[vlc-devel] [vlc-commits] picture: round the number of visible lines to the upper multiple

[Rémi Denis-Courmont]

Le maanantaina 16. huhtikuuta 2018, 19.50.55 EEST Steve Lhomme a écrit :
> Le 16/04/2018 à 18:32, Rémi Denis-Courmont a écrit :
> > Le maanantaina 16. huhtikuuta 2018, 19.25.54 EEST Steve Lhomme a écrit :
> >> Le 16/04/2018 à 17:09, Rémi Denis-Courmont a écrit :
> >>> Le maanantaina 16. huhtikuuta 2018, 17.56.34 EEST Tristan Matthews a
> >>> écrit
> >>> 
> >>>> On Mon, Apr 16, 2018 at 10:38 AM, Steve Lhomme <git at videolan.org> 
wrote:
> >>>>> vlc | branch: master | Steve Lhomme <robux4 at ycbcr.xyz> | Mon Apr 16
> >>>>> 15:35:23 2018 +0200| [82f649983443292bd893962cac0484b7df8d1c89] |
> >>>>> committer: Steve Lhomme
> >>>>> 
> >>>>> picture: round the number of visible lines to the upper multiple
> >>>> 
> >>>> This breaks make check for me:
> >>> Same here.
> >>> 
> >>> On which platform was the test suite run before submitting for review?
> >> 
> >> For the n-th time: make check doesn't work on Windows.
> > 
> > So what?
> > 
> > Passing the test suite on a platform is not optional.
> 
> If you mean submitted code should pass the test suite, yes I agree
> (unless the test is wrong, of course).

> If you mean all platform should pass the test suite then it's been
> optional forever.

I asked which platform. The instruction say to make ti pass on *a* platform. 
It should probably recommend X86 because of the extra assembler tests.

Passing on all platforms do not even mean anything. There is not a finite list 
of platforms, or where to draw the boundary between two similar platforms.

Besides, the test suite will never work in cross-compilation trees, on under-
provisioned systems, with critical system library bugs, with broken a tool 
chain, or any other combination too far off the autotools(/Meson)-expected 
Unix-like native build environment.

And yet we don't excuse OS/2, iOS or rPI devs from a passing test run.

-- 
Реми Дёни-Курмон
http://www.remlab.net/