Thanks for @ferdinand awesome answer!
I found DescLevel and DescId always confused to me, but them do can access more than simple set item, time to dive deeper to the DescLevel part

Cheers~
DunHou

Hey community,

This my first dive into graphview node, sorry if this is a stupid basic question.

I want to set condition node input port (GvPort) value, but it seems I can not do this via __settitem__ / SetParameter, is this is a limit or I do it in wrong way?

Cheers~
DunHou

import c4d

def iterateNodes(node):
    while node:
        description = node.GetDescription(c4d.DESCFLAGS_DESC_0)
        data: c4d.BaseContainer
        pid: c4d.DescID
        for data, pid, _ in description:
            value: any | None = None
            try:
                value = node[pid]
            except:
                value = "Inaccessible in Python"
            
            if str(data[c4d.DESC_NAME]).startswith("Input"):
                print(f"\tInput '{data[c4d.DESC_NAME]}'(c4d.{data[c4d.DESC_IDENT]}) = {value} type: {type(value)}")
                
        
        node[2000, 1001] = 2 # TypeError: __setitem__ got unexpected type 'int'.
        node.SetParameter([2000, 1002],2,0) # no error but not worrk.
        print(node[2000, 1001].GetDataInstance()) # AttributeError: parameter access failed
        
        if node.IsGroupNode():
            iterateNodes(node.GetDown())

        node = node.GetNext()

def main():
    # Checks if selected object is valid
    if op is None:
        raise ValueError("op is none, please select one object.")

    # Retrieves the xpresso Tag
    xpressoTag = op.GetTag(c4d.Texpresso)
    if xpressoTag is None:
        raise ValueError("Make sure the selected object get an Xpresso Tag.")

    # Retrieves the node master
    gvNodeMaster = xpressoTag.GetNodeMaster()
    if gvNodeMaster is None:
        raise RuntimeError("Failed to retrieve the Node Master.")

    # Retrieves the Root node (the Main XGroup) that holds all others nodes
    gvRoot = gvNodeMaster.GetRoot()
    if gvRoot is None:
        raise RuntimeError("Failed to retrieve the Root Node.")

    # Iterates overs all nodes of this root node.
    iterateNodes(gvRoot)


if __name__ == '__main__':
    main()

I think you need to override RenderEngineCheck to false, the document said:

Bool MyRenderer::RenderEngineCheck(const BaseVideoPost* node, Int32 id) const
{
  switch (id)
  {
    case RENDERSETTING_STATICTAB_MULTIPASS:
    case RENDERSETTING_STATICTAB_ANTIALIASING:
    case RENDERSETTING_STATICTAB_OPTIONS:
    case RENDERSETTING_STATICTAB_STEREO:
      return false;
  }
 
  return true;
}

Cheers~
DunHou

@ferdinand love the new mxutils functions, Great work!

btw, python SDK change notes seems link to C++ docs.

Cheers~
DunHou

Hey @ferdinand , thanks for your information, Iterate graph is one of the solutions to listen to name change, but what I want is the "state" of any text input.

As I had posted on the beta forums (Typing software issue with Cinema 4D.), if someone used non-Latin languages, like Chinese for example, If you use a typing software like IME( Mircosoft Pinyin ), it should ship an English input. When we want to change something like an object name, it will set the typing to English mode. It is very annoying when you have tons of objects to rename one by one.

So I wonder if I can get the "typing mode" so I can set the IME always in Chinese mode when I input a text, and keep in English when not typing text to use shortcuts.

I am a very stupid beginner in C++, but you know what I mean. I try to use GetGUIThreadInfo and GUI_CARETBLINKING to get the text cursor, but it seems not work.

Hope it is clear.

Cheers~
DunHou

Hey community.

I wonder if we can get the text input event in Cinema like this. That is to say, when pressing the keyboard, it is the action of inputting text.

Another corresponding state is the shortcut key state, such as pressing C to collapse a certain object.

How can I get this state via Python or in C++?

