Re: [PyOpenGL-Users] glVertexAttribPointer(); loading platform-specific functions

[Mike C. F. <mcf...@vr...>]

Joshua Davis wrote:
...
>> that's extremely old now.  Can you add these lines to the top of your
>> script and run it, please:
>>
>>     import logging
>>     logging.basicConfig( level = logging.DEBUG )
>>     import OpenGL
>>     OpenGL.FULL_LOGGING = True
>>
>> it'll dump every call it makes onto the console, include all the
>> parameters it's passing.  That should let me inspect and see if we're
>> getting the wrong values showing up at the GL level.
>
>
> This is a great debugging tool! Thanks. So is it a Numpy vs. Numeric
> problem?
Nope, the info-log there is just telling you that the registered Numeric
(old Numpy) handler isn't loadable.  That's normal and desirable, you
don't want to be using Numeric in modern code, it's available for
backward compatibility with old software that doesn't want to convert to
Numpy.

The trace looks fine.  There's an unlikely possibility, which is that
the shader compilation happening before mainloop/display callback might
fail on OSX's GLUT.  To test that, comment out the calls to
initializeShaders and handleDisplayEvent at the top-level of your test
script and add:

    if not shaderProgram:
        initializeShaders()

to the top of your handleDisplayEvent() function.  I doubt *very*
strongly that is the problem, as *all* OpenGLContext initialization
happens in that same state (post context creation, pre mainloop) and
it's never been a problem before.  Didn't get to trying the code on
friend's Mac last night either.

Would be nice if we could confirm that this machine can handle
attributes at all (so we know whether this is a problem with the wrapper
or the driver).

Good luck,
Mike

-- 
________________________________________________
  Mike C. Fletcher
  Designer, VR Plumber, Coder
  http://www.vrplumber.com
  http://blog.vrplumber.com