threads / discuss / 34099

patch, 2 partsRe: v3 [PATCH 2/2] status:introduce status.branch to enable --branch by default

Subject: Re: v3 [PATCH 2/2] status:introduce status.branch to enable --branch by default

## tl;dr

2 messages between Jun 11, 2013 and Jun 11, 2013. Diffs are folded; open one to read it.

replies: 1people: 2as markdown or json

Ramkumar Ramachandra· Jun 11, 2013, 07:30 UTC · lore
<y@ensimag.imag.fr> wrote:
> [...]
Good change.
> diff --git a/t/t7508-status.sh b/t/t7508-status.sh
> index 9a07f15..958617a 100755
> --- a/t/t7508-status.sh
> +++ b/t/t7508-status.sh

I expected one test. Two, at most. This is a bit overzealous. I don't mind leaving it as it is, but as a note for the future: this kind of overkill is not necessarily a good thing; it makes it hard to add/remove tests, multiple tests fail when one tiny feature doesn't work as expected, and the testsuite takes longer to run.

Matthieu Moy· Jun 11, 2013, 07:51 UTC · re: Ramkumar Ramachandra · lore
Ramkumar Ramachandra <artagnon@gmail.com> writes:
Show 11 quoted lines
> <y@ensimag.imag.fr> wrote:
>> [...]
>
> Good change.
>
>> diff --git a/t/t7508-status.sh b/t/t7508-status.sh
>> index 9a07f15..958617a 100755
>> --- a/t/t7508-status.sh
>> +++ b/t/t7508-status.sh
>
> I expected one test.  Two, at most.  This is a bit overzealous. 

I actually had this discussion with Jorge (in real life). I first thought this was too much, and I think we already removed one rendundant test. But actually, this should be "two at least", since you really want to test whether --[no-]branch takes precedence over status.branch or not, in addition to testing status.branch alone.

The last two tests are less important, but they also test something (in short: what happens when branch.status is set to something other than "true"). So I think it makes sense to keep them.

-- 
Matthieu Moy
http://www-verimag.imag.fr/~moy/

← back to recent threads