This thread is for all tests of pyglet 1.2dev in Mac OS X 10.7 (Lion),
using any 32-bit python.
In my case I am using ActivePython 2.7.2.5, on a Mac Mini with an AMD GPU.
I have not had time to do complete tests and probably won't for another 2
days or so. But my partial tests uncovered enough bugs to be worth
mentioning. It would be good if someone else with Lion can repeat any of
these bugs (or try and fail to repeat them), either with ActivePython
2.7.2.5 or any other python.
(I will retry them myself with other pythons on the same machine, when I
have time.)
The definite bugs I found:
- changing a fullscreen window to non-fullscreen doesn't work (window
remains fullscreen),
this is done by window.WINDOW_SET_FULLSCREEN when you press 'g'
and results in this traceback (which I've also consistently seen in my
Outlawn game when trying to do the same thing):
on_key_press(symbol=G, modifiers=)
Leaving fullscreen.
Traceback (most recent call last):
File "_ctypes/callbacks.c", line 313, in 'calling callback function'
File "tests/../pyglet/window/carbon/__init__.py", line 673, in
_on_key_down
self.dispatch_event('on_key_press', symbol, modifiers)
File "tests/../pyglet/window/__init__.py", line 1151, in dispatch_event
if EventDispatcher.dispatch_event(self, *args) != False:
File "tests/../pyglet/event.py", line 355, in dispatch_event
if handler(*args):
File "tests/../tests/window/WINDOW_SET_FULLSCREEN.py", line 33, in
on_key_press
self.w.set_fullscreen(False)
File "tests/../pyglet/window/__init__.py", line 662, in set_fullscreen
self.screen.restore_mode()
File "tests/../pyglet/canvas/carbon.py", line 217, in restore_mode
_oscheck(carbon.CGDisplayRelease(self.id))
File "tests/../pyglet/libs/darwin/__init__.py", line 44, in _oscheck
raise RuntimeError('Carbon error %d' % result)
RuntimeError: Carbon error 1001
on_key_release(symbol=G, modifiers=)
- the test window.MODE_SWITCH effectively hangs my machine (with a dark
screen) when i press option 'a' (the first one I tried); I had to power
toggle the machine
- in several preceding tests which correctly entered a mode in which the
dock and main menubar were gone (including the one for keyboard exclusive
mode, but also some others), the dock and main menubar did not come back
when I ended the test, but only when I clicked in some other app.
- I might as well also mention, though it's been mentioned before on this
list, that many pyglet processes which I run on this Lion machine, as well
as many Panda3D processes, print 2 or 3 messages like this to the console
shortly after starting (a few seconds), and perhaps in some cases
thereafter:
<date> <hostname> Python[438] <Error>: kCGErrorInvalidConnection:
CGSGetWindowTags: Invalid connection
<date> <hostname> Python[438] <Error>: kCGErrorFailure: Set a breakpoint @
CGErrorBreakpoint() to catch errors as they are logged.
I am not sure if this happens when I run them from python2.6 (since I
rarely do that these days and can't repeat them from that right now). It
often happens from 32-bit ActivePython 2.7.2.5 for pyglet, and from Apple's
system python 2.5 for Panda3D. It happened to someone else too, who
mentioned it on this list.
There were other tests which may have had incorrect results, but to be sure
I'd need to understand the desired behavior better and/or retry them, so
I'll report on those later (as well as on the tests I didn't try yet).
(Sorry for the incomplete report (including sketchy descriptions above),
but since I have no time for a better one now I figure some info is better
than no info.)
- Bruce Smith
--
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.
