Re: Towards a better testsuite: Attributions

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Towards a better testsuite: Attributions

fantasai
On 03/24/2016 01:00 PM, Geoffrey Sneddon wrote:

>
> I went through all of the all of the metadata in
> <https://lists.w3.org/Archives/Public/public-css-testsuite/2015Oct/0004.html>;
> there's rough agreement on removing much of what we currently have.
> That said, I think it's worthwhile to reiterate that requiring *any*
> metadata causes friction. Tests written by browser vendors are rarely
> a file or two which is quick to add metadata too. I know in general
> people seem interested in using the same infrastructure to run both
> web-platform-tests and csswg-test, which essentially requires the
> metadata required to run the tests be identical across the two.

Wrt authoring information... It would be convenient not to
have to include these in the test files. However, in most
projects his information is in the checkin logs, and we
don't really have those for tests that are shared across
repositories like this.

Its certainly possible the browser vendors don't care, but
individual contributors and other organizations that contribute
tests (which we want more of, not less of) might appreciate being
acknowledged for their contributions to a particular test suite.

I don't have a good answer, but I think we should have one.
Simplest case, we can keep <link rel="author">, but make it
optional.

~fantasai

Reply | Threaded
Open this post in threaded view
|

Re: Towards a better testsuite: Attributions

Florian Rivoal-4

> On Apr 9, 2016, at 02:00, fantasai <[hidden email]> wrote:
>
> On 03/24/2016 01:00 PM, Geoffrey Sneddon wrote:
>>
>> I went through all of the all of the metadata in
>> <https://lists.w3.org/Archives/Public/public-css-testsuite/2015Oct/0004.html>;
>> there's rough agreement on removing much of what we currently have.
>> That said, I think it's worthwhile to reiterate that requiring *any*
>> metadata causes friction. Tests written by browser vendors are rarely
>> a file or two which is quick to add metadata too. I know in general
>> people seem interested in using the same infrastructure to run both
>> web-platform-tests and csswg-test, which essentially requires the
>> metadata required to run the tests be identical across the two.
>
> Wrt authoring information... It would be convenient not to
> have to include these in the test files. However, in most
> projects his information is in the checkin logs, and we
> don't really have those for tests that are shared across
> repositories like this.

I don't know about mercurial (and other VCS that might be in use), but at least
as far as git is concerned, author and committer information are tracked separately,
and preserving author information should be doable even when migrating TCs across repos, even if the committer information is lost/replaced by that of the sync script.

> I don't have a good answer, but I think we should have one.
> Simplest case, we can keep <link rel="author">, but make it
> optional.

Regardless of the VCS capabilities, we should do that. It costs nothing to the people who don't care and gives a chance to the people who do to make sure this information is captured.

 - Florian