harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Stepan Mishura" <stepan.mish...@gmail.com>
Subject [buildtest] Are there obsolete fields in the test's description? (Was: Re: [bti][p-unit] finally, it is ready)
Date Tue, 30 Oct 2007 08:54:57 GMT
Hi Alexei,

Thanks for your comments - so your idea that "date-of-creation" and
"authors" fields in the test's description became obsolete.

See my comment below

On 10/29/07, Alexei Fedotov <alexei.fedotov@gmail.com> wrote:
> Hello Stepan,
> I'm gratefull for your interest to this patch.
>
> >  why did you remove logging for "date-of-creation" and
> > "authors" field?
>
> I believe these fields are now excessive. A date of creation may be
> got from svn as well as an author. I was not a supporter or author
> removal idea because svn lists only committers, but this was accepted
> anyway, and now we should to follow.
>

- date of creation != date of first check in
- list of authors != list of committers

So I'd call these fields (as well the corresponding options) -
developer's fields.

> I agree that obsolete warnings may be removed by decreasing logging
> level, but this is not a correct way to follow to my personal point of
> view.
>

What made them "obsolete"?

> > Is is necessary for p-unit integration? I guess no -
> > you can change logging level (or fix tests description files) to avoid
> > unwanted logging.
>
> And you are correct. It is not necessary for p-unit integration. The
> goal is not p-unit integration at all.
>The goal is adding new tests
> and integrating them seamlessly, and absence of obsolete warnings is
> important for seamless integration.

I'm not sure that for adding new tests these warning have to be removed.

Also please keep in mind that the harness is used by several suites
and developers. I think if some options/fields/functionality became
obsolete (and should be removed due to different reasons) then it is
better to discuss it first. If there are no objections then we can
remove it. And it should be removed entirely, not partly as your patch
does.

In this case I don't know if "date-of-creation" and "authors" fields
are useful in test's description or they put additional burden on
developers. So I've changed the subject to to hear opinions of the
harness's users .

Thanks,
Stepan.

>
> Thanks!
On 10/29/07, Stepan Mishura <stepan.mishura@gmail.com> wrote:
> On 10/26/07, Alexei Fedotov <alexei.fedotov@gmail.com> wrote:
> > Stepan, sure. Thank you for your interest to this patch.
> >
> > For reviewer convinience the harness patch is split into two:
> > Harmony-style formatting changes were separated from removing two
> > warning outputs and authorship [1]. While I respect a contribution of
> > Andrey Tyuryushkin, I have learned the the author removal trick from
> > Sveta.
> >
>
> Alexei,
>
> Thanks for providing new patch that free from formatting changes -
> that made the review easier. Now I have a question to your updates to
> the harness: why did you remove logging for "date-of-creation" and
> "authors" field? Is is necessary for p-unit integration? I guess no -
> you can change logging level (or fix tests description files) to avoid
> unwanted logging.
>
> Thanks,
> Stepan.
>
> > Thanks.
> >
> > [1] http://issues.apache.org/jira/secure/attachment/12368462/harness_review.patch

Mime
View raw message