Kyle Stanley <aeros...@gmail.com> added the comment:
I now have a working implementation, for both ThreadPoolExecutor and ProcessPoolExecutor. I've also ensured that the tests I added are not vulnerable to race conditions with the following: ``` [aeros:~/repos/aeros-cpython]$ ./python -m test test_concurrent_futures --match test_cancel_futures -j200 -v -F [snip] Ran 4 tests in 2.865s OK 0:03:24 load avg: 143.25 [1018] test_concurrent_futures passed -- running: test_concurrent_futures (2 min 36 sec), test_concurrent_futures (35.8 sec) test_cancel_futures (test.test_concurrent_futures.ProcessPoolForkProcessPoolShutdownTest) ... 0.57s ok test_cancel_futures (test.test_concurrent_futures.ProcessPoolForkserverProcessPoolShutdownTest) ... 0.80s ok test_cancel_futures (test.test_concurrent_futures.ProcessPoolSpawnProcessPoolShutdownTest) ... 0.53s ok test_cancel_futures (test.test_concurrent_futures.ThreadPoolShutdownTest) ... 0.20s ok ``` I'll attach a PR to the issue once I finish writing the documentation and "What's New" entry. Note: I was originally going to do this in two separate PRs, one for each executor, but it seemed to make more sense to just have it as a single cohesive PR since Executor.shutdown() shares the same documentation for both executors. ---------- _______________________________________ Python tracker <rep...@bugs.python.org> <https://bugs.python.org/issue39349> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
