bug#81395: 32.0.50; diff-apply-hunk should be case-sensitive

Stefan Monnier <[email protected]> writes:

>>> Indeed the matching is case-insensitive.  I remember making this
>>> decision consciously.  🙂
>>> I'm OK if you make it case-sensitive, tho.
>>
>> See the attached patch.
>
> LGTM, feel free to push it.

Applied, onto master.
https://cgit.git.savannah.gnu.org/cgit/emacs.git/commit/?id=e0d2d80a628

> I think we should push this to the UI instead.  E.g. in the case you
> showed the question should be "already applied, undo instead?" but
> something like "Patch applicable but could also be undone elsewhere.
> Really apply?"  So the users can say `n` and then `C-u C-c C-a` if
> they'd rather undo the change.

+1

> I can imagine similarly changing the IU in case where the patch can be
> applied at several different places, where we'd ask the user where to
> apply it.

FYI, the common pattern that triggers this bug is repetitive code that
is changed. For example, when refactoring function calls after function
rename. So, having a good heuristics about the best place to apply the
chunk would still be valuable.

However, as a minimum, if a chunk can be both applied and reverted,
users should be able to choose between two variants. Although,
repetitive question like that would be annoying if there are many
duplicate changes.

-- 
Ihor Radchenko // yantar92,
Org mode maintainer,
Learn more about Org mode at <https://orgmode.org/>.
Support Org development at <https://liberapay.com/org-mode>,
or support my work at <https://liberapay.com/yantar92>