Not sure if it is related to your crash, but I have never seen someone using isinstance() this way.
In the C4D API, there is IsInstanceOf() to check for inheritance or type. See also Type.
Solved Still not possible to find out what's visible in the Timeline?
@PluginStudent isinstance is the plain Python class comparison. IsInstanceOf() does not work here because it compares against type constants, and there is (AFAIK) no such constant for BaseDocument or BaseList2D or GeListHead.
EDIT: Oh, I'm stupid. I just saw that such a constant is used in the original example... okay, let's try that...
@PluginStudent No, didn't work with IsInstanceOf either.
For completeness: Tbasedocument and Tbaselist2d exist but are only documented in the C++ manual; missing in Python. Tgelisthead does not exist in either, so I still need to check the class against that with the Pythonic isinstance.
Hi Just to let you know its seems there is currently a bug somewhere (still have to figure out) but to avoid any issue check if the returned GeListHead and then if you iterate over all its children that each c4dAtom returned is alive with C4DAtom.IsAlive before doing anything on it(e.g.s print them).
That should fix your issue.
Cheers,
Maxime.
MAXON SDK Specialist
@m_adam Thank you, unfortunately that didn't work either... I check the object, the list head, and the list nodes all for being alive, and I do not get any output that they aren't. The crashes continue.
I have compiled the SDK, and the "Active Object Properties" dialog works fine even in places where the Python script hangs. So, it might be a Python-only issue.
So far, I think that these places are when the script encounters a "Tags" object (type 200001048). The first place I saw this was under ProjektHooks/PLKHUD/PSUNDOHEAD/PKHOP. When I skipped that branch explicitly, the next place of crash was under Modeling Objects Hooks/Modeling Objects Branch/Plane Manipulator and Pivot Manipulator (both branches contain a "Tags" element). Didn't continue after that.
However, trying to exclude these node types didn't help either. Apparently, the objects are so volatile that any way of touching them creates the crash already.
(Since the scene tags also hide under these nodes, I do not want to exclude them, anyway.)
Here's the code I tried last:
import c4d
from c4d import gui
emCount = 0
handledList = []
def showBranchInfo(depth, obj, f):
global emCount, handledList, maxCount, dialogCount
try:
# Validity check
if obj == None: return
if obj in handledList:
print "Already handled! : ", obj
print >>f, "Already handled! : ", obj
f.flush()
return
else:
handledList.append(obj)
if not obj.IsAlive():
print "Object not alive!"
print >>f, "Object not alive!"
f.flush()
return
if obj.GetType() == 200001048:
print "Object is 200001048!"
print >>f, "Object is 200001048!"
f.flush()
return
except:
print >>f, "ERROR in validity check"
f.flush()
return
try:
# Object output
emCount += 1
print " " * depth, "Count=", emCount, "(object)"
print >>f, " " * depth, "Count=", emCount, "(object)"
f.flush()
if isinstance(obj, c4d.documents.BaseDocument):
print " " * depth, "DOC:", obj.GetDocumentName()
print >>f, " " * depth, "DOC:", obj.GetDocumentName()
f.flush()
# this is also a BaseList2D so we can continue
elif not isinstance(obj, c4d.BaseList2D):
print " " * depth, "Not a BaseList2D: ", obj
print >>f, " " * depth, "Not a BaseList2D:", obj
f.flush()
return
else:
print " " * depth, "OBJ:", obj.GetName()
print >>f, " " * depth, "OBJ:", obj.GetName()
f.flush()
except:
print >>f, "ERROR in object output"
f.flush()
return
try:
child = obj.GetDown()
if child != None:
while child:
showBranchInfo(depth + 1, child, f)
child = child.GetNext()
except:
print >>f, "ERROR in subobject recursion"
f.flush()
return
try:
bis = obj.GetBranchInfo()
if bis == None:
print "ERROR: no branch info retrieved"
print >>f, "ERROR: no branch info retrieved"
f.flush()
return
# empty branch info is allowed, so no len() check!
except:
print >>f, "ERROR in GetBranchInfo"
f.flush()
return
try:
for bi in bis: # list of dict{‘head’, ‘name’, ‘id’, ‘flags’}
try:
if len(bi) == 0:
print "ERROR: no entries in dict list"
print >>f, "ERROR: no entries in dict list"
f.flush()
return
listHead = bi['head']
nodeName = bi['name']
# print bi['id']
# print bi['flags']
except:
print >>f, "ERROR in dict list check"
f.flush()
return
try:
# List head output
emCount += 1
node = None
print " " * (depth+1), "Count=", emCount, "(list head)", nodeName
print >>f, " " * (depth+1), "Count=", emCount, "(list head)", nodeName
f.flush()
# Crash irgendwo hier
if listHead == None:
print " " * (depth+1), "No list head"
print >>f, " " * (depth+1), "No list head"
f.flush()
elif not listHead.IsAlive():
print "List head not alive!"
