Re: [PATCHv2 0/8] Add color test for range-diff, simplify diff.c

[Junio C Hamano <gitster@pobox.com>]

Stefan Beller <sbeller@google.com> writes:

> On Wed, Aug 1, 2018 at 12:13 PM Junio C Hamano <gitster@pobox.com> wrote:
>>
>> Stefan Beller <sbeller@google.com> writes:
>>
>> > Stefan Beller (8):
>> >   test_decode_color: understand FAINT and ITALIC
>> >   t3206: add color test for range-diff --dual-color
>> >   diff.c: simplify caller of emit_line_0
>> >   diff.c: reorder arguments for emit_line_ws_markup
>> >   diff.c: add set_sign to emit_line_0
>> >   diff: use emit_line_0 once per line
>> >   diff.c: compute reverse locally in emit_line_0
>> >   diff.c: rewrite emit_line_0 more understandably
>> >
>> >  diff.c                  | 94 +++++++++++++++++++++++------------------
>> >  t/t3206-range-diff.sh   | 39 +++++++++++++++++
>> >  t/test-lib-functions.sh |  2 +
>> >  3 files changed, 93 insertions(+), 42 deletions(-)
>>
>> As I cannot merge this as is to 'pu' and keep going, I'll queue the
>> following to the tip.  I think it can be squashed to "t3206: add
>> color test" but for now they will remain separate patches.
>>
>> Subject: [PATCH] fixup! t3206: add color test for range-diff --dual-color
>
> Thanks for dealing with my stubbornness here.

I didn't know you were stubborn---I just thought you were busy in
other things.

> I assumed that the contribution would be a one time hassle
> during git-am, not an ongoing problem during the whole time
> the patch flows through pu/next/master, which is why I punted
> on doing this change and resending in a timely manner.

It is a bit worse, in that it won't be at pu/next/master boundary
but each and every time the integration branches are rebuilt, which
happens a few times a day.  Whitespace breakage after a merge is
detected and my automation stops to ask for manual inspection.