Cheers~
DunHou

Thanks @ferdinand for your great solution!

Cheers~
DunHou

Hey @ferdinand ,

thanks for details.

Cheers~
DunHou

---

About forum, I can get notification about my own topic, And I clear all the cache, but nothing changed.

After I tried to reset my account settings manually, then it worked as expected.

Hey @ferdinand ,

Thanks for your answer, unfortunately, the answer is the same as I thought, and I am powerless to do anything about it. I can only wait for the exposed parameters.

Based on your suggestion, there may be many issues, and it is not worth spending a lot of time debugging them. Fortunately, this is not a very urgent task.

Cheers~
DunHou

Hey @ferdinand,

Thanks for your tips! Disabled validateAbsolutePaths with lazy paths can also work in this case, but if you interested, here is the console without disabled:

I know third party APIs are not supported, just want to know if it is possible due to Arnold. It looks like it is due to my poor knowledge

And as I know, Arnold has demo plugin that only has watermark on render, I use the demo to create plugins and it it worked as same as the full version.

Cheers~
DunHou

Traceback (most recent call last):
  File "scriptmanager", line 19, in <module>
  File "C:\Program Files\Maxon Cinema 4D 2025\resource\modules\python\libs\python311\maxon\frameworks\nodes.py", line 548, in ApplyDescription
    res: DataDictionary = GraphDescription._ApplyDescription(
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Program Files\Maxon Cinema 4D 2025\resource\modules\python\libs\python311\maxon\decorators.py", line 495, in Auto
    ExecStaticMethod(*args)
RuntimeError: The node attribute reference '>output.r' (space: 'com.autodesk.arnold.nodespace', lang: 'en-US', node: 'com.autodesk.arnold.shader.image') is not associated with any IDs. [graphdescription_impl.cpp(2003)]
In:
{ <- Error in this scope ->
   $type: #~.image
   #<filename: asset:///file_2e316c303b15a330
   #<multiply -> #>output.r: {

     $type: #~.image
     #<filename: asset:///file_2e316c303b15a330
  }
} [graphdescription_impl.cpp(203)]

---

I test on a brand new Macbook, and refresh my PC browser, and try to use private mode. None of them worked, but the notification of my question does show in the Ring icon, but the sort does not work. not sure about UNREAD, but I do miss the recent topic.

Hey community,

I want to get the explicit render engine name of the default import plugin but not the index, can I do this?

e.g. I want to set the importor to Standard or Redshift, but if I remove some renderer like CentiLeo, the index of Standard will changed.

Cheers~
DunHou

Hey @ferdinand ，

I notice that GraphDescription also have two undo steps even I create graph and insert before apply description, how can I avoid this.

the description dict is steal from GraphDescription Manual

Cheers~
DunHou

import c4d
import maxon

data = {
    "$type": "Output",
    "Surface -> outColor": {
        "$type": "Standard Material",
        "Base/Metalness": 1.,
        "Base/Color": {
            "$type": "Texture",
            "Image/Filename/Path": maxon.Url("asset:///file_eb62aff065c50a2b"),
            "Scale": {
                "$type": "Value",
                "$id": "texTransform",
                "Input": maxon.Color(2, 2, 1)
            }
        },
        "Reflection/Roughness": {
            "$type": "Texture",
            "Image/Filename/Path": maxon.Url("asset:///file_8bf49bb0b9992dbe"),
            "Scale": "#texTransform"
        },
        "Geometry/Bump Map": {
            "$type": "Bump Map",
            "Input Map Type": 0,
            "Height Scale": .2,
            "Input": {
                "$type": "Texture",
                "Image/Filename/Path": maxon.Url("asset:///file_e706df60a3a533d2"),
                "Scale": "#texTransform"
            }
        }
    }
}


def main() -> None:
    """Called by Cinema 4D when the script is being executed.
    """
    doc = c4d.documents.GetActiveDocument()
    doc.StartUndo()
    material = c4d.BaseMaterial(c4d.Mmaterial)
    node_space = maxon.Id("com.redshift3d.redshift4c4d.class.nodespace")
    graph: maxon.NodesGraphModelRef = maxon.GraphDescription.GetGraph(material, node_space, True)
    doc.InsertMaterial(material)
    doc.AddUndo(c4d.UNDOTYPE_NEWOBJ, material)
    
    maxon.GraphDescription.ApplyDescription(graph, data)
    doc.EndUndo()
if __name__ == '__main__':
    main()

