Details
-
Improvement
-
Status: Closed
-
Major
-
Resolution: Fixed
-
None
-
None
Description
While checking why the TAP5-2645 patch had a failing test, I noticed I was wrong about an assumption I made about TypeCoercer: it doesn't favor exact match coercions (i.e. coercions with the exact same source and target classes) over other ones, leaving the possibility of an intermediate (i.e. non-direct) one to be picked up even when an exact match one exists.