[bind10-dev] reporting disabled tests?

JINMEI Tatuya / 神明達哉 jinmei at isc.org
Mon Jul 19 17:35:45 UTC 2010


At Mon, 19 Jul 2010 07:59:22 -0500 (CDT),
"Jeremy C. Reed" <jreed at isc.org> wrote:

> > Some of our tests *temporarily* disabled.  I emphasized "temporarily"
> > because it should basically be considered a short term workaround but
> > in reality we tend to forget/ignore them.  This may be more so
> > recently because the warning messages from gtest are often hidden from
> > python test reports (you can see it by running make check on top of
> > the source tree)
> 
> Can you please provide some specific examples?
> 
> I see a couple "make check" output with "YOU HAVE n DISABLED TESTS" 
> and I see 
> "[ RUN      ] CacheTest.disabled
> [       OK ] CacheTest.disabled (0 ms)"
> 
> And I see code with DISABLED_foo.
> 
> But I don't see any correlation in the "make check" output itself to 
> recognize the specific disabled checks.

What do you mean by that?

> > So I wonder whether we should be bugged about those periodically.
> > 
> > Does it make sense (and if so is it easy) to have buildbot report
> > disabled tests if there's any with a moderate rate limiting (e.g. at
> > most once a week)?
> 
> I will do this weekly. Can you please provide a regular expression 
> pattern(s) I can use to match them (since I can't see them now)?

Again, I'm not sure what you mean...oesn't the above pattern suffice?,
i.e.,
> "[ RUN      ] CacheTest.disabled
> [       OK ] CacheTest.disabled (0 ms)"
and so on.

---
JINMEI, Tatuya



More information about the bind10-dev mailing list