Hey community,

I am working on something with GraphDescription, but seems I can not use explicit connection with Arnold, but Redshift and VRay worked as I want.
I know 3rd is not full supported, but is this a problem with SDK design or Arnold's lack of adaptation, should I ask to Arnold dev?
I tried match label/id/lazy id, but none of them work.

btw: The notification and sorting of the forum have not been fixed yet. I have also tested it on a new computer and browser

Cheers~
DunHou

this is want I want:

Code:

import maxon

desc = {
    "$type": "#~.material",
    "#<shader":
        {
            "$type": "#~.standard_surface",
            "#<base_color": {
                "$type": "#~.image",
                "#<filename": maxon.Url("asset:///file_2e316c303b15a330"),
                "#<multiply -> #>output.r": {
                    "$type": "#~.image",
                    "#<filename": maxon.Url("asset:///file_2e316c303b15a330")
                }
            }
        }
}

maxon.GraphDescription.ApplyDescription(
    maxon.GraphDescription.GetGraph(name="test"), desc)

Hi community,

I notice that I can not get the newest and unread topic, when I open the form the sort seems not worked as before, I need to changed it manually to NEWEST TO OLDEST. Has anyone else encountered a similar situation？

Cheers~
DunHou

Maybe you can try Google Chrome browser. The auto-fill doesn't work well in Edge sometimes, even manually passing your info. I can not login old plugin cafe with Edge, but it can work with Chrome, @ferdinand reminds me that in the past, I think it is good to mention here.

Cheers~
DunHou

Hey @ferdinand,

Sorry to interject here, I'm just curious why it's recommended to put the return type at the end (of course, I prefer this Python-like writing style), but as far as I know, this is a relatively old-fashioned writing style (c++11), and the code I observed during my learning process was all of the predecessor types, this is quite confusing for a C++ beginner like me.

Cheers~
DunHou

Result<GraphNode> getConnectedNode(const GraphNode& input, const NODE_KIND kind);```

Hey @m_adam ,

Yes , I had a look at this example, but I'm confusing to understand how it can apply to my case, sorry for my foolish brain.

If I keep InputEvent as @ferdinand did in the example, HandleMouseDrag will return True if I have a little move, I try to stop this,
so I want to execute HandleDragEvent only when mouse "obvious interaction outside the ua"，to avoid the accompanying effect of dragging when lightly clicked (download) .

    def InputEvent(self, msg: c4d.BaseContainer) -> bool:
        """Called by Cinema 4D when the user area receives input events.

        Here we implement creating drag events when the user drags from this user area. The type of
        drag event which is initiated is determined by the drag type selected in the combo box
        of the dialog.
        """
        # When this is not a left mouse button event on this user area, we just get out without
        # consuming the event (by returning False).
        if (msg.GetInt32(c4d.BFM_INPUT_DEVICE) != c4d.BFM_INPUT_MOUSE or
                msg.GetInt32(c4d.BFM_INPUT_CHANNEL) != c4d.BFM_INPUT_MOUSELEFT):
            return False

        # Get the type of drag event that should be generated, and handle it.
        dragType: int = self._host.GetInt32(self._host.ID_DRAG_TYPE)
        return self.HandleDragEvent(msg, dragType)

my code did stop the HandleDragEvent execute if I didn't want to.

but in this situation, even if I print and execute HandleDragEvent successfully, C4D does not accept drag events, which means the material cannot be assigned to the object. I don't know what is preventing this.

    def InputEvent(self, msg: c4d.BaseContainer) -> bool:
        """Called by Cinema 4D when the user area receives input events.