print >>f, "List head not alive!"
f.flush()
elif not isinstance(listHead, c4d.GeListHead):
print " " * (depth+1), "Not a GeListHead:", listHead
print >>f, " " * (depth+1), "Not a GeListHead:", listHead
f.flush()
else:
print " " * (depth+1), nodeName, listHead
print >>f, " " * (depth+1), nodeName, listHead
f.flush()
node = listHead.GetFirst()
except:
print >>f, "ERROR in list head output"
f.flush()
return
try:
# Content of sublist branch
locCount = 0
if node == None:
print " " * (depth+2), "Empty list"
print >>f, " " * (depth+2), "Empty list"
f.flush()
while node != None:
if node.IsAlive():
if node.GetType() != 200001048:
emCount += 1
print " " * (depth+2), "El:", locCount, node
print >>f, " " * (depth+2), "El:", locCount, node
f.flush()
locCount += 1
if isinstance(node, c4d.BaseList2D):
showBranchInfo(depth + 3, node, f)
# no else case; node was already printed
else:
print "List node is 200001048!"
print >>f, "List node is 200001048!"
f.flush()
return
else:
print "List node not alive!"
print >>f, "List node not alive!"
f.flush()
node = node.GetNext()
except:
print >>f, "ERROR in sublist branch handling"
f.flush()
return
# missing here: recurse into children of "node" or "obj"
except:
print >>f, "ERROR in dictionary loop"
f.flush()
return
try:
print >>f, " " * depth, "Returning from depth:", depth
f.flush()
except:
print >>f, "ERROR in return"
f.flush()
return
def main():
print " "
print "GetBranchInfo"
print "============="
f = open("L:/getbranchinfo.txt", "at")
showBranchInfo(1, doc, f)
f.close()
if __name__=='__main__':
main()
You can happily ignore the try/except blocks, as they never caught any error.
Considering that the code now consists of 80% error checking, this is a very frustrating exercise.
Interestingly, my initial example that uses GetCTracks() works fine (for what it does) and does not crash, even when handling (scene) tags. I suspect the issue ultimately lies in the GetBranchInfo() function somewhere.
creating a cube and call op.GetBranchInfo is enough to trigger a crash, since its a C++ crash, there is no room for python to catch it.
MAXON SDK Specialist
@m_adam oops... didn't expect that, since the Active Object Properties sample works fine (solely in C++).
Okay, it's definitely the Tags node. I added some code to check for the name "Tags" in the list returned by GetBranchInfo() to skip that dictionary entry before even the list head is touched. That works, the recursion/iteration is running through without crash.
Of course, that makes the scene tags disappear from the list, so I also added in an else case that explicitly calls GetTags() on the original object, and continues the recursion with any returned tags.
For testing, I also added a collector list for the tracks (which were the original goal of this exercise...) and print the tracks at the end of the code. For a few test scenes, this apparently works; I have no idea about the overall soundness and completeness of this workaround though.
Here's the code for anyone who might face the same issue:
import c4d
from c4d import gui
emCount = 0
handledList = []
tracks = []
def showBranchInfo(depth, obj, f):
global emCount, handledList, maxCount, dialogCount, tracks
# Validity check
if obj == None: return
if not obj.IsAlive():
print "Object not alive!"
print >>f, "Object not alive!"
f.flush()
return
if obj.GetType() == 200001048:
print "Object is 200001048!"
print >>f, "Object is 200001048!"
