Re: [PATCH] merge-recursive: ignore_case shouldn't reject intentional removals

From: Elijah Newren <newren@gmail.com>
Date: Nov 27, 2017, 16:40 UTC
Message-ID: <CABPp-BELci79tuiPFVRKQwm_-j6tN-DxqcKsvjD2xG_8ZWNxGw@mail.gmail.com>
In-Reply-To: <xmqq4lpgfkyb.fsf@gitster.mtv.corp.google.com>

[Removed cc's that just bounce]

On Sun, Nov 26, 2017 at 7:40 PM, Junio C Hamano <gitster@pobox.com> wrote:

> Elijah Newren <newren@gmail.com> writes:
>
>> In commit ae352c7f3 (merge-recursive.c: fix case-changing merge bug,
>> 2014-05-01), it was observed that removing files could be problematic on

<snip>

Show 8 quoted lines

>>
>> If that description leaves more questions than answers, we may need to
>> augment the above commit message with the following explanation...
>> ...
>>  merge-recursive.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> As a fix, this sorely wants something new in t/ directory.

Well, then perhaps I was wrong to submit it independent of my directory rename series. As noted in the (very lengthy) extended commit message explanation, the assumption the previous code made just happened to work until a few extra tweaks (from directory renames) caused us to want to remove a file from the working copy that was found at stage 0 in the index. Thus, the only testcase we can really use for this commit, is testcase 7b of the new t6043 added by that other patch series, and it's only valid with the code from that other series.

When I submitted this patch, I was thinking about just including this fix with the next reroll of my rename-directory-detection series but it partially felt like an independent fix...but maybe I chose wrong.

Would you prefer I include it in my next en/rename-directory-detection reroll?