cPickle fails on manually compiled and executed Python function

[Jan Gosmann]

On 07/18/2017 01:07 AM, dieter wrote:
> "Jan Gosmann" <jan at hyper-world.de> writes:
>
>> [...]
>> fn = load_pyfile('fn.py')['fn']
>> [...]
> "pickle" (and "cpickle") are serializing functions as so called
> "global"s, i.e. as a module reference together with a name.
> This means, they cannot handle functions computed in a module
> (as in your case).
Note that I'm assigning the computed function to a global/module level 
variable. As far as I understand the documentation 
<https://docs.python.org/2/library/pickle.html#what-can-be-pickled-and-unpickled> 
that should be all that matters because only the function name will be 
serialized.
> I am quite convinced that "pickle" will not be able to deserialize (i.e. load)
> your function (even though it appears to perform the serialization
> (i.e. dump).
Actually the deserialization works fine with either module. That is both 
pickle.loads(pickle.dumps(fn)) and cPickle.loads(pickle.dumps(fn)) give 
me back the function.

By now I realized that a pretty simple workaround works. Instead of 
doing `fn = load_pyfile('fn.py')['fn']` the following function 
definition works with both pickle modules:

_fn = load_pyfile('fn.py')['fn']
def fn(*args, **kwargs):
     return _fn(*args, **kwargs)