Many thanks for the detailed feedback...
>
> I run the suite with --no-regression-check and all image tests looked
> OK except for image.TEXTURE_GRID. No window came up when I ran the
> test on its own, and I get an access violation if I run it after the
> regression tests.
>
I can't replicate this; created issue #50.  (The traceback doesn't  
seem particular to this test, perhaps it is the result of some  
external event?)

>
> I got some more errors I think I must have accidentally edited out
> last time (sorry for the n00b question, but any easy way to capture
> the output to a file?).

Yes: --log-file=log.txt (see tests/test.py -h for all options).

> Press return to begin test...
> INFO:root:Begin unit tests for image.TEXTURE_GRID
> INFO:root:6 tests run
> [P]assed test, [F]ailed test: f
> Enter failure description:
>> Nothing happened.
> ERROR:root:User marked fail for image.TEXTURE_GRID
> ERROR:root:Nothing happened.

This time it looks like it didn't crash (it's not supposed to produce  
any output, as of r777 it doesn't ask for confirmation).

> INFO:root:Testing model.obj.OBJ_LOAD.
etc.

I've marked all these tests (r778) that are clearly far from working  
as developer (though these are mostly Richard's domain, they may only  
need simple fixes).

>
> The first window test seems broken:
>
> INFO:root:Testing window.CONTEXT_SHARE.
> INFO:root:Begin unit tests for window.CONTEXT_SHARE
> ERROR:root:Error in window.CONTEXT_SHARE
> ERROR:root:Traceback (most recent call last):
>   File "c:\pyglet\tests\window\CONTEXT_SHARE.py", line 45, in
> test_context_nosha
> re_list
>     factory = get_factory()
> NameError: global name 'get_factory' is not defined

Ok, fixed in r779.

> In the window.EVENT_SHOW_HIDE test, no initial show event is printed
> out. Subsequent events are printed OK.

Well spotted, fixed in r780.

> In the window.WINDOW_CAPTION test, the counting caption window is too
> small to see the numbers, and I can't resize it. The unicode title
> looks right.

These windows are larger and resizable now (r781).  I'm actually not  
convinced by this unicode test, it only uses code points < 255.  If  
anyone speaks a higher-byte language I'd appreciate another test case  
(however there could be problems with missing fonts on some OSs).

>
> I suggest you print again the description of the expected results for
> each test right before asking if it passed or failed. Some tests have
> a lot of output so this text goes out of the terminal window, and the
> user may forget some details about what he's expected to check for.

Good thinking, I've implemented this and also disabled regression  
checking by default in r782.

> Although the window resize test looked right, I don't know if the size
> printed would correspond to the the client area only, or to the whole
> window including title bars and borders. A test for the former may be
> to draw two rectangular outlines; a green one that is just contained
> within the window, and a red one that is just outside it [i.e., the
> green rectangle has corners (x0=0, y0=0), (x=client_width - 1,
> y=client_height - 1), and the red one has (x0=-1, y0=-1),
> (x=client_width, y=client_height)]. Then tell the tester he should see
> a green outline in the border of the window, and no red at all.

Another good idea, implemented in r783, thanks.

> In the test that captures the mouse, I closed the window when the
> mouse was captured. After that, the mouse cursor would appear hidden
> in all tests. This is true even if I Ctrl-C out of test.py and restart
> with --start-with. I'm going to try rebooting, so I send this right
> away.
>

That was a tricky one!  Fixed in r785.

>
> WINDOW_SET_VSYNC fails with
> INFO:root:Begin unit tests for window.WINDOW_SET_VSYNC
> ERROR:root:Error in window.WINDOW_SET_VSYNC
> ERROR:root:Traceback (most recent call last):
>   File "c:\pyglet\tests\window\WINDOW_SET_VSYNC.py", line 45, in
> test_open_windo
> w
>     print 'vsync is %r' % self.w1.get_vsync()
>   File "c:\pyglet\pyglet\window\win32\__init__.py", line 355, in
> get_vsync
>     return bool(wglGetSwapIntervalEXT())
> NameError: global name 'wglGetSwapIntervalEXT' is not defined
>

Fixed in r786.

> The rest of non-developer tests run fine. I run the font and layout
> tests with --developer and I only got one failure and a warning:

I'll get around to fixing the font test cases eventually, the layout  
ones are a long way off stable.

Thanks again for the excellent reports.  If you can add any info to  
issue #50 that would be great.
Please also add an issue for TEXTURE_3D if that is still failing for  
you (can't replicate your problem from previous post here).

Cheers
Alex.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"pyglet-users" group.
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