        Here we implement creating drag events when the user drags from this user area. The type of
        drag event which is initiated is determined by the drag type selected in the combo box
        of the dialog.
        """
        # When this is not a left mouse button event on this user area, we just get out without
        # consuming the event (by returning False).
        if msg[c4d.BFM_INPUT_DEVICE] != c4d.BFM_INPUT_MOUSE and msg[c4d.BFM_INPUT_CHANNEL] != c4d.BFM_INPUT_MOUSELEFT:
            return False
        
        dragType: int = self._host.GetInt32(self._host.ID_DRAG_TYPE)
        
        if msg.GetBool(c4d.BFM_INPUT_DOUBLECLICK):
            print("Double click detected, generating drag event")
            return True
   
        mx = int(msg[c4d.BFM_INPUT_X])
        my = int(msg[c4d.BFM_INPUT_Y])
        mx -= self.Local2Global()["x"]
        my -= self.Local2Global()["y"]
        # print(f"Start mouse: {mx}, {my}")

        state = c4d.BaseContainer()

        self.MouseDragStart(c4d.BFM_INPUT_MOUSELEFT,mx,my,c4d.MOUSEDRAGFLAGS_DONTHIDEMOUSE|c4d.MOUSEDRAGFLAGS_NOMOVE)
        isFirstTick = True
        s = 0
        dua = 0
        
        while True:
            res, dx, dy, channels = self.MouseDrag()
            if res != c4d.MOUSEDRAGRESULT_CONTINUE:
                break  

            mx -= dx
            my -= dy
            
            self.GetInputState(c4d.BFM_INPUT_MOUSE, c4d.BFM_INPUT_MOUSELEFT, state)
            
            # mouse released, this can triggered by a click or a drag end
            if state[c4d.BFM_INPUT_VALUE] == 0:
                dua = time.perf_counter() - s
                print (f"Released Mouse in {dua:.4f} seconds")
                
                # drag too short, not starting drag event
                if dua < 0.15:
                    return False
                break                

            if isFirstTick:
                isFirstTick = False
                s = time.perf_counter()
                print(f"\t-- first click : {mx}, {my}")
                continue

        endState = self.MouseDragEnd()
        if endState == c4d.MOUSEDRAGRESULT_FINISHED:
            print(f"\t-- drag finished : {mx}, {my}")
            
            # drag end inside ua, not generating drag event
            if (0 <= mx <= 0 + self.width and 0 <= my <= 0 + self.height): 
                return False

        print('Now ,try to start drag event')        
        return self.HandleDragEvent(msg, dragType)

Cheers~
DunHou

Hey @ferdinand ,

Sorry for silence, quite busy last days, anyway, back to the question.

It should be noted that from the results, I have found an alternative way to implement it as mentioned above. I just want to review the diagram to understand the logic of the input event
I'm not questioning the rationality of this setting, I just want to know if there's a way to control it more finely

A more precise question may be control over clicking and dragging, e.g. :

• Assume Cinema consider a click down and release under 0.1sencond and mouse position didn't change more than 2 pixels as a BFM_INPUT_MOUSELEFT, and hold on more than 0.1s and mouse move as HandleMouseDrag.
• Under this premise, if the user accidentally touches and causes the mouse to quickly click and the cursor to move some distance, this behavior will be considered as dragging.
• I hope this behavior is seen as a click （I know this is a bit strange, I just want to try and see if it can be achieved）
• I tried to use c4d.BFM-INPUT_LUE to retrieve mouse release, but as far as I know, this needs to be executed in a while loop.
• When there is a loop in InputEvent, the conditions I set are judged correctly, but the drag time is masked, which means that the drag behavior cannot be triggered(e.g., no plus under the cursor, and no material created).

like our asset browser, click or drag between the green arrow will not trigger the download progress. only click download/double click/drag outside will download and assign the asset.

