me about that. And just do it themselves.

I used to think of merges as something special. I now really only look at them as being special *during* the merge process, as you may need to generate that recursive base. But otherwise its just "diff M^1..M". So why isn't it?

:-)

"Shawn O. Pearce" <spearce@spearce.org> writes:

Actually, this turns to be a trick question ;-)

If there was an accidental clean merge between A and B, and a part of the duplicated same changes in the accidental clean merge sits between C and A, replaying the difference from A to M on top of C would not yield the same result as merging B into C.