After digging through some older threads and testing things, I found that this approach works — based on the discussion here:
https://developers.maxon.net/forum/topic/13555/python-how-to-get-axis-scale-from-the-preference-settings/2

import c4d

doc: c4d.documents.BaseDocument  # The currently active document.
op: c4d.BaseObject | None  # The primary selected object in `doc`. Can be `None`.

def main() -> None:

    # get preferences
    pref = c4d.plugins.FindPlugin(465001632) 
    # Memory ((465001628, 1, 465001632), (888, 133, 465001632))
    desc = c4d.DescID(
        c4d.DescLevel(465001628, 1, 465001632),
        c4d.DescLevel(888, 133, 465001632)
    )
    
    
    memoryPref = pref[desc]
    
    memoryPref[c4d.PREF_MEMORY_PVHARDFOLDER] = "c:\\blubb"
    c4d.EventAdd()
    
    


if __name__ == '__main__':
    main()

To discover all available IDs inside a preference node, you can list them like this:

    for bc, descid, _ in pref.GetDescription(0):
        name = bc[c4d.DESC_NAME]
        
        print(name,descid)

So this topic is solved

Hey everyone,

I’m trying to roll out a few default preference settings across all company workstations, but I can’t find any solid reference or example on how to do it properly through Python.

Basically, I’d like to change things like:

• some Memory settings (Picture Viewer cache, etc.)
• a few File options (Autosave paths and intervals)
• certain Redshift preferences
• and enable the Verify Script check

My goal is to make sure every workstation starts with the same defaults, ideally via a startup script or a deployed config.

Unfortunately, I haven’t found any working example in the documentation or SDK that shows how to access and write these preferences in C4D 2025 / 2026

Does anyone have a current example or best practice for setting preferences programmatically in recent C4D versions?

Thanks a lot!
— Tobias

Could you provide an snipped where you show how to cache it in a plugin?

Works! Thank you!

I have a Python Tag that loads on button press a json with some animation data. Position / Rotation working fine but i still haven't figured out how to create a track on a userdata with the type vector. i think i'm missing the desc ids or something like that

def CreateOrUpdateKey(curve, time, value, interpolation):
    keyDict = curve.FindKey(time)
    if keyDict:
        key = keyDict["key"]
        key.SetValue(curve, value)
        key.SetInterpolation(curve, interpolation)
    else:
        keyDict = curve.AddKey(time)
        if not keyDict:
            raise MemoryError("Failed to create key")
        key = keyDict["key"]
        index = keyDict["nidx"]
        key.SetValue(curve, value)
        curve.SetKeyDefault(doc, index)
        key.SetInterpolation(curve, interpolation)

def LoadJsonToTracks(obj, path):
    if not os.path.isfile(path):
        raise FileNotFoundError(f"JSON file not found: {path}")
    with open(path, "r") as f:
        rawData = json.load(f)

    fps = doc.GetFps()
    interpolation = c4d.CINTERPOLATION_LINEAR

    # --- POSITION TRACKS ---
    pos_curves = []
    for axis in [c4d.VECTOR_X, c4d.VECTOR_Y, c4d.VECTOR_Z]:
        descID = c4d.DescID(
            c4d.DescLevel(c4d.ID_BASEOBJECT_REL_POSITION, c4d.DTYPE_VECTOR, 0),
            c4d.DescLevel(axis, c4d.DTYPE_REAL, 0)
        )
        track = obj.FindCTrack(descID)
        if not track:
            track = c4d.CTrack(obj, descID)
            obj.InsertTrackSorted(track)
        curve = track.GetCurve()
        curve.FlushKeys()
        pos_curves.append(curve)

