Python Opengl Plugin Error

[stealth-kid]

hi. i'm trying to write a plugin, using python, that draws some stuff into the viewport. i want to translate this c++ code into python:

CODE
float modelview[16];
int i,j;
glPushMatrix();
glGetFloatv(GL_MODELVIEW_MATRIX , modelview);
for( i=0; i<3; i++ )
    for( j=0; j<3; j++ ) {
        if ( i==j )
            modelview[i*4+j] = 1.0;
        else
            modelview[i*4+j] = 0.0;
    }
glLoadMatrixf(modelview);
// draw something here
glPopMatrix();

the python code looks like this (i'll put just the relevant parts):
CODE
view.beginGL()
glFT.glPushMatrix()
modelview = [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]
glFT.glGetFloatv(OpenMayaRender.MGL_MODELVIEW_MATRIX , modelview)
for i in range(0, 3, 1):
    for j in range(0, 3, 1):
        if ( i==j ):
            modelview[i*4+j] = 1.0
        else:
             modelview[i*4+j] = 0.0
glFT.glLoadMatrixf(modelview)
glFT.glBegin(OpenMayaRender.MGL_LINE_LOOP)
#draw a square
glFT.glEnd()
glFT.glPopMatrix()
view.endGL()

when i load the plugin, i get this error:
QUOTE
# Traceback (most recent call last):
# File "/path/to/my/script.py", line 381, in draw
# glFT.glGetFloatv(OpenMayaRender.MGL_MODELVIEW_MATRIX , modelview)
# TypeError: in method 'MGLFunctionTable_glGetFloatv', argument 3 of type 'MGLfloat *' //

i also tryed modelview = glFT.glGetFloatv(OpenMayaRender.MGL_MODELVIEW_MATRIX), but it says it expects 3 arguments, not just 2.

so.... anyone has any idee how to fix this? thanks.

[uiron]

first parameter of the method is "self", so the error is actually how you initialize modelview. i don't know how to initialize it tbh, couldn not find any info on MGLfloat.

anyway... this whole C code is just an ugly way to construct an identity matrix, and you can safely minimize it to:

glPushMatrix()
glLoadIdentity()

// draw something here
glPopMatrix()

[stealth-kid]

thank you for the reply. you're right about the identity matrix, i posted that piece of code mainly because i get that kind of error when i try other stuff too and i figured that if someone coud tell me exactly how to fix that one, i might be able to fix the others.

anyway.... new problem: i need functions for cross product and normalize vectors. i found out they are included in the OpenGLContext.utilities module, but how can i import them?

[leovitch]

To rephrase the real question being asked, how does one read, and then set, the OpenGL modelview matrix explicitly from Python code? For related reasons, I want to do that.

A big part of the answer can be found at http://www.rtrowbridge.com/blog/2009/02/15...r-python-users/. Ryan explains a lot of tricks with the MScriptUtil class, which is clearly key to getting things done here. It provides the way to get the required pointer types. For instance, you can read back the viewport information with this code:

[codebox]su = OpenMaya.MScriptUtil()
print glFT.glGetFloatv(OpenMayaRender.MGL_VIEWPORT,su.asFloatPtr())
for i in xrange(4):
print su.getFloatArrayItem(su.asFloatPtr(),i)[/codebox]

Accordingly, you might think this code would read back the modelview matrix:

[codebox]su = OpenMaya.MScriptUtil()
print glFT.glGetFloatv(OpenMayaRender.MGL_MODELVIEW_MATRIX,su.asFloatPtr())
for i in xrange(16):
print OpenMaya.MScriptUtil.getFloatArrayItem(su.asFloatPtr(),i)[/codebox]

Instead, this code hard-crashes Maya. Clearly, something needs to change, and presumably what need to change is to somehow get MScriptUtil to allocate the 16 storage locations ahead of time. MScriptUtil's documentation is kind of thin, though, and it's not clear exactly how to do that. I tried using setFloatArray ahead of time:

[codebox]su = OpenMaya.MScriptUtil()
fptr = su.asFloatPtr()
for i in xrange(16):
OpenMaya.MScriptUtil.setFloatArray(fptr,i,float(i))
print glFT.glGetFloatv(OpenMayaRender.MGL_MODELVIEW_MATRIX,fptr)
for i in xrange(16):
print OpenMaya.MScriptUtil.getFloatArrayItem(fptr,i)[/codebox]

That keeps Maya alive a little bit longer, but it still crashes after a few repeated executions. Another variant was to use createFromList:

[codebox]su = OpenMaya.MScriptUtil()
su.createFromList(map(float,xrange(16)),16)
print glFT.glGetFloatv(OpenMayaRender.MGL_MODELVIEW_MATRIX,su.asFloatPtr())
for i in xrange(16):
print OpenMaya.MScriptUtil.getFloatArrayItem(su.asFloatPtr(),i)[/codebox]

This doesn't crash at all, but instead of printing the expected matrix, it always prints the initialized values (i.e., [0.0, 1.0, ... 15.0]). Anyway, if anyone has figured out how to read back the OpenGL Matrix from Python code, the answer would be much appreciated!

[david-coen]

have only been scripting maya for two days now (would usually program 3dstudio max in c++) and not that familiar with python, but took the time to register on this board as you got further than me on this before i gave up and looked online.

the following worked for me, gives the illusion of working, not certain if i can take the time to play around with getting it nicer

CODE
    testMatrix = ( 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00 )
    testMatrix2 = [ 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00 ]

    suFloatArray = maya.OpenMaya.MScriptUtil()
    #thought createFromList would have made a suDoubleArray? seems to make something big enough for glGetFloatv to fill
    suFloatArray.createFromList( testMatrix, 16 )
    suPointer = suFloatArray.asFloatPtr()

    glFT.glGetFloatv( maya.OpenMayaRender.MGL_PROJECTION_MATRIX, suPointer )

    for index in range(16):
      testMatrix2[ index ] = maya.OpenMaya.MScriptUtil.getFloatArrayItem(suPointer, index)

    print( str(testMatrix2) )

[david_corral]

I'm trying to apply this code for a camera alignment object, maya crash all the time :S
Anyone was able to make this work in python?
There are a lot of examples in C++ and seems to be way easier :S

This is my code:

CODE
#get camera inverse matrix
camView = om.MDagPath()
view.getCamera( camView )            
camInvMatrix = camView.inclusiveMatrix().inverse()                                            
matrix2DoubleArray = [om.MScriptUtil.getDoubleArrayItem(camInvMatrix[0],0), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[0],1), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[0],2), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[0],3),om.MScriptUtil.getDoubleArrayItem(camInvMatrix[1],0), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[1],1), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[1],2), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[1],3),om.MScriptUtil.getDoubleArrayItem(camInvMatrix[2],0), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[2],1), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[2],2), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[2],3),om.MScriptUtil.getDoubleArrayItem(camInvMatrix[3],0), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[3],1), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[3],2), om.MScriptUtil.getDoubleArrayItem(camInvMatrix[3],3)]                    

su = om.MScriptUtil()
su.createFromList(matrix2DoubleArray,16)
suPointer = su.asFloatPtr()
test = glFT.glGetFloatv(omRender.MGL_MODELVIEW,suPointer)#MGL_VIEWPORT

#set camera matrix to opengl
glFT.glPushMatrix()

glFT.glLoadMatrixd( test )#>>>>>>>>>>>CRAAAASSH

#start render
......

Everything seems to be ok but when i'm applying the GLFloat* array it crash....
Help!