f.flush()
return
if obj in handledList:
print "Already handled! : ", obj
print >>f, "Already handled! : ", obj
f.flush()
return
else:
handledList.append(obj)
# Object output
emCount += 1
print " " * depth, "Count=", emCount, "(object)"
print >>f, " " * depth, "Count=", emCount, "(object)"
f.flush()
if isinstance(obj, c4d.documents.BaseDocument):
print " " * depth, "DOC:", obj.GetDocumentName()
print >>f, " " * depth, "DOC:", obj.GetDocumentName()
f.flush()
# this is also a BaseList2D so we can continue
elif not isinstance(obj, c4d.BaseList2D):
print " " * depth, "Not a BaseList2D: ", obj
print >>f, " " * depth, "Not a BaseList2D:", obj
f.flush()
return
else:
print " " * depth, "OBJ:", obj.GetName()
print >>f, " " * depth, "OBJ:", obj.GetName()
f.flush()
# assemble the tracks of the object
addTracks = obj.GetCTracks() # returns a list
tracks.extend(addTracks)
# go through the children of the object if present
child = obj.GetDown()
if child != None:
while child:
showBranchInfo(depth + 1, child, f)
child = child.GetNext()
# check the branch info and go into all dependent branches
bis = obj.GetBranchInfo()
if bis == None or len(bis) == 0:
print " " * (depth+1), "No branch info retrieved, or empty"
print >>f, " " * (depth+1), "No branch info retrieved, or empty"
f.flush()
return
for bi in bis: # list of dict{‘head’, ‘name’, ‘id’, ‘flags’}
if len(bi) == 0:
print "ERROR: no entries in dict list"
print >>f, "ERROR: no entries in dict list"
f.flush()
return
nodeName = bi['name']
if nodeName != "Tags":
listHead = bi['head']
# print bi['id']
# print bi['flags']
# List head output
emCount += 1
node = None
print " " * (depth+1), "Count=", emCount, "(list head)", nodeName
print >>f, " " * (depth+1), "Count=", emCount, "(list head)", nodeName
f.flush()
# Crash irgendwo hier
if listHead == None:
print " " * (depth+1), "No list head"
print >>f, " " * (depth+1), "No list head"
f.flush()
elif not listHead.IsAlive():
print "List head not alive!"
print >>f, "List head not alive!"
f.flush()
elif not isinstance(listHead, c4d.GeListHead):
print " " * (depth+1), "Not a GeListHead:", listHead
print >>f, " " * (depth+1), "Not a GeListHead:", listHead
f.flush()
else:
print " " * (depth+1), nodeName, listHead
print >>f, " " * (depth+1), nodeName, listHead
f.flush()
node = listHead.GetFirst()
# Content of sublist branch
locCount = 0
if node == None:
print " " * (depth+2), "Empty list"
print >>f, " " * (depth+2), "Empty list"
f.flush()
while node != None:
if node.IsAlive():
if node.GetType() != 200001048:
emCount += 1
print " " * (depth+2), "El:", locCount, node
print >>f, " " * (depth+2), "El:", locCount, node
f.flush()
locCount += 1
if isinstance(node, c4d.BaseList2D):
showBranchInfo(depth + 3, node, f)
# no else case; node was already printed
else:
print "List node is 200001048!"
print >>f, "List node is 200001048!"
f.flush()
return
else:
print "List node not alive!"
print >>f, "List node not alive!"
f.flush()
node = node.GetNext()
else:
print "Tags skipped!"
print >>f, "Tags skipped!"
f.flush()
if isinstance(obj,c4d.BaseObject):
taglist = obj.GetTags()
print " " * (depth+2), "Tag list, length=", len(taglist)
print >>f, " " * (depth+2), "Tag list, length=", len(taglist)
f.flush()
for tag in taglist:
showBranchInfo(depth + 3, tag, f)
# missing here: recurse into children of "node" or "obj"
print >>f, " " * depth, "Returning from depth:", depth
f.flush()
def main():
global tracks
print " "
print "GetBranchInfo"
print "============="
f = open("L:/getbranchinfo.txt", "at")
showBranchInfo(1, doc, f)
for track in tracks:
print track
print >>f, track
f.close()
if __name__=='__main__':
main()
(also, unsolicited advertising because I really spend too much time with this)
https://www.patreon.com/cairyn
there you go!
Hi, I found the issue, unfortunately, nothing on your side you can do to fix it.
I marked this topic as solved, and will bump the topic once a fix is included in a public release.
Keep in mind this is a Python only issue.
Cheers,
Maxime.
MAXON SDK Specialist
@m_adam okay, good that it's found... I guess I can go with the code above for now that excludes tags explicitly and gathers them in a different recursion.
As per the original issue (there is no function in the API that returns the currently visible tracks in timeline n), I guess I would need to make a requirement/suggestion on Maxon's site? Or is something like this planned already?