@miguelinux314: Thanks for catching this. In theory, __file__ could be set to anything, right? So, to mitigate a TypeError, would it be better to either type-check (i.e. with type) or to catch a TypeError? If None is the most common case, then you could check that as you have now first...

Thank you @mmckerns for the quick reply.

Yes, the __file__ attribute of any module (built-in or not) can be manually set to anything.

On the other hand, under certain conditions (see https://github.com/python/cpython/blob/423459be2f0b6d007e5f235f39d80044cb099faf/Lib/importlib/_bootstrap.py#L667, caused by https://bugs.python.org/issue32305) importlib does set it to None without complaining or manually fidgeting with the __file__ attribute.

Based on this, I don't think one can get the __file__ attribute to differ both from a valid path string and None without actively trying to break the code. That's why I checked only for None and left other cases to crash. However, if you feel the dill library would be sturdier considering the manual fiddling cases, my recommendation would be a try-catch of lines 1620 to 1629 with TypeError. I'd be happy with either solution.

Thanks again for you time and help

I'll take it as is. thanks