[PyOpenGL-Devel] [ pyopengl-Bugs-1592164 ] Regression: glutInit does not accept zero arguments

[SourceForge.net <no...@so...>]

Bugs item #1592164, was opened at 2006-11-07 19:07
Message generated for change (Comment added) made by dmurmann
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105988&aid=1592164&group_id=5988

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: GLUT
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: David Murmann (dmurmann)
Assigned to: Mike C. Fletcher (mcfletch)
Summary: Regression: glutInit does not accept zero arguments

Initial Comment:
regarding OpenGL-ctypes 3.0.0a4:

glutInit had this check in PyOpenGL 2.x:

interface/GLUT.i: line 87

PyObject *_glutInit(PyObject* args)
{
...
if (argc == 0)
{
	argv = PyMem_New(char*, argc = 1);
	argv[0] = "foo";
}
...
}

Something similar should be added to glutInit in
OpenGL/GLUT/special.py line 226.


----------------------------------------------------------------------

>Comment By: David Murmann (dmurmann)
Date: 2006-11-09 11:09

Message:
Logged In: YES 
user_id=1455846

Nope, sorry, still not quite. I attached a patch against
current HEAD that resembles the old behavior more closely
(probably should have done that in the first place, i don't
seem to be very good at explaining the problem ;-).

----------------------------------------------------------------------

Comment By: Mike C. Fletcher (mcfletch)
Date: 2006-11-07 23:10

Message:
Logged In: YES 
user_id=34901

Okay, attempted to fix again.  Please give it a try.

----------------------------------------------------------------------

Comment By: David Murmann (dmurmann)
Date: 2006-11-07 21:57

Message:
Logged In: YES 
user_id=1455846

First, thanks for the fast response, but this is not what i
meant
(sorry if my report was inaccurate). The python function
glutInit
did never accept zero arguments (i.e. glutInit() was always
illegal).
But glutInit([]) was valid, i.e. the argument vector could
be empty
(in which case PyOpenGL 2.x added a "foo" first argument). I
suspect
my GLUT implementation (win32) expects argv to be non-empty
with the
first entry being the program name.

This is the traceback I get in my app with glutInit([]):

D:\develop\nash\public\glslide>python glslide.py
Traceback (most recent call last):
  File "glslide.py", line 35, in <module>
    backend = OpenGLBackend(filename)
  File "D:\develop\nash\public\glslide\opengl_backend.py",
line 186, in __init__
    glutInit([])
  File
"D:\develop\python25\lib\site-packages\opengl-3.0.0a4-py2.5.egg\OpenGL\GLUT\special.py",
line 252, in glutInit
    _base_glutInit( ctypes.byref(count), holder )
WindowsError: exception: access violation reading 0x00000000


----------------------------------------------------------------------

Comment By: Mike C. Fletcher (mcfletch)
Date: 2006-11-07 19:31

Message:
Logged In: YES 
user_id=34901

Thanks for the report.  Fixed in CVS.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105988&aid=1592164&group_id=5988