bpo-46198: rename duplicate tests and remove unused code#30297
Conversation
|
|
Sorry, something went wrong.
|
I am addressing feedback from https://bugs.python.org/issue46198, so I am increasing the scope of this PR to also include other similar cases. I am using Amount of tests:
|
Sorry, something went wrong.
|
Title of the PR needs to change! |
Sorry, something went wrong.
|
Rebased to solve conflicts in |
Sorry, something went wrong.
|
In general, please do merges for CPython rather than rebases and force pushes: https://devguide.python.org/pullrequest/ |
Sorry, something went wrong.
|
@raghavthind2005 hello 👋🏽 Reviews by non-core developers are useful if you carefully consider the bug report, the changes, the discussions. Sometimes you need to get the branch locally, build python and try out the code to see if it does what it’s supposed to. It is useful if people who are not core devs do these things and write a message on the PR to say that. But just giving +1 without engaging with the process or the people has little value. |
Sorry, something went wrong.
JelleZijlstra
left a comment
There was a problem hiding this comment.
Thanks, I'm planning to merge this (cc @gvanrossum).
The duplicate test cases are quite bad and we should definitely fix them. The duplicate imports aren't as bad but it's still cleaner to get them fixed.
Sorry, something went wrong.
|
@JelleZijlstra Looks fine to me. (In general I think you don't have to CC me on things you plan to merge any more, unless you really want me to have another look. You're doing great!) |
Sorry, something went wrong.
|
Great, thanks! I was thinking of asking when I could leave new core dev quarantine :) |
Sorry, something went wrong.
|
Thanks @sobolevn for the PR, and @JelleZijlstra for merging it 🌮🎉.. I'm working now to backport this PR to: 3.9, 3.10. |
Sorry, something went wrong.
|
Sorry @sobolevn and @JelleZijlstra, I had trouble checking out the |
Sorry, something went wrong.
|
Sorry, @sobolevn and @JelleZijlstra, I could not cleanly backport this to |
Sorry, something went wrong.
|
I'll backport |
Sorry, something went wrong.
…onGH-30297). (cherry picked from commit 6c83c8e) Co-authored-by: Nikita Sobolev <mail@sobolevn.me>
…nGH-30297). (cherry picked from commit 6c83c8e) Co-authored-by: Nikita Sobolev <mail@sobolevn.me>
|
#31796 is the 3.10 backport, not sure why the bot didn't pick it up |
Sorry, something went wrong.
pythonGH-30297 removed a duplicate `from test import support` statement from `test_asyncio.test_sslproto`. However, in between that PR being filed and it being merged, pythonGH-31275 removed the _other_ `from test import support` statement. This means that `support` is now undefined in `test_asyncio.test_sslproto`, causing the CI to fail on all platforms for all PRS.
GH-30297 removed a duplicate `from test import support` statement from `test_asyncio.test_sslproto`. However, in between that PR being filed and it being merged, GH-31275 removed the _other_ `from test import support` statement. This means that `support` is now undefined in `test_asyncio.test_sslproto`, causing the CI to fail on all platforms for all PRS.
…nGH-30297) (pythonGH-31797) (cherry picked from commit 6c83c8e) Co-authored-by: Nikita Sobolev <mail@sobolevn.me>
Right now, two tests have the same name:
cpython/Lib/test/test_email/test__header_value_parser.py
Line 304 in 8e11237
cpython/Lib/test/test_email/test__header_value_parser.py
Line 398 in 8e11237
The first one is always skipped.
I think that NEWS should not be added.
https://bugs.python.org/issue46198