Repository navigation
3.11.5 test_httpservers fails on unhandled DeprecationWarning ("invalid escape sequence '\\?'") #108569
Description
Bug report
Checklist
- I am confident this is a bug in CPython, not a bug in a third-party project
- I have searched the CPython issue tracker,
and am confident this bug has not been reported before
CPython versions tested on:
3.11
Operating systems tested on:
Linux
Output from running 'python -VV' on the command line:
Python 3.11.3 (main, Jun 5 2023, 09:32:32) [GCC 13.1.1 20230429]
A clear and concise description of the bug:
Hi! When attempting to upgrade the python package for Arch Linux to 3.11.5 I ran into the following test issue:
[..]
test.test_httpservers
FAIL
======================================================================
FAIL: test_all (test.test___all__.AllTest.test_all)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/build/python/src/Python-3.11.5/Lib/test/test___all__.py", line 129, in test_all
self.check_all(modname)
File "/build/python/src/Python-3.11.5/Lib/test/test___all__.py", line 43, in check_all
with warnings_helper.check_warnings(
File "/build/python/src/Python-3.11.5/Lib/contextlib.py", line 144, in __exit__
next(self.gen)
File "/build/python/src/Python-3.11.5/Lib/test/support/warnings_helper.py", line 185, in _filterwarnings
raise AssertionError("unhandled warning %s" % reraise[0])
AssertionError: unhandled warning {message : DeprecationWarning("invalid escape sequence '\\?'"), category : 'DeprecationWarning', filename : '/build/python/src/Python-3.11.5/Lib/test/test_httpservers.py', lineno : 445, line : None}
----------------------------------------------------------------------
Ran 1 test in 0.621s
FAILED (failures=1)
test test___all__ failed
0:00:05 load avg: 3.34 [ 12/432/1] test___future__ -- test___all__ failed (1 failure)
test_attributes (test.test___future__.FutureTest.test_attributes) ... ok
test_names (test.test___future__.FutureTest.test_names) ... ok
----------------------------------------------------------------------
Ran 2 tests in 0.000s
[..]
Full build and check logs:
python-3.11.5-1-x86_64-build.log
python-3.11.5-1-x86_64-check.log
Build script for the package: https://gitlab.archlinux.org/archlinux/packaging/packages/python/-/blob/main/PKGBUILD
cc @felixonmars
Activity
hugovk commentedon Aug 28, 2023
The f-string was turned into a raw f-string for
mainand 3.12, but not 3.11:test___all__.AllTest.test_allis failing on every PR #105821@pablogsal @AlexWaygood Shall we backport it to 3.11?
AlexWaygood commentedon Aug 28, 2023
That makes sense to me; the change is very low-risk 👍
@dvzrv, are you running your tests with
-W/-Werror? (I'm curious why we're not seeing the same test failures in CI over at CPython!)pablogsal commentedon Aug 28, 2023
Let's do the backport! 👍
Also let's investigate as well why we are didn't experience this in our own Ci. 🤔
AlexWaygood commentedon Aug 28, 2023
I triggered the backport, and I've enabled automerge:
dvzrv commentedon Aug 28, 2023
Not that I am aware of (definitely not explicitly).
Rebuilding with the patch applied now to see where this leads and will report back. Thanks for the super quick responses here!
dvzrv commentedon Aug 28, 2023
Building with #108576 applied on top of 3.11.5 fixes the issue for us. Thank you! :)
hugovk commentedon Aug 29, 2023
Because we're no longer running
Lib/test/test___all__.pyon the CI.Here's the last build before the backport (33eee43):
https://github.com/python/cpython/actions/runs/6000276301/job/16272033246
Checking out 33eee43 locally, building and running the test fails:
(Note: running it a second time passes, some intermediary files need to be deleted first.)
test___all___was added to the CI in #26640 and skipped in #30897 (both by @pablogsal) and then skipped in a different way in #31632.