Hi myosis, thanks for reaching us out.
With regard to your request, I confirm that is not possible to script using our Python API the use of the Outline Selection tool.
The options on your end are either to write your own outline selector by making use of the CPolygon, Neighbor and BaseSelect classes or take advantage of the selection scripts already developed by César Vonc.
Best, Riccardo
Hi Daniel, thanks for reaching out us.
With regard to your issue, the question is what values we are passing to the CalculateVisiblePoints. The function is actually designed to receive an array specifying the world position of points in camera space. On top of this the function is supposed to receive a result array initialized to 1.
For your convenience the solution should look like:
// check passed parameter
if (!doc)
return maxon::IllegalArgumentError(MAXON_SOURCE_LOCATION);
// check for an active object
BaseObject* op = doc->GetActiveObject();
if (!op)
return maxon::OK;
// check for current BaseDraw
BaseDraw* activeBD = doc->GetActiveBaseDraw();
if (!activeBD)
return maxon::UnexpectedError(MAXON_SOURCE_LOCATION);
// check it's an instance of a Opolygon
if (!op->IsInstanceOf(Opolygon))
return maxon::OK;
// cast active BaseObject in PolygonObject
PolygonObject* polyobj = static_cast<PolygonObject*>(op);
// get the point count
const Int32 pointCount = polyobj->GetPointCount();
if (pointCount == 0)
return maxon::OK;
// get the active object points
const Vector* points = (Vector*)polyobj->GetPointR();
if (!points)
return maxon::UnexpectedError(MAXON_SOURCE_LOCATION);
// allocate a temporary array of points (same count of active object points)
Vector* pointFromCam = nullptr;
pointFromCam = NewMem(Vector, pointCount) iferr_return;
// calculate the a transformation of the active object as seen from the camera
Matrix mgObjFromCamera = activeBD->GetMi() * polyobj->GetMg();
// fill the temporary array of points with active object points as seen from the camera space
for (Int32 p = 0; p < pointCount; p++)
pointFromCam[p] = mgObjFromCamera * points[p];
// allocate an array to store the result of the visiblity check
UChar *pointVisRes = nullptr;
pointVisRes = NewMem(UChar, pointCount) iferr_return;
// fill the visibility result array with 1
FillMemType(UChar, pointVisRes, pointCount, 1);
// execute the visibility check
Bool res = CalculateVisiblePoints(activeBD, polyobj, pointFromCam, pointVisRes, true);
if (!res)
{
DeleteMem(pointFromCam);
DeleteMem(pointVisRes);
return maxon::UnexpectedError(MAXON_SOURCE_LOCATION);
}
// return the results in a more convenient format
maxon::String resString;
for (Int i = 0; i < polyobj->GetPointCount(); i++)
resString += FormatString(" @"_s, maxon::Int(pointVisRes[i]));
// print the output
ApplicationOutput("visibility array:"_s + resString);
// free the allocated resources
DeleteMem(pointFromCam);
DeleteMem(pointVisRes);
return maxon::OK;
Finally this stuff in Python is:
if doc is None:
return
if op is None:
return
activeBD = doc.GetActiveBaseDraw()
if activeBD is None:
return
visres = c4d.utils.CalculateVisiblePoints(activeBD, op, True)
print visres
Best, Riccardo
Hi Fritz, thanks for reaching out us.
With regard to your issue, as already pointed about by @kbar , please refer to the General Settings section of the ProjectTool Manual. There you'll find the projectdefinition parameter Kent just mentioned in his post.
Best, Riccardo
Hi Daniel,
for the time being I see three options:
Best, Riccardo
@C4DS said in R21 CommandData with option - not working on macOS:
Is a solution being looked at, in order to provide a work around to the known limitation, or provide an alternative way of having the CommandData option available?
If I properly understood your question, for the time being, the workaround should take in consideration that when the command is called from the "Extension" menu (on Cinema uses macOS menu bar as its menu bar) the value of the subid parameter is zero.
Hope it makes sense (but I might have misunderstood your comment)
Cheers, R
Hi fyeng, thanks for reaching out us.
With regard to your question, I confirm that the workflow you're trying to achieve using the python executable found among the Linux CLR modules is not possible since only c4dpy, as you've already seen, has access to all the needed modules.
Best, Riccardo
Hi Daniel, first and foremost I apologize for the delay getting back here. It simply slipped through our daily checking for a few days and we're indeed grateful to you for having reported the focus on it.
With regard to the reported behavior I confirm that it's a limitation we already know and it appears only on macOS when the Cinema 4D top menu bar is embedded in the macOS menu bar.
If Cinema runs with its own top menu bar, the issue doesn't occur.
What instead looks like a bug to me is that when Cinema uses the macOS menu bar, clicking on the command doesn't fire the Execute() method but rather the ExecuteOptionID() with the subid parameter equal to 0. On the opposite when Cinema uses its own top menu bar clicking on the command just fires the Execute() as expected.
I'm going to file a bug report and see how it evolves.
Best, Riccardo
I'm sorry Daniel you're giving it up here, but I dunno how I can be of any further help. I will ask today to one of the other members of the team to replicate the same steps on his system to see if they are consistent with my findings or yours to grant information correctness for those coming in the future and reading the post.
Best, R.
Hi Daniel, thanks for following up.
I've reproduced your cases and I warmly encourage you to test them again since I found a few discrepancies.
<user prefs>/plugins folder (even though unchecked) and from the <whatever>/R20plugins folder still checked in the search path. This behavior is consistent with the finding in my previous post since R20.xxx loads by default anything it finds in the <user prefs>/plugins folder<user prefs>/plugins, the second because of the presence of the <user prefs>/plugins in the search path, the third because of the presence of the <whatever>/R21plugins in the search path<user prefs>/plugins folder (even though unchecked) and from the <user prefs>/plugins folder still checked in the search path. This behavior is consistent with the finding in my previous post since R21.xxx doesn't exclude the <user prefs>/plugins folder if is specified in the search pathFinally with regard to
However, the recommendations for plugin installation is quite a surprise, in particular the second bullet:
"do not install a plugin in the Cinema 4D user's preferences folder".
That is exactly the location where the plugin folder gets created at first launch of the application, so I am confused this location is not recommended.
we think that plugins preferences or user-dependent plugin files might be stored in the <user prefs>/plugins while the actual plugin should be stored in a more convenient location. Last but not least although not planned in the short future, the <user prefs> folder might change in structure and location (usually it depends on OS) and this might cause side effects to plugins stored over there.
Best, Riccardo
Hi Paul_, thanks for reaching out us.
With regard to your issue, we're not aware of any change in the Xpresso in R21 that could lead to this unexpected behavior.
Could you please simplify your setup and provide a testing scene to replicate it here?
Thanks, Riccardo