Re: ds/reachable (was Re: What's cooking in git.git (Jul 2018, #03; Wed, 25))

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

Derrick Stolee <stolee@gmail.com> writes:

>>   Stuck in review?
>>   cf. <20180723203500.231932-1-jonathantanmy@google.com>
>
> This comments on the initial values of 'struct ref_filter' (that are
> not used). All we need is the diff below squashed into "test-reach:
> test commit_contains".
>
>>   cf. <20180723204112.233274-1-jonathantanmy@google.com>
> This comment asks why "parse_commit()" instead of
> "parse_commit_or_die()" but the _or_die() would create a change in
> behavior that is not the purpose of the series.
>>   cf. <CAGZ79kb7tWV=cmboA+nsChAUaiC+fVVM-GBCuWfsypC+-wyaVg@mail.gmail.com>
>
> I just responded to Stefan's comment about sorting. I don't believe
> any change is needed. Some tests output multiple results and the order
> is not defined by the method contract, so 'test-tool reach <verb>'
> will always sort the output (by OID).

Just to let everybody know, there is no point responding to all of
"cf. <message>" comments in "What's cooking" report.  Because it is
*NOT* meant as an exhaustive list of things that need to be fixed,
refuting each and every one of them would not make the topic
"objection free" anyway.  The list of cf.'s are there to have just
enough of them to remind me to refrain from merging a topic to
'next' too hurriedly.  The discussion thread in the list archive is
the authoritative record of the discussion; treat "What's cooking"
as my personal note, nothing more.

> (Sorry for the delay. I'm on vacation.)

That's OK, and enjoy your time off.  We are not in a hurry.


> Thanks,
> -Stolee
>
> ---
>
> diff --git a/t/helper/test-reach.c b/t/helper/test-reach.c
> index eb21103998..ca30059117 100644
> --- a/t/helper/test-reach.c
> +++ b/t/helper/test-reach.c
> @@ -117,6 +117,7 @@ int cmd__reach(int ac, const char **av)
>     struct ref_filter filter;
>     struct contains_cache cache;
>     init_contains_cache(&cache);
> +    memset(&filter, 0, sizeof(filter));
>
>     if (ac > 2 && !strcmp(av[2], "--tag"))
>         filter.with_commit_tag_algo = 1;