pyopengl-users — Discussion of all PyOpenGL matters.

[Pyopengl-users] [ pyopengl-Bugs-459735 ] Build failures with 2.0.0.44

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

Bugs item #459735, was opened at 2001-09-07 21:59
You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=105988&aid=459735&group_id=5988

Category: build
Group: v2.0
Status: Open
Resolution: None
Priority: 5
Submitted By: Richard Jones (richard)
Assigned to: Tarn Weisner Burton (twburton)
Summary: Build failures with 2.0.0.44

Initial Comment:
Out of the box, without SWIG installed, I got build 
errors.

I have attached the entire build log. The build was 
attempted using the PyOpenGL-2.0.0.44.tar.gz source 
distribution.

With SWIG installed, I forced a build_w and it still 
broke. That was using swig 1.3a5.

Relevant installed packages:
python-2.1.1-3mdk
python-devel-2.1.1-3mdk
python-numeric-20.1.0-1mdk
python-numeric-devel-20.1.0-1mdk
tkinter-2.1.1-3mdk
Mesa-3.4.1-4mdk
Mesa-common-3.4.1-4mdk
Mesa-common-devel-3.4.1-4mdk
swig-1.3a5-1mdk

Note: the log also contains some other type warnings.



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

You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=105988&aid=459735&group_id=5988

[Pyopengl-users] [ pyopengl-Bugs-459730 ] Linux-Mandrake build uses wrong X11 dirs

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

Bugs item #459730, was opened at 2001-09-07 21:31
You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=105988&aid=459730&group_id=5988

Category: build
Group: v2.0
Status: Open
Resolution: None
Priority: 5
Submitted By: Richard Jones (richard)
Assigned to: Tarn Weisner Burton (twburton)
Summary: Linux-Mandrake build uses wrong X11 dirs

Initial Comment:
The config file for the linux build specifies 
/usr/X11/[include|lib] for finding includes/libs. 
This is not correct for mandrake (at least 8.0) which 
has them installed in /usr/X11R6/[include|lib]. The 
python setup.py has detection of libraries using 
various paths, perhaps your setup.py could use 
similar mechanisms?



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

You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=105988&aid=459730&group_id=5988

[Pyopengl-users] Mesa 3.1 and 3.2 fix

[Tarn W. B. <twb...@ma...>]

As some may know there are some problems compiling PyOpenGL2 on Mesa 3.1 or
3.2.  This is through no fault of PyOpenGL, but instead occurs because Mesa
has a misbehavin' GLU header.  I've added some stuff to the setup script to
detect this and execute a work around.  The modified setup script is in
"mesa-setup-patch.tar.gz".  In all, the possible work arounds are:

1) Use the modified setup script (first extract "PyOpenGL-2.0.0.44.tar.gz"
then "mesa-setup-patch.tar.gz" from the same folder),
2) or comment out the line "#define GLU_VERSION_1_2 1" in the GLU header
"GL/glu.h" ,
3) or upgrade to the SGI GLU library which supports GLU 1.3 and is available
via RPM at the Mesa homepage,
4) or upgrade Mesa, also available at the Mesa homepage.

thanks all,
Tarn

RE: [Pyopengl-users] patch to trackball.py

[Tarn W. B. <twb...@ma...>]

| I've got OpenGLContext installed, maybe I'll have a look at those. The
| demos in the OpenGL directory don't use vertex arrays at all (as far as
| I grepped).

da/dots.py uses glVertexPointerd

Tarn

RE: [Pyopengl-users] patch to trackball.py

[Alastair (L. T. <liq...@op...>]

Hi Mike

On 05 Sep 2001 18:25:20 -0400, Mike C. Fletcher wrote:
> The Trackball in the base distribution (PyOpenGL) likely hasn't been used,
> save possibly by some of the TK demos.  I use a different Trackball (and
> Quaternion) implementation in OpenGLContext.  Suppose we need some tests for

I just checked the changelog and it seems like trackball.py was only
added since the last point release. I'm happy that it least its working
now :). I was planning to use the OpenGLContext, but I was porting an
OpenGL program I was working on in C to Python to check out the
performance differences. So using plain OpenGL + GL* was more suitable
than figuring out OpenGLContext.

