From: Harald Nordgren <haraldnordgren@gmail.com>
Date: Mon, 12 Oct 2026 14:31:22 GMT
Subject: Re: [PATCH v2 0/2] branch: let --delete-merged find squash merged branches
Message-ID: <CAHwyqnUP8Bgd4S=oE9g0TonqL-ohy4LXXG66AY8SeVk=6HCNUg@mail.gmail.com>
In-Reply-To: <03b187ae-bef2-426f-b6a4-def9f70c4a3f@gmail.com>

> I think it would be helpful if the cover letter explained in more detail
> what this series actually does to detect squashed merges and compared
> that to other approaches such as using patch ids or test merges.

Ok.

> As far
> as I can see the approach here is, to a first approximation, checking
>
>         $(git merge-tree B C) == C^{tree}
>
> to see if C is a squashed merge of B, only without using merge-ort but
> by calling merge_ll() on each path instead. That means we're not
> checking if C is the same as merging B into C^, C may contain additional
> changes as well. I think it also means we're missing out on rename
> detection.

The check is "does C already contain everything B changed", not "is C
exactly the squash of B". For deleting the branch it could be enough
since nothing is lost.

But it means "landed as" can point to a commit with other changes in
it as well, maybe that's not good.

I think following renames is a bad basis for assuming merged.
Sometimes the goal of a topic is a rename, and then it should not be
dropped.


Harald