Re: [PATCH 2/4] merge-recursive: increase marker length with depth of recursion

[Elijah Newren <newren@gmail.com>]

On Sun, Oct 14, 2018 at 10:12 PM Junio C Hamano <gitster@pobox.com> wrote:
>
> Elijah Newren <newren@gmail.com> writes:
>
> > When using merge.conflictstyle=diff3 to have the "base version" be shown
> > in conflicts, there is the possibility that the base version itself has
> > conflicts in it.  This comes about when there are more than one merge
> > base, and the merging of those merge bases produces a conflict.
> > Since this process applies recursively, it is possible to have conflict
> > markers nested at an arbitrary depth; to be able to differentiate the
> > conflict markers from different nestings, we make them all of different
> > lengths.
>
> I know it is possible that the common ancestor part that is enclosed
> by the outermost makers can have arbitrary conflicts, and they can
> be even recursive conflicts.  But I fail to see why it is a problem.
> Perhaps that is because I am not particularly good at resolving
> merge conflicts, but as long as the common part of the outermost
> merge is identifyable, would that really matter?  What I would do
> while looking at common ancestor part with conflicts (not even a
> recursive one) is just to ignore it, so...
>
> Not that I strongly oppose to incrementing the marker length at
> every level.  I do not think it would hurt, but I just do not see
> how it would help.

Fair enough.  The real motivation for these changes was the
modification to rename/rename(2to1) conflicts (and rename/add
conflicts) to behave like add/add conflicts -- that change means we
can have nested conflict markers even without invoking the recursive
part of the recursive machinery.  So, I needed a way to increase the
length of the conflict markers besides just checking
opts->virtual_ancestor.  Just using a fixed extra indent seemed
problematic, because if I also had to worry about even one
virtual_ancestor, then I was already dealing with the possibility of
triply nested conflict markers and only one of them from a virtual
merge base.  See t6036 in
https://public-inbox.org/git/20181014020537.17991-3-newren@gmail.com/.

However, that series was long enough, so to try to simplify review I
split as much as I could out of it.  That resulted, among other
things, in me submitting this marker nesting change as part of this
series using a more limited rationale.

Would you like me to edit the commit message to include this more
difficult case?