    # --- ROTATION TRACKS ---
    rot_curves = []
    for axis in [c4d.VECTOR_X, c4d.VECTOR_Y, c4d.VECTOR_Z]:
        descID = c4d.DescID(
            c4d.DescLevel(c4d.ID_BASEOBJECT_REL_ROTATION, c4d.DTYPE_VECTOR, 0),
            c4d.DescLevel(axis, c4d.DTYPE_REAL, 0)
        )
        track = obj.FindCTrack(descID)
        if not track:
            track = c4d.CTrack(obj, descID)
            obj.InsertTrackSorted(track)
        curve = track.GetCurve()
        curve.FlushKeys()
        rot_curves.append(curve)

    # --- Add keyframes from JSON ---
    for entry in rawData:
        frame = entry.get("frame")
        pos = entry.get("position", [0, 0, 0])
        rot = entry.get("rotation", [0, 0, 0])
        if frame is None or len(pos) != 3 or len(rot) != 3:
            continue

        time = c4d.BaseTime(frame, fps)

        for i in range(3):
            CreateOrUpdateKey(pos_curves[i], time, pos[i], interpolation)
            CreateOrUpdateKey(rot_curves[i], time, rot[i], interpolation)

I would also be interested.

I cache all kinds of things, but not the mesh at the end.

#         check if any dependency changed
    def CacheCheck(self, op, doc):


        """

        :param op:
        :param doc:
        :return:
        """

        if op[res.GRIDDER_CACHE] is not None:

            preOption = op[res.GRIDDER_CACHE].GetString(1001)
            preNoise = op[res.GRIDDER_CACHE].GetFloat(1002)
            preField = op[res.GRIDDER_CACHE].GetFloat(1003)
            preFrame = op[res.GRIDDER_CACHE].GetFloat(1004)

            rd = doc.GetActiveRenderData()
            self.rs.SetStartFrame(rd[c4d.RDATA_FRAMEFROM].GetFrame(doc.GetFps()))

            force = op[res.GRIDDER_EVERYFRAME]
            frame = doc.GetTime().GetFrame(doc.GetFps())
            noise, shaderAnimated = self.CacheShader(op, doc)

            if force is 1 and preFrame != frame:
                return False

            if preOption != self.HelperGetOptions(op):
                return False

            if preNoise != noise:
                return False

            field = self.CacheField(op, doc)

            if preField != field:
                return False

            return True

        return False

#-----------------------------------------------------------------
# after everything is calculated:
        bc = c4d.BaseContainer()

        bc.SetData(1001, self.HelperGetOptions(op))
        bc.SetData(1002, self.CacheShader(op, doc)[0])
        bc.SetData(1003, self.CacheField(op, doc))

        if doc != None:
            bc.SetData(1004, frame)

        op[res.GRIDDER_CACHE] = bc

Hey sorry for gravedigging.

i have the problem that my tag plugin recognises the difference between rendering and not rendering. unfortunately it also always changes the state in the doc itself and not only in the doc that is rendered

hey,

found the error. works exactly as i had planned. my code was correct, but i forgot to switch off use fields on the new vertex i mapped.

thanks for the help anyway!

i use a vertext map to manipulate the motion blur in redshift from an object. positive directions are no problem, but the vertext map only seems to store positive values. manipulating it afterwards via python script didn't help either.

which surprises me in particular because the documentation mentions a float as the data type.

in other tools it is easily possible to use negative float values to create the desired effect.

this script breaks an poly object into island, colorized it by random, offset to center of bounding box and some noise. after that i connects all islands again.

its super fast compared to my older "solution"

import c4d
from c4d import gui
import random

def AddVertexColor(s,center,rad):

    cnt = s.GetPointCount()
    tag = c4d.VariableTag(c4d.Tvertexcolor, cnt)


    data = tag.GetDataAddressW()

    bs = s.GetPointS()
    bs.SelectAll(cnt)
    done = []

    points = s.GetAllPoints()


    r = random.random()



    for i in range(cnt):

        g = (points[0]-center).GetLength()    / rad
        b = c4d.utils.noise.Noise(points[0]*0.01,rad)
        c = c4d.Vector(r,g,b)

        c4d.VertexColorTag.SetColor(data, None, None, i, c)


    s.InsertTag(tag)


def main():
    random.seed(666)


    selection  = doc.GetActiveObjects(c4d.GETACTIVEOBJECTFLAGS_NONE)
    for s in selection:
        center = s.GetMp()
        rad    = s.GetRad().GetLength()
        c4d.CallCommand(12298)
        c4d.CallCommand(17891)
        mesh = s.GetDown()

        while mesh:

