Before, if an object A was parented to an object B, the instance of A would have no parent.
We are fixing that behavior (as it is breaking rendering when used with glTF) and thus now the instance.parent === source.parent
Before, if an object A was parented to an object B, the instance of A would have no parent.
We are fixing that behavior (as it is breaking rendering when used with glTF) and thus now the instance.parent === source.parent
LOL ![]()
Reminds me of the company Accenture back then. They were giving only advices we all already knew about
![]()
Are vertices in the child assumed to be in the local space of the parent or in top-level space?
I think this is equivalent to asking: is it
child.parent = parent;
Or
child.setParent(parent);
The former
In the real world it’s: parent.makeChild(partner)
Should there be an option to keep the old behavior for compatibility?
Unless its
//...
protected makeChild(partner) { /*...*/}
And remember always avoid
//...
public makeChild(partner) { /*...*/}
As I consider this behavior a bug, I would not do that. If we have kept the original behavior by default and propose an option to use the new one, I would have been ok.
But here it is clearly a bug that we are fixing. And it is quite simple to restore previous behavior by setting instance.parent = null