[issue23096] Implementation-depended pickling floats with protocol 0

[Alexandre Vassalotti]

Alexandre Vassalotti added the comment:

The repr for floats was changed some time ago to use a shorter decimal strings when possible. Both representations should yield the same float value once decoded. If we want to make the C and Python implementations of pickle consistent, then we should pick on what the Python version does currently: i.e., call repr on the float.

----------

_______________________________________
Python tracker <report at bugs.python.org>
<http://bugs.python.org/issue23096>
_______________________________________