            AddVertexColor(mesh,center,rad)
            mesh = mesh.GetNext()

        c4d.CallCommand(100004768)
        c4d.CallCommand(16768)
# Execute main()
if __name__=='__main__':
    main()

Hey,

i use

c4d.documents.MergeDocument(doc , f,  c4d.SCENEFILTER_OBJECTS | c4d.SCENEFILTER_MATERIALS | c4d.SCENEFILTER_MERGESCENE)

now i would like to group this new scene under a NULL. if i do the whole thing via obj.InsertUnder(null) i miss the materials.
which is totally logical but unfortunately not what i have in mind.

So one idea would be to insert the materials separately into the scene by checking the tags. The problem is that I can't be 100% sure that all the materials that are needed will be migrated.

Another approach would be to check which objects are already in the scene and then simply move the newly loaded ones under the zero.

So what is the best way to extend the functionality of MergeDocument to be able to specify a null object as root?

I have several object plugins and they all have the same problem - my custom caching functions work as expected until the document is saved and reopened. In that case the generator object returns an empty object. is there any kind of event I could use to force an update?

I got a setup with a stationary camera. i need to keep the object centered in my view and maximize its size but i can't change the rotation / position of my camera.

what i have so far is the following:

i calculate the center of the objekt and change the offset accordingly to center the object. now if i want to zoom in my offset calculations are wrong again.

fovProblem.c4d

edit:

can someone please move the thread into the right subforum?

here is a script i use to color polygon groups. unfortunately a bit slow

import c4d
from c4d import gui
import random

def main():
    random.seed(666)

    c4d.CallCommand(12139) # point mode
    selection  = doc.GetActiveObjects(c4d.GETACTIVEOBJECTFLAGS_NONE)
    for s in selection:

        cnt = s.GetPointCount()
        tag = c4d.VariableTag(c4d.Tvertexcolor, cnt)


        data = tag.GetDataAddressW()

        bs = s.GetPointS()

        done = []

        for i in range(cnt):

            if i in done: continue

            r = random.random()
            g = random.random()
            b = random.random()

            c = c4d.Vector4d(r,g,b,1)

            bs.DeselectAll()
            bs.Select(i)

            c4d.CallCommand(12557)

            sel = bs.GetAll(cnt)

            for index, selected in enumerate(sel):
                if not selected: continue

                done.append(index)
                c4d.VertexColorTag.SetColor(data, None, None, index, c)

            done = list(set(done))

        s.InsertTag(tag)

# Execute main()
if __name__=='__main__':
    main()

forget about it. i removed the c4d.OBJECT_INPUT flag during development and forgot to add it back.

@m_magalhaes said in HOW TO REMOVE GENERATOR CHILDS WHEN CONVERTING VIA "C":

Something you can do is touching the objects either with a dependency list or directly.

Ah - i thought that touch only works with GVO. Never tried it to be honest. This solve at least the visibilty problem.

Hey,

i have built a spline gernerator that modifies a spline, when i convert the object i get my modified spline but also the hirachie remains. however i only need the new spline object.

Since GetContour() unfortunately doesn't support HirachyHelper I don't know how to hide the original objects and how to prevent the original splines from being present after the conversion.

https://gumroad.com/l/spliner

<3<3<3<3

The script change the compression method to zip not to Zips or what i expected: RLE because rle was the given value.

I need to change my exr compression to one scanline at a time but beside using a c4d default document is there any other way to change it directly in a script?