I am facing a strange problem with Blender. I start Blender with a script as parameter which is supposed to apply all transformations to an imported fbx model. The import itself works fine.

Part of the code is following:

for o in bpy.context.scene.objects:
        o.select = True
        bpy.context.scene.objects.active = o
        bpy.ops.object.transform_apply(location = True, scale = True, rotation = True)
        o.select = False

When running the code like that, Blender does not apply the transformations. Instead it prints:

Warning: Objects have no data to transform

for every single object.

The really starnge thing is, when I paste the exact same script snippet into the editor field and run it, Blender executes it like expected and applies the transformation.

I really tried to find a solution, but did not find one. Thank you in advance!

  • $\begingroup$ Also related in that you can run the operator once per scene by selecting the objects first. $\endgroup$
    – batFINGER
    May 7 '18 at 10:11
  • $\begingroup$ What command line are you using? the order is important Surely this can be done without a workaround such as the one suggested in your answer. $\endgroup$
    – batFINGER
    May 7 '18 at 16:10

Make sure that you actually need to apply transformations. If you are writing an exporter, multiplying the object.matrix_world by the vertex position gives you the transformed value without altering the original mesh.

In case you still need to do it, do what batFINGER suggested.

  • $\begingroup$ Thank you for the answers. :-) I am working on a rendering pipeline and need the objects measurement for scaling and positioning purposes in the scene. The assets our client delivered have often many groups or parent/child relations, many different scalings, rotations, pivot positions in all layers and so on. I need also to reset the empties values, because their children inherit the transformations if a empty is a grouphead. $\endgroup$
    – S.Hermes
    May 7 '18 at 13:48

The answer from CoDEmanX to this question actually helped me solve my problem.

How to programmatically load a python script in Text Editor and launch it

I did exactly what is the topic of that post. I was not able to run the code from commandline argument script so I loaded it to the blender text editor and run it from there.

The code in the main function:

filename = os.getcwd() + "\\\src\\python\\blender\\render\\apply.py"
text = bpy.data.texts.load(filename)

for area in bpy.context.screen.areas:
    if area.type == 'TEXT_EDITOR':
        area.spaces[0].text = text # make loaded text file visible

        ctx = bpy.context.copy()
        ctx['edit_text'] = text # specify the text datablock to execute
        ctx['area'] = area # not actually needed...
        ctx['region'] = area.regions[-1] # ... just be nice


... and apply.py

import bpy

for o in bpy.context.scene.objects:
    o.select = True
    bpy.context.scene.objects.active = o
    bpy.ops.object.transform_apply(location=True, rotation=False, scale=True)

It also looks like as if initially a blend file with a Text Editor field has to be loaded to make that run.

Thanks again for your time and I hope I was also able to help solving future problems.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.