Re: js/bisect-in-c

From: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Date: May 27, 2022, 12:52 UTC
Message-ID: <220527.86k0a72klr.gmgdl@evledraar.gmail.com>
In-Reply-To: <xmqqo7zkxdvu.fsf@gitster.g>

On Thu, May 26 2022, Junio C Hamano wrote:

Show 16 quoted lines

> Junio C Hamano <gitster@pobox.com> writes:
>
>>>>  Final bits of "git bisect.sh" have been rewritten in C.
>>>>
>>>>  Will merge to 'next'?
>>>>  source: <pull.1132.v3.git.1653144546.gitgitgadget@gmail.com>
>>>
>>> This topic has outstanding regressions in CLI parsing. I.e. we'll now
>>> offer to start bisection where we previously errored out on invalid
>>> command usage. See my replies in that thread.
>>
>> Pointers?
>
> I guess you meant this one?
>
> https://lore.kernel.org/git/220521.86zgjazuy4.gmgdl@evledraar.gmail.com/

Yes, sorry about the lack of specific reference upthread.

As noted after in https://lore.kernel.org/git/220523.865ylwzgji.gmgdl@evledraar.gmail.com/ we're rewriting code in this case that doesn't have test coverage to catch these regressions, so more than just a narrow bugfix I'd really like to see a re-roll that gives some confidence that there aren't other such issues by first adding missing test coverage.