Re: Testing packages with mock and Xnest?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

On Wed, Apr 25, 2012 at 10:46 AM, Toshio Kuratomi <a.badger@xxxxxxxxx> wrote:
> On Fri, Apr 20, 2012 at 12:15:45PM -0500, Richard Shaw wrote:
>> Here's the draft:
>> Comments/additions welcomed!
> Cool.  Ive read the draft now.  I thought it was going to be about using
> Xnest with graphical unittests.  Oops.  Doesn't 100% fit in the Packaging
> Guidelines but perhaps we can link to it when we mention that reviewers
> should test that a package functions as intended.

Hmm... I think that would be quite a bit more complicated. I don't
think Xnest would be the right tool for that. I really don't know
enough about X to guess. I would think that the build server probably
runs without X and even if it could run some sort of headless X just
for testing purposes we probably would have to make sure that each
package building in parallel that needed this feature (however
unlikely) ran on a different display. What's the limit on the number
of displays anyway?

The main reason I include it under the Packaging Guidelines umbrella
is that I use it to run rpmlint on installed packages, especially
libraries, where you can find additional problems that rpmlint can't
detect from checking the RPMs. I'm not saying we should make it a
requirement, but perhaps "strongly recommended" would be a good idea.

> /me notices that although that instruction is in the ReviewGuidelines we
> dont seem to have it on the main Guidelines page.

I assume you're talking about?

devel mailing list

[Fedora Announce]     [Fedora Kernel]     [Fedora Testing]     [Fedora Legacy Announce]     [Home]     [Fedora Tools]     [Fedora PHP Devel]     [Kernel List]     [Fedora Legacy]     [Fedora Maintainers]     [Fedora Maintainers]     [Fedora Desktop]     [PAM]     [Red Hat Development]     [Big List of Linux Books]     [Gimp]     [Yosemite News]

Add to Google Powered by Linux