Probability Selector Copy & Paste Bug

[asimParticipant]

I noticed a bug while evaluating Node Canvas.  See description below.  Do you have a fix for this ?

Issue :
Copy and paste (duplicate nodes) does not work correctly on probability selector node.  Repro steps (1) Make a sub tree with probability selector with 2 action children (2) select all nodes and duplicate the sub-tree.  Notice that the copied probability node has 6 weight entries instead of 2. See also attached picture.

[attachment file=”Probability Selector.jpg”]

[asimParticipant]

Is this a known issue ?

[GavalakisKeymaster]

Hello there,

Sorry for the late reply.

This is actually not a known issue and I wonder how I missed that :). Thanks a lot for reporting this bug.

I will take a look at fixing this tomorrow and let you know here with an updated post.

Thanks!

[asimParticipant]

Thank you Gavalakis.
Kindly let me know when you have  fix I could merge.
Before I can “unleash” this package on others in the company, I would to have this resolved. I have had fun using it so far.

[asimParticipant]

Hello Gavalakis,

Has this issue been addressed ?  I have been using your BT and FSM for a while now and this is the only source of pain for me. I would appreciate it if you could take a look at this.

Thanks you,

 

Asim.

 

[GavalakisKeymaster]

Hello,
Here is a quick fix (even though I will more properly handle this in the next version).
Please open up PrioritySelector.cs and change the OnChildConnectedMethod to be like this:

1 2 3 4 5 6 7

public override void OnChildConnected(int index) {             if ( priorities.Count < outConnections.Count ) {                 priorities.Insert(index, new BBParameter<float> { value = 1, bb = graphBlackboard });             }         }

In a similar fashion, please open ProbabilitySelector.cs and change it to this:

1 2 3 4 5 6 7

public override void OnChildConnected(int index) {             if ( childWeights.Count < outConnections.Count ) {                 childWeights.Insert(index, new BBParameter<float> { value = 1, bb = graphBlackboard });             }         }

This handles the issue when duplicating a PrioritySelector along with connected children. If you duplicate a PrioritySelector alone however, you will still have original “weights” and when connecting a child node those will be used first before adding more.

Thanks.

[Author]

Posts