copy transform

[Sebastian Schoellhammer]

Hello there,

I want to copy the transformation from one object onto another while keeping its shape in the exact same position.

I couldn't figure out a better way than first applying the transform matrix and then transforming all the meshes vertexes back by the inverse matrix.

Heres my script:

import maya.cmds as mc
from cgkit.cgtypes import mat4


def copyTransform(source, toCopy):
    #get the matrix
    m = mc.xform(source, worldSpace = True, matrix=True, query=True)
    mInv = mat4(m).inverse()
    mc.select(toCopy)
    #reset transform
    mc.makeIdentity(apply=True, t=1, r=1, s=1, n=0)
    mel.eval("performResetTransformations(0)")
    #apply source transform
    mc.xform(toCopy, worldSpace = True, matrix=m)

    #transform all verts inversely
    numVerts = mc.polyEvaluate(toCopy,vertex=True)
   
    for i in range(0,numVerts):
        n = toCopy+".vtx["+str(i)+"]"
        mc.xform(n, matrix=mInv.toList(1))
    mel.eval("DeleteHistory")

sel = mc.ls(sl=True)
last = sel.pop()
for s in sel:
    copyTransform(last, s)

It's seems a bit silly to me and im sure there must be a better way... Anyone?

Thanks so much,

seb

--
--
Sebastian Schoellhammer
３DArtist・Technical Director
www.subdivme.com/sebscorner

[Sylvain Berger]

using constraints then deleting the constraints might be a quick and dirty way to do what you want

2008/4/14 Sebastian Schoellhammer <sschoel...@gmail.com>:

--
They say, "Evil prevails when good men fail to act." What they ought to say is, "Evil prevails."
Nicolas Cage as Yuri Orlov in Lord of War.

[Olivier Renouard]

Hi,

I guess you don't want to use pivots. You could do this which amounts to the same, but could be faster by using
mc.makeIdentity (apply=True, translate=True, rotate=True, scale=True) :

Just try this, the idea is, apply inverse source transform, freeze transform, apply source transform

def copyTransform(source, toCopy):
    #get the matrix
    m = mc.xform(source, worldSpace = True, matrix=True, query=True)
    mInv = mat4(m).inverse()

    mi = []
    for i in xrange(4) :
        for j in xrange(4) :
            mi.append(mInv[j][i])
    # apply source inverse transform
    mc.xform(toCopy, worldSpace = True, relative = True, matrix=mi)
    #reset transform
    mc.makeIdentity(toCopy, apply=True, t=1, r=1, s=1, n=0)
    #apply source transform
    mc.xform(toCopy, worldSpace = True, relative = True, matrix=m)

I see you use cgkit for the matrix math. I'd like to know who on list would be for the use of cgkit, or Numpy or Maya's api for doing matrix , vector operations and such related things ? What's your thoughts ?

Olivier

[Chadrik]

i have not used cgkit or numpy, but a cursory glance shows that they
both need to be compiled per platform. the fact that MVector and
MMatrix offer a pre-compiled solution makes them quite attractive just
from the distribution standpoint. also, the consistency of the
interface with the rest of maya is a plus. furthermore numpy seems a
bit of overkill for maya.

however, like i said, i don't have experience with them, so i'd like
to hear what others have to say, particularly cgkit versus maya api.

-chad




On Apr 16, 2:44 pm, Olivier Renouard <o.renou...@attitude-studio.com>
wrote:

> >www.subdivme.com/sebscorner<http://www.subdivme.com/sebscorner>

[Ofer Koren]

just happened to use numpy recently but it was for a more complicated matrix calculation, using matrices bigger than the 4x4 that maya allows (it does allow only 4x4 right?)
Otherwise I use maya's or pymel's.

2008/4/16 Chadrik <cha...@gmail.com>:

--


- Ofer
www.mrbroken.com