Pool's map, imap, apply, apply_async, etc fail on Windows

[mmckerns]

>>> import multiprocess as mp
>>> import time
>>> 
>>> def sleepy_squared(x):
…     time.sleep(0.2)
…     return x**2
…
>>> mp.Pool().map(sleepy_squared, range(2))

On Windows, this fails with a NameError -- global name time not defined.

However, it can be avoided by using:
dill.settings['trace'] = True
Results then are as expected.

[mmckerns]

There appear to be other similar (yet different) test failures when dill.settings['recurse'] = True has been set as the dill default -- even on a rebuild of dill and multiprocess.

It looks like the best workaround for the time being is to include the import inside the function whenever possible.

>>> import multiprocess as mp
>>> 
>>> def sleepy_squared(x):
…     import time
…     time.sleep(0.2)
…     return x**2
…
>>> mp.Pool().map(sleepy_squared, range(2))
[0, 1]

[mmckerns]

Note that this works as originally expected on non-windows systems.

Python 2.7.9 (default, Dec 11 2014, 01:21:43) 
[GCC 4.2.1 Compatible Apple Clang 4.1 ((tags/Apple/clang-421.11.66))] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import multiprocess as mp
>>> import time
>>> 
>>> def sleepy_squared(x):
...   time.sleep(0.2)
...   return x**2
... 
>>> mp.Pool().map(sleepy_squared, range(2))
[0, 1]
>>> 
>>> import dill
>>> dill.settings['trace'] = True
>>> 
>>> mp.Pool().map(sleepy_squared, range(2))
[0, 1]

[mmckerns]

closed due to uqfoundation/dill#363