[Xcb] adding testcases to xcb/demo, my brother will write testcases for xinput for two weeks. (Re: [PATCH demo 1/1] make xcb/demo compile again.)

[Christian Linhart]

On 2017-03-20 18:48, Eric Anholt wrote:
> Christian Linhart <chris at DemoRecorder.com> writes:
>> [...]
> Are these tests going in here going to be useful as X server regression
> tests?  I've wondered where the right home would be for little X
> regression tests, and would like to get more into piglit for keeping the
> X server working.
These tests will primarily be aimed at testing whether the xcb protocol-definitions are correct,
and whether the generator generates correct code.
The tests should be usable as regression tests for that.

But, as a nice side-effect, they'll do some X-Server testing, too.
Because if you (ideally) test all aspects of the protocol, you'll also get quite a good test-coverage for the X-server. So you definitely got your intuition right about that.

It is not yet clear how far we'll get though with that test suite.

> If they are regression tests, xcb/demo does seem like a funny place for
> them.  Maybe a new xcb/tests forked from demo?
Yes, we could do that mid term.
Forking from demo seems to be a good idea. (and then cleaning up to leave only the regression tests)
Short term I suggest to use xcb/demo and see how far we'll get.

BTW, who has the right to create new repos such as xcb/test?
And how long does it usually take to get attention from them?
(or do I have the rights and I am just not aware of it?)

I know that I have commit rights to at least xcb/proto and xcb/libxcb.
But I guess this does not include the right to create new repos...

Cheers,

Chris

>
>
> _______________________________________________
> Xcb mailing list
> Xcb at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/xcb


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/xcb/attachments/20170320/3cde0833/attachment.html>