Hi @Danchyg1337, thanks for reaching out us.

With regard to your question, I'm a bit confused because iso-parametrics are actually lightweight representations of geometrical elements which being governed by parameters can be drawn using curves defined by the driving parameters having the same value.
This usually applies to rational and non-rational surfaces where the s and t parameters can represent s and t iso-parametric curves defined by the locus of points characterized by the same parameter value.

In the case of a generator, this is meaningful because, depending on the math behind the generator, you could take advantage of the iso-params representation but for a modifier I really don't get it. Usually modifiers when applied in a representation where iso-params are shown, automatically deform the isoparametric curves (check for example applying the bend deformer to a sphere represented via iso-params)

Assuming that the code you've posted is running in a BaseObject::ModifyObject() method then BaseObject::SetIsoparm() can not be called from there.
Finally, to my knowledge, deformers like the Bevel Deformer does nothing with regard to iso-params representation since being it applicable only to polygon objects iso-param representation coincides exactly with the edges representation.

Cheers, R

@zipit said in Change a rendering parameter in python xpresso node:

Hi,
I am a bit confused. If you have the plugin id, I do not quite understand what exactly the problem is.

You're right @zipit, nothing to get confused about it was just me taking a different mind path. Your proposed approach is indeed correct and also more effective. Thanks for pointing it out!

Hi @Passion3D, thanks for reaching out us.

With regard to your request, please consider that you can access the currently active document using the doc instance already existing in the Xpresso Python node. From this instance you can retrive the active RenderData using BaseDocument::GetActiveRenderData and from here retrieve the first BaseVideoPost using RenderData::GetFirstVideoPost().
Given the first BaseVideoPost, you can iterate over the videoposts until you find the right one and then change the parameter by accessing by reference its BaseContainer via BaseList2D::GetDataInstance().

The code below attempts to modify the Filter Size of the ProRender videopost given the FilterDriver as Float parameter of the Xpresso Python Node.

import c4d

def main():
    # check doc
    if doc is None:
        return

    # get the active RenderData
    activeRD = doc.GetActiveRenderData()
    if activeRD is None:
        return

    # get the first BaseVideoPost
    firstVP = activeRD.GetFirstVideoPost()

    # check and eventually make the change
    if firstVP.IsInstanceOf(1037639):
        firstVP.GetDataInstance()[c4d.VP_GPURENDERER_FILTERSIZE_OFFLINE] = FilterDriver

Hi @SolarPH, thanks for reaching out us.

With regard to your question, actually initializing the InExcludeData has nothing to do with having the list not including duplicates.
The v parameter pointed out in the documentation is used to fill the to-be-created InExcludeData with the values found in the passed InExcludeData as @zipit has shown in line 48.
To avoid duplication instead you've to manually browse the items in the lists and pickup from the second only those not found in the first. Again @zippit code perfectly shows this.

Cheers, R

Thanks @moGRR for reaching out us.

With regard to your question, as pointed out by @zipit, it's not possible to have a Python Generator to generate object which can be recognized as real spline by generators/deformers expecting a spline as input due to the Python Generator flags used in its registration.

Beside the option mentioned by @zipit I also invite to have a look at this post where it's presented a generator outputting spline and accepting splines as input.

Best, R

Hi Daniel, thanks for reaching out us.

With regard to your question, @Cairyn's observations are indeed correct and there's nothing in our API automatically copying layer upon objects being copied.

Cheers, R

Hi @C4DS, thanks for reaching out us.

I had very little time to better investigate the issue but on top of your behavior, I also spotted different results between mac and win. I'll dig more next week.

Best, R

Kudos to @PluginStudent saying in Best way to fill a maxon::BaseArray:

maxon::BaseArraymaxon::Int testArray;
testArray.Append({ 1,2,3 }) iferr_return;

I confirm that only BaseArray::Append(), BaseArray::Insert() and BaseArray::Insert() can use an initialization list to add data to a BaseArray.

Cheers, R

Hi @C4DS, thanks for reaching out us.

I confirm that the behavior you've reported is a bug and, as anticipated by @zipit, it's due to the fact that when the dialog is opened for the second time, an object then exists and the code follows a different path resulting in a different position of the dialog compared to the mouse position.

You can mitigate the issue by using effective pointer position

	if (GetInputState(BFM_INPUT_MOUSE, BFM_INPUT_MOUSELEFT, bc))
	{
		Int32 xpos = bc.GetInt32(BFM_INPUT_X);
		Int32 ypos = bc.GetInt32(BFM_INPUT_Y);
		
		const Int32 w = 50;
		const Int32 h = 50;
		mDlg.Open(DLG_TYPE::ASYNC_POPUPEDIT, TEST_COMMAND_PLUGIN_ID, xpos, ypos + h, w, h, 0);
	}

Best,

Hi Frank, I confirm that it's a flaw of macOS. Creating a 32-bit grayscale image with an image-editing software and pressing the space-bar on the selected image just ends up with the same results.

Cheers,