Close #10128: don't rerun __main__.py in multiprocessing
- backports issue #10845's mitigation of incompatibilities between
the multiprocessing module and directory and zipfile execution
- Multiprocessing on Windows will now automatically skip rerunning top
level __main__.py modules in spawned processes, rather than failing
with AssertionError
diff --git a/Lib/multiprocessing/forking.py b/Lib/multiprocessing/forking.py
index 6bddfb7..d393817 100644
--- a/Lib/multiprocessing/forking.py
+++ b/Lib/multiprocessing/forking.py
@@ -470,12 +470,26 @@
process.ORIGINAL_DIR = data['orig_dir']
if 'main_path' in data:
+ # XXX (ncoghlan): The following code makes several bogus
+ # assumptions regarding the relationship between __file__
+ # and a module's real name. See PEP 302 and issue #10845
+ # The problem is resolved properly in Python 3.4+, as
+ # described in issue #19946
+
main_path = data['main_path']
main_name = os.path.splitext(os.path.basename(main_path))[0]
if main_name == '__init__':
main_name = os.path.basename(os.path.dirname(main_path))
- if main_name != 'ipython':
+ if main_name == '__main__':
+ # For directory and zipfile execution, we assume an implicit
+ # "if __name__ == '__main__':" around the module, and don't
+ # rerun the main module code in spawned processes
+ main_module = sys.modules['__main__']
+ main_module.__file__ = main_path
+ elif main_name != 'ipython':
+ # Main modules not actually called __main__.py may
+ # contain additional code that should still be executed
import imp
if main_path is None: