You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I spent some time debugging an "inconsistent" behavior in c2Collide.
In c2Collide if you input a C2_TYPE that is not a c2Poly then c2x transform is not used to transform the points of that shape.
This is inconsistent with other functions in the API that use the transform in all shapes, for example in c2Toi (c2GJK one) if you input a C2_TYPE that is not a c2Poly then the c2x transform is applied to that shape.
I think that maybe what can be improved is to treat the points in the shapes as local points and always transform these points in each function in the api. (If the user passes nullptr no transformation is applied ofc)
I know that this is a minor issue but from a users perspective i think it's a bit inconsistent.
Thoughts?
The text was updated successfully, but these errors were encountered:
roig
changed the title
cute_c2.h c2x usage in c2Collide and consistency with the API.
cute_c2.h c2x usage in c2Collide and consistency int the API.
May 6, 2020
We can update c2Collide's behavior, though this is a fairly time-consuming task. It would help to reduce time it takes to implement the changes to know which functions aren't transforming points like you expected. Do you have a list of the ones you've noticed so far?
Hello Randy,
I spent some time debugging an "inconsistent" behavior in c2Collide.
In c2Collide if you input a C2_TYPE that is not a c2Poly then c2x transform is not used to transform the points of that shape.
This is inconsistent with other functions in the API that use the transform in all shapes, for example in c2Toi (c2GJK one) if you input a C2_TYPE that is not a c2Poly then the c2x transform is applied to that shape.
I think that maybe what can be improved is to treat the points in the shapes as local points and always transform these points in each function in the api. (If the user passes nullptr no transformation is applied ofc)
I know that this is a minor issue but from a users perspective i think it's a bit inconsistent.
Thoughts?
The text was updated successfully, but these errors were encountered: