Re: Some thoughts about our tests and the build time

Hey Miklos,

On Tue, May 17, 2016 at 5:30 PM, Miklos Vajna <vmiklos@collabora.co.uk>
wrote:

> Hi Markus,
>
> On Tue, May 17, 2016 at 10:26:07AM +0200, Markus Mohrhard <
> markus.mohrhard@googlemail.com> wrote:
> > It is just a toplevel make. So it only covers unitcheck and slowcheck and
> > not subsequentcheck. Therefore we have no java tests in these numbers.
>
> I see. If you perform your checks again, then please consider timing
> 'make check', not 'make' next time -- I hope that subsequentchecks are
> not in the "nobody runs them" category.

Well. I think most people don't run them. Even I don't run them most of the
time. But to be able to have a comparison here the numbers for a time make
check:

real    11m26.067s
user    67m2.876s
sys    1m24.788s


So based on that at least for the dbgutil build the java tests are actually
not that slow anymore. Well there are not that many and if you look how
many tests the slow cppunit tests I think they are still much slower. But
the numbers are now around 2:1 for the cppunit tests.


> Thanks,
>
> Miklos
>
> _______________________________________________
> LibreOffice mailing list
> LibreOffice@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/libreoffice