> Vertex arrays certainly work for me.  Can you tell us what your system
> configuration is?  Do you have OpenGLContext installed?  If so, it's got
> lots of tests in the tests directory that do basic vertex, normal, colour,

I've got OpenGLContext installed, maybe I'll have a look at those. The
demos in the OpenGL directory don't use vertex arrays at all (as far as
I grepped).

Thanks,

Alastair.

-- 
________________________________________________________________________
 Alastair Tse -=- https://www.liquidx.net/ -=- al...@cs...
     ___                                   -=- li...@op...
    /__/\  - opengl for ascii              -=- icq - 118396999
    \__\/  
" I'm sorry, our software is perfect.  The problem must be you."
  -- Dogbert

RE: [Pyopengl-users] patch to trackball.py

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

The Trackball in the base distribution (PyOpenGL) likely hasn't been used,
save possibly by some of the TK demos.  I use a different Trackball (and
Quaternion) implementation in OpenGLContext.  Suppose we need some tests for
the one in the base distribution.  Also need to work on unifying the two
some day I suppose.

Vertex arrays certainly work for me.  Can you tell us what your system
configuration is?  Do you have OpenGLContext installed?  If so, it's got
lots of tests in the tests directory that do basic vertex, normal, colour,
etceteras array drawing.  Those all run fine on my machine, so if they don't
on your machine it's likely a compilation or similar problem rather than
user error.

Thanks for the patches, I'll let Tarn do the integration, since it's his
module these days.  Enjoy,
Mike

-----Original Message-----
From: pyo...@li...
[mailto:pyo...@li...]On Behalf Of Alastair
(LiQUiDx) Tse
Sent: September 5, 2001 11:06
To: pyo...@li...
Subject: [Pyopengl-users] patch to trackball.py
...

1. trackball.py
...
2. Vertex Arrays in PyOpenGL 2.0.44

Are the working ? I just get seg faults everytime I try glDrawElement()
and others ? Maybe it's just me :)
...

[Pyopengl-users] [ pyopengl-Feature Requests-458589 ] Real Time Shadow Library

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

Feature Requests item #458589, was opened at 2001-09-04 18:47
You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=355988&aid=458589&group_id=5988

Category: new module
Group: v2.1
Status: Open
Resolution: None
Priority: 3
Submitted By: Mike C. Fletcher (mcfletch)
Assigned to: Tarn Weisner Burton (twburton)
Summary: Real Time Shadow Library

Initial Comment:
A wrapper for a library similar to 
http://reality.sgi.com/mjk/tips/rts/ would be very 
cool.  I'm guessing there are better libraries 
somewhere, but I haven't found them with some 
preliminary searching.

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

>Comment By: Mike C. Fletcher (mcfletch)
Date: 2001-09-05 14:32

Message:
Logged In: YES 
user_id=34901

Let's ask Mark for a clarification of the license, my 
reading is that it's fine, but we'll make sure before we go 
ahead with it.

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

Comment By: Tarn Weisner Burton (twburton)
Date: 2001-09-04 18:58

Message:
Logged In: YES 
user_id=21784

This was already done back in 2.0a2, and killed in 2.0a3.  
I killed it because I wasn't sure about the license.  It 
has the same license as GLUT.  Do you me want to resurrect 
it?

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

You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=355988&aid=458589&group_id=5988

RE: [Pyopengl-users] patch to trackball.py

[Alastair (L. T. <liq...@op...>]

Hi Tarn,

On 05 Sep 2001 10:44:17 -0500, Tarn Weisner Burton wrote:
> I applied some of your changes, but not all.  One thing that worries me is
> the line
> 
> phi = 2.0 * math.asin(t)

Hehe .. It must have been the comment I wrote below that threw you :)

Actually, I was debugging something and I forgot to remove it. That was
basically from the SGI code that implemented the trackball. The original
trackball.py was right (at least to my little trig knowledge). You can
ignore that line. So the bit should go :

t = max(min(t, 1.0), -1.0)
			
scale = 1.0/math.sqrt(a[0]**2 + a[1]**2 + a[2]**2)
q = [math.sqrt(1.0-t**2)] + map(lambda x, y: x*y, a, [t*scale]*3) 

> In the future please submit patches via the Patch Tracker at the sourceforge
> page.  Also you can always check the current status of code to see if a bug
> has already been fixed using the CVS browser.

