bug#25105: 26.0.50; diff navigation is broken

[Tino Calancha]

Mark Oteiza <mvoteiza@udel.edu> writes:

> I am guessing this is a consequence of bug#17544. From -Q:
>
> 1. Do C-x v d RET = in a repository with a bunch of worktree changes
> 2. Hit n. Point is now at the top of the SECOND hunk
> 3. Go to end of buffer.
> 4. Hit p. Point is now at the top of the PENULTIMATE hunk
FWIW, to me this behaviour is very annoying and it has being
around already a while without a fix (4 months).
I would propose to revert the commit causing this misbehaviour.
Then, once the patch is mature enough, it can be pushed again without
affecting users.

Dima Kogan <dima@secretsauce.net> writes:

> I pushed the fix to this. What do yall want to do about the other logic?
If for 'the other logic' you mean Mark's recipe above, then what i want
is:
above recipe behaves exactly as it does in Emacs-25; that behaviour is
very convenient to read diffs.

Tino