New submission from Eric Appelt: While testing my understanding in order to document PEP525 asynchronous generators in the Language Reference (issue 28091) I noticed that the implemented behavior deviates from PEP525, specifically the PEP states that:
"When an asynchronous generator is iterated for the first time, it stores a reference to the current finalizer. If there is none, a RuntimeError is raised. This provides a strong guarantee that every asynchronous generator object will always have a finalizer installed by the correct event loop." I created an asynchronous generator to try to run calling __anext__ interactively without an event loop to check the behavior, and was surprised when I didn't get a RuntimeError even though I had not called sys.set_asyncgen_hooks() to set a finalizer function. I looked at the function async_gen_init_hooks defined in Objects/genobject.c and it appears that if the thread state async_gen_finalizer is NULL, then it just returns zero and allows the calling function to continue. I'm not sure if this is a bug, or this is intentional and the finalizer mechanism should be optional. If it is the latter should PEP525 get updated? ---------- components: Interpreter Core messages: 282737 nosy: Eric Appelt, yselivanov priority: normal severity: normal status: open title: Async generator does not raise RuntimeError if finalizer not set type: behavior versions: Python 3.6, Python 3.7 _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue28910> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