Yeah, thanks for the suggestion. I will check the CVS code next time. In
fact I only had a look at the trackball.py code today and was in a rush
to get it working.

> Yep.  Look at the documentation.  The array functions have slightly
> difference prototypes that the spec so this might be the cause.

Thanks for the pointer. I'll take a look at the documentation again.

Cheers,

Alastair.

> 
> thanks,
> Tarn Weisner Burton
> 
> 
> _______________________________________________
> Pyopengl-users mailing list
> Pyo...@li...
> https://lists.sourceforge.net/lists/listinfo/pyopengl-users
-- 
________________________________________________________________________
 Alastair Tse -=- https://www.liquidx.net/ -=- al...@cs...
     ___                                   -=- li...@op...
    /__/\  - opengl for ascii              -=- icq - 118396999
    \__\/  
" I'm sorry, our software is perfect.  The problem must be you."
  -- Dogbert

RE: [Pyopengl-users] patch to trackball.py

[Tarn W. B. <twb...@ma...>]

| 1. trackball.py
|
| I have been playing around with trackballing under C, and I find that
| the trackball.py is horribly broken. It makes me wonder whether it was
| included there by accident. After spending quite a bit of time looking
| at it, I found that there were a fair number of bugs in it, which I have
| fixed up.

No, quaternion was a late addition and I obviously didn't look at trackball
close enough.

I applied some of your changes, but not all.  One thing that worries me is
the line

phi = 2.0 * math.asin(t)

in your source.  This is not in any CVS version.  Which
distribution/platform did you install?

In the future please submit patches via the Patch Tracker at the sourceforge
page.  Also you can always check the current status of code to see if a bug
has already been fixed using the CVS browser.

| 2. Vertex Arrays in PyOpenGL 2.0.44
|
| Are the working ? I just get seg faults everytime I try glDrawElement()
| and others ? Maybe it's just me :)

Yep.  Look at the documentation.  The array functions have slightly
difference prototypes that the spec so this might be the cause.

thanks,
Tarn Weisner Burton

[Pyopengl-users] patch to trackball.py

[Alastair (L. T. <liq...@op...>]

Hi PyOpenGL Developers,

First things first, Thanks for the OpenGL python class that you have all
created. It's very impressive indeed!

I have a few issues that I want to bring up,

1. trackball.py

I have been playing around with trackballing under C, and I find that
the trackball.py is horribly broken. It makes me wonder whether it was
included there by accident. After spending quite a bit of time looking
at it, I found that there were a fair number of bugs in it, which I have
fixed up.

The patch is attached. It's made against the PyOpenGL-2.0.44
OpenGL/trackball.py using diff -u.

Basically, the quaternion.py included has the scalar as the first
argument rather than the last as in most implementations. Other changes
are included in the patch - so I won't bore you guys here :)

2. Vertex Arrays in PyOpenGL 2.0.44

Are the working ? I just get seg faults everytime I try glDrawElement()
and others ? Maybe it's just me :)

Thanks guys,

Alastair
-- 
________________________________________________________________________
 Alastair Tse -=- https://www.liquidx.net/ -=- al...@cs...
     ___                                   -=- li...@op...
    /__/\  - opengl for ascii              -=- icq - 118396999
    \__\/  
" I'm sorry, our software is perfect.  The problem must be you."
  -- Dogbert

[Pyopengl-users] RE: [PyOpenGL] installation wizard bug report

[Tarn W. B. <twb...@ma...>]

|   I am trying to install the OpenGLPy package for Windows and I am running
| into trouble with the installation
| utility. On the second page of the install wizard it prompts me for an
| installation directory. I can't get focus
| for this text box no matter what, its not active.

Looks like the install can't find the current python installation.  Try
reinstalling Python.

