So regarding pixel testing, are you referring to true "ignorant" pixel 
testing (eg reading frame buffer values)? I was recently talking to a 
friend who is a very good QA tester. He has done pixel-level UI testing, 
and says it has certain inherent difficulties which are not at first glance 
obvious. Things like the rgb values for *all* pixels in the frame buffer 
changing even though the test inputs are identical and the results also 
look identical to human eyes. FWIW, I asked him if there was any free way 
to do this kind of testing, and he pointed out "T-Plan Robot"; I'm 
unassociated with the project: http://sourceforge.net/projects/tplanrobot/.

Then again, if you're reading OpenGL memory instead, that would sidestep 
these kinds of problems entirely. If this is so, I apologize for 
interjecting an ignorant comment :)

On Saturday, June 23, 2012 2:58:15 PM UTC-4, Winston wrote:
>
> I was thinking along the same lines--both testing pixels for specific 
> colors and having tests be in a separate process so they can be killed. 
>
> -ww 
>
> On Jun 23, 2012, at 10:03 AM, Jonathan Hartley wrote: 
>
> > I think some of the manual tests could be automated. e.g. the font ones 
> could look for font-colored pixels in the output, to establish the 
> approximate rectangle that the font has been rendered in. 
> > 
> > If fatal crashes terminating the test run early is a persistent problem, 
> an idea for the future might be to convert tests which actually fire up a 
> window to do so in a new process, so that it the process-under-test could 
> crash without terminating the test run. 
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"pyglet-users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/pyglet-users/-/et7cxMCuYVkJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/pyglet-users?hl=en.

Reply via email to