Cheers~
DunHou

Hey @ferdinand,

Is there any deficiency in pathlib ? I just think its writing style is more convenient and it produces less code. Lazy me

@ferdinand said in ImportSymbols with single file didn't return as expect.:

does not support passing a file as an input but only directories

I'm a bit confused, does this mean that ImportSymbols only support directories, but the documentation says that files can be passed for parsing, or did I misunderstand again...

Cheers~
DunHou

Hey @ferdinand ,

An extension issue, which is actually the original intention of this issue, I followed your suggestion and implemented "user seamless interaction" using a timer, but there is still one issue that troubles me.

How to exclude mouse clicks before dragging event.

• The reason I need to do this is that before clicking, there may not be any data on the hard drive to generate the material. After dragging and dropping, I want to perform a data check, but I don't want to trigger a download task every time I click
• I tried using MouseDragStart polling to determine the position of the mouse when it is released, in order to determine whether the action should be considered a click, but at this point, the print ('try to start drag event ') did trigger, but there was no HandleMouseDrag action, and there was no interaction when the mouse moved over the object.
• I followed the suggestion here and tried using GetInputState to determine mouse release, but I couldn't get rid of this limitation

Uncertain alternative methods require testing

• Generate empty material, if it is a valid dragging, check data and replace materials after successful execution, which may be feasible but requires a lot of additional code.
• Generate an empty material, but download the address in the texture path after success, and then refresh the material.

These methods certainly cannot separate clicking and dragging intuitively. Is it possible to achieve this?

get idea from https://developers.maxon.net/forum/topic/16267/marquee-selection-of-items-in-geuserarea

Cheers~
DunHou

    def InputEvent(self, msg: c4d.BaseContainer) -> bool:

        
        mx = int(msg[c4d.BFM_INPUT_X])
        my = int(msg[c4d.BFM_INPUT_Y])
        gx,gy = mx,my
        mx -= self.Local2Global()["x"]
        my -= self.Local2Global()["y"]

        channel = msg[c4d.BFM_INPUT_CHANNEL]
        state = c4d.BaseContainer()
        mousex = mx
        mousey = my

        if channel == c4d.BFM_INPUT_MOUSELEFT:
            # res, dx, dy, channels = self.MouseDrag()
            # print(res, dx, dy, channels)
            self.MouseDragStart(
                c4d.BFM_INPUT_MOUSELEFT,
                mx,
                my,
                c4d.MOUSEDRAGFLAGS_DONTHIDEMOUSE,
            )
            while True:
                res, dx, dy, channels = self.MouseDrag()
                
                if res == c4d.MOUSEDRAGRESULT_ESCAPE:
                    self._drag_enabled = False
                    break                
                elif res == c4d.MOUSEDRAGRESULT_CONTINUE:
                    mx -= dx
                    my -= dy
                    
                if not self.GetInputState(c4d.BFM_INPUT_MOUSE, c4d.BFM_INPUT_MOUSELEFT, state):
                    break
                
                if state[c4d.BFM_INPUT_VALUE] == 0:
                    print ("Released Left Mouse")
                    break

                if dx == 0 and dy == 0:
                    continue
            
                mousex += dx
                mousey += dy
                print("Mouse Dragging at position [%f,%f]" % (mousex, mousey))
                
                if res == c4d.MOUSEDRAGRESULT_FINISHED:
                    print("dra finished")
                    # print(mx,my)
                    # print(f"inside: {self.is_inside(mx, my)}")
                    if self.is_inside(mx, my):
                        print("end inside")
                        if min(abs(gx-mx), abs(gy-my)) < 20:
                            print("move little, treat as click")
                            self._drag_enabled = False
                            return True
                        break
                    else:
                        self._drag_enabled = True
                        print("end outside, drag enabled")
                        break

            self.MouseDragEnd()

        if self._drag_enabled:
            print('try to start drag event')
            dragType: int = self._host.GetInt32(self._host.ID_DRAG_TYPE)
            return self.HandleDragEvent(msg, dragType)
        return True