The case here is as follows:
- create random unity sprite prefab
- place sprite's pivot point outside of the sprite (say, pivot U is 2.53 and pivot V is -1.22)
- try to attach this sprite via the SpriteAttacher script
Expected result here would be seeing the sprite sticking with it's pivot point to the placeholder.
Actual result here looks like the sprite's pivot point is somewhere in U: 1 and V: 0.
The bug is caused by Mathf.InverseLerp - it just cannot return values outside of [0; 1] range.
So the solution would be to make our own version of InverseLerp, which handles such situations correctly.
* spine-csharp updated to 3.2.00.
* Some cleanup, catchup and fixes.
* Mix transform constraint scale and shear offsets.
* Update readmes.
* Clean up.
* Formatting.
* Fixed single bone IK with nonuniform scale. Improved two bone IK.
* SkeletonRendererCustomMaterials: correctly reapply changes when slot name changes (#554)
* Made sure private methods are private. Editor uses reflection.
* Also removed public properties.
* Added transform constraint rotate, scale, and shear offsets and mixes.
* Added TransformConstraintTimeline.
* Added bone shearing.
* Added ShearTimeline.
* Changed Skeleton#updateCache.
* Changed JSON and binary formats. Some binary format fields and enums were rearranged for consistency -- sorry, but it's better for the long term. Docs for both are up to date.
http://esotericsoftware.com/spine-json-formathttp://esotericsoftware.com/spine-binary-format
* per-material material overrides, separate MonoBehaviour for setting overrides from inspector
* clean-up and fix SkeletonRendererMaterialOverride
* SkeletonRendererCustomMaterials: initialize SkeletonRenderer to ensure skeletonRenderer.skeleton != null
* rename SkeletonRendererCustomMaterials folder for consistency