[issue43068] test_subprocess: test_specific_shell() fails on AMD64 FreeBSD Shared 3.x

STINNER Victor added the comment: The two FreeBSD 3.x buildbot are back to green, I close the issue. -- resolution: -> out of date stage: -> resolved status: open -> closed ___ Python tracker

[issue43068] test_subprocess: test_specific_shell() fails on AMD64 FreeBSD Shared 3.x

Kubilay Kocak added the comment: P.S gdb and bash have been rebuilt :) -- ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue43068] test_subprocess: test_specific_shell() fails on AMD64 FreeBSD Shared 3.x

Kubilay Kocak added the comment: Note: Not suggesting dep removals, just a try-except-skip block in the (hopefully) one place the sub-process handler in the test framework that runs then :) -- ___ Python tracker

[issue43068] test_subprocess: test_specific_shell() fails on AMD64 FreeBSD Shared 3.x

STINNER Victor added the comment: > As far as I'm aware, the best practice and recommendation for these cases is > that 'tests should not test dependencies', and to skip/skipif these > (external) dependencies are not available or fail to load/run. In theory, we should limit external

[issue43068] test_subprocess: test_specific_shell() fails on AMD64 FreeBSD Shared 3.x

Kubilay Kocak added the comment: Rebuilding bash & gdb now to pickup new/updated ncurses library. Still suggest picking up "subprocess.CalledProcessError" generically in test framework for tests that use subprocess/external commands -- ___ Python

[issue43068] test_subprocess: test_specific_shell() fails on AMD64 FreeBSD Shared 3.x

Kubilay Kocak added the comment: The failure mode is a generic "an external dependency expected by a test suite is not available or doesn't run", applying to many core Python test that invokes an external utilities, and not specific to OS. As far as I'm aware, the best practice and

[issue43068] test_subprocess: test_specific_shell() fails on AMD64 FreeBSD Shared 3.x

STINNER Victor added the comment: > Would you like me to rebuild bash/gdb to fix the error, or wait for the tests > to be fixed so that they can handle the error condition? I don't think that Python test suite should handle the case of a broken FreeBSD setup. Could you either disable the

[issue43068] test_subprocess: test_specific_shell() fails on AMD64 FreeBSD Shared 3.x

Kubilay Kocak added the comment: FreeBSD base was updated on the FreeBSD workers, between versions that changes the way the base curses library was built, and what it built. The net results was that any software (from ports) linked against base curses needed to be rebuilt, and until then

[issue43068] test_subprocess: test_specific_shell() fails on AMD64 FreeBSD Shared 3.x

Change by STINNER Victor : -- nosy: +koobs ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue43068] test_subprocess: test_specific_shell() fails on AMD64 FreeBSD Shared 3.x

New submission from STINNER Victor : https://buildbot.python.org/all/#/builders/483/builds/758 FAIL: test_specific_shell (test.test_subprocess.POSIXProcessTestCase) -- Traceback (most recent call last): File