Tarn

[Pyopengl-users] FW: [PyOpenGL] installation wizard bug report

[Tarn W. B. <twb...@ma...>]

Hello,

  I am trying to install the OpenGLPy package for Windows and I am running
into trouble with the installation
utility. On the second page of the install wizard it prompts me for an
installation directory. I can't get focus 
for this text box no matter what, its not active.

Thanks in advance,

David Schere

[Pyopengl-users] [ pyopengl-Feature Requests-458589 ] Real Time Shadow Library

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

Feature Requests item #458589, was opened at 2001-09-04 18:47
You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=355988&aid=458589&group_id=5988

Category: new module
Group: v2.1
Status: Open
Resolution: None
Priority: 3
Submitted By: Mike C. Fletcher (mcfletch)
Assigned to: Tarn Weisner Burton (twburton)
Summary: Real Time Shadow Library

Initial Comment:
A wrapper for a library similar to 
http://reality.sgi.com/mjk/tips/rts/ would be very 
cool.  I'm guessing there are better libraries 
somewhere, but I haven't found them with some 
preliminary searching.

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

>Comment By: Tarn Weisner Burton (twburton)
Date: 2001-09-04 18:58

Message:
Logged In: YES 
user_id=21784

This was already done back in 2.0a2, and killed in 2.0a3.  
I killed it because I wasn't sure about the license.  It 
has the same license as GLUT.  Do you me want to resurrect 
it?

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

You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=355988&aid=458589&group_id=5988

[Pyopengl-users] [ pyopengl-Feature Requests-458589 ] Real Time Shadow Library

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

Feature Requests item #458589, was opened at 2001-09-04 18:47
You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=355988&aid=458589&group_id=5988

Category: new module
Group: v2.1
Status: Open
Resolution: None
Priority: 3
Submitted By: Mike C. Fletcher (mcfletch)
Assigned to: Tarn Weisner Burton (twburton)
Summary: Real Time Shadow Library

Initial Comment:
A wrapper for a library similar to 
http://reality.sgi.com/mjk/tips/rts/ would be very 
cool.  I'm guessing there are better libraries 
somewhere, but I haven't found them with some 
preliminary searching.

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

You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=355988&aid=458589&group_id=5988

[Pyopengl-users] (no subject)

[Gil T. [ateliermobile] <gi...@at...>]

[Pyopengl-users] RE: [PyOpenGL] Macintosh OS9 problems

[Tarn W. B. <twb...@ma...>]

| I've had no success at all installing pyopengl 2.0.0.42beta with Python 
| 2.1.1 for Mac OS.  I installed pyopengl on NT last week with some 
| effort, but I did not have to change any installation scripts. However, 
| The dist.py script in setup bombs. First there was an embedded path for 
| 'OpenGL/Demo' that bombed, I had to change the script to ':OpenGL:Demo' 
| just to get past the problem . 

Fixed in cvs. 

| Now, the install scripts are reading .h 
| files and trying to interpret them. eh? I thought that Codewarrior 
| would have been called to compile the C code as VC++ was called in NT.

What do you mean?  Which setup script?


Tarn Weisner Burton

[Pyopengl-users] FW: Macintosh OS9 problems

[Tarn W. B. <twb...@ma...>]

Greetings,

I've had no success at all installing pyopengl 2.0.0.42beta with Python 
2.1.1 for Mac OS.  I installed pyopengl on NT last week with some 
effort, but I did not have to change any installation scripts. However, 
The dist.py script in setup bombs. First there was an embedded path for 
'OpenGL/Demo' that bombed, I had to change the script to ':OpenGL:Demo' 
just to get past the problem . Now, the install scripts are reading .h 
files and trying to interpret them. eh? I thought that Codewarrior 
would have been called to compile the C code as VC++ was called in NT.

Anyway, if anyone out there has ever successfully installed pyopengl on 
Mac OS9, please let me know. The new Mac OS version of Python has 
distutils and includes Numeric, so it would be ideal to get pythonogl 
working. 

I've got Mac OSX on another machine too. Does anyone know if pyopengl 
works with OSX?

Thanks,

Bryce aka:  mas...@ea...

[Pyopengl-users] Post Test

[Tarn W. B. <twb...@ma...>]

Mailing list post test.

Tarn Weisner Burton