gh-119517: Fixes for pasting in pyrepl #120251
Conversation
…uentially (pythonGH-118864) (python#118927) The free-threaded build currently immortalizes some objects once the first thread is started. This can lead to test failures depending on the order in which tests are run. This PR addresses those failures by suppressing immortalization or skipping the affected tests. (cherry picked from commit b309c8e) Co-authored-by: Sam Gross <[email protected]>
…ve them (pythonGH-118747) (python#118931) This ensures we don't lose races that occur in subprocesses or interleave races from workers running in parallel. Log files are collected and packaged into a zipfile that can be downloaded from the "Artifacts" section of the workflow run. (cherry picked from commit b88889e) Co-authored-by: mpage <[email protected]>
…ythonGH-118923) (python#118933) (cherry picked from commit 35c4361) Co-authored-by: Tian Gao <[email protected]>
…ationWarning (pythonGH-117668) (pythonGH-118956) pythongh-117655: Prevent `test_strptime` from raising a DeprecationWarning (pythonGH-117668) * Fix `test_strptime` raises a DeprecationWarning * Ignore deprecation warnings where appropriate. * Update Lib/test/datetimetester.py This is follow on work to silence unnecessary warnings from the test suite that changes for python#70647 added. (cherry picked from commit abead54) Co-authored-by: Nice Zombies <[email protected]>
…ss (pythonGH-118902) (python#118968) pythongh-118899: Add tests for `NotImplemented` attribute access (pythonGH-118902) (cherry picked from commit ec1398e) Co-authored-by: Nikita Sobolev <[email protected]>
…il with --enable-experimental-jit (pythonGH-118959) (cherry picked from commit 5b941e5) Co-authored-by: Savannah Ostrowski <[email protected]>
…ional-only arguments with **kwargs (pythonGH-103404) (python#118985) (cherry picked from commit 9c15202) Co-authored-by: Jacob Walls <[email protected]>
…ythonGH-118964) (python#118991) Co-authored-by: Nikita Sobolev <[email protected]>
…nGH-118880) Fixes an AttributeError that occurs when checking if ns.temp is an absolute path during building from source on Windows. (cherry picked from commit d8a82cc) Co-authored-by: I-Shen Leong <[email protected]>
… f_trace is not set (pythonGH-118979) (python#119007) pythongh-58933: Make pdb return to caller frame correctly when f_trace is not set (pythonGH-118979) (cherry picked from commit f526314) Co-authored-by: Tian Gao <[email protected]>
…ls.update_wrapper` (pythonGH-119012) (python#119013) pythongh-119010: Adds docs about `__type_params__` to `functools.update_wrapper` (pythonGH-119012) (cherry picked from commit b04c497) Co-authored-by: Nikita Sobolev <[email protected]> Co-authored-by: Jelle Zijlstra <[email protected]>
…memodule` (pythonGH-118999) (python#119018) pythongh-118998: Handle errors correctly in `tmtotuple` in `timemodule` (pythonGH-118999) (cherry picked from commit fc75792) Co-authored-by: Nikita Sobolev <[email protected]>
…present (pythonGH-119020) (cherry picked from commit e04cd96) Co-authored-by: Michał Górny <[email protected]>
…h path starting with multiple slashes and no authority (pythonGH-113563) (pythonGH-119023) (cherry picked from commit e237b25) Co-authored-by: Serhiy Storchaka <[email protected]>
…honGH-119031) (pythonGH-119037) (cherry picked from commit 331d385) Co-authored-by: Serhiy Storchaka <[email protected]>
…GH-119005) (python#119038) (cherry picked from commit a9328e2) Co-authored-by: Alex Waygood <[email protected]>
I honestly forgot this slipped into 3.13, but I think it's worth highlighting more, as it is a PEP-sized change that makes the type system significantly more powerful. @Yhg1s I think it's also worth mentioning in your release announcements. (cherry picked from commit ee13797) Co-authored-by: Jelle Zijlstra <[email protected]>
…pythonGH-119056) (cherry picked from commit 94591dc) Co-authored-by: Steve Dower <[email protected]>
…string (pythonGH-119071) (cherry picked from commit 7d722b7) Co-authored-by: Thomas Grainger <[email protected]>
…dows (pythonGH-118782) (pythonGH-119072) For unknown reasons some arguments for Tkinter binding can be created as a 1-tuple containing a Tcl_Obj when wantobjects is 2. (cherry picked from commit 5b88d95) Co-authored-by: Serhiy Storchaka <[email protected]>
…119074) Co-authored-by: Rafael Fontenelle <[email protected]>
…n#119083) Use literal syntax in origin property (pythonGH-119029) (cherry picked from commit 66b73e9) Co-authored-by: Rafael Fontenelle <[email protected]>
…h in tests (pythonGH-119065) (pythonGH-119087) (cherry picked from commit 0152dc4) Co-authored-by: Serhiy Storchaka <[email protected]>
…onGH-119091) (cherry picked from commit b683994) Co-authored-by: Serhiy Storchaka <[email protected]>
…ses docs (pythonGH-119082) (python#119097) Co-authored-by: Shantanu <[email protected]>
(cherry picked from commit ab73bcd) Co-authored-by: Miro Hrončok <[email protected]> Co-authored-by: Erlend E. Aasland <[email protected]>
…cils.h (pythonGH-119101) (cherry picked from commit 4702b7b)
Fix typos in documentation (pythonGH-119092) (cherry picked from commit 65de194) Co-authored-by: Xie Yanbo <[email protected]> Co-authored-by: Alex Waygood <[email protected]>
hugovk
removed request for
ncoghlan,
iritkatriel,
pfmoore,
ericsnowcurrently,
hugovk,
pganssle,
rhettinger,
willingc,
gvanrossum,
pradyunsg,
serhiy-storchaka,
corona10,
ethanfurman,
markshannon,
pablogsal,
FFY00,
hauntsaninja,
gaogaotiantian,
erlend-aasland,
sethmlarson,
terryjreedy,
lysnikolaou,
ezio-melotti,
brandtbucher,
isidentical,
kumaraditya303 and
AlexWaygood
Don't worry about it, but please don't rebase and force push. If to want to update from https://devguide.python.org/getting-started/pull-request-lifecycle/#quick-guide |
Ah, the thing I tried to do that failed horribly wasn't a rebase, it was retargetting the PR to main instead of 3.13 via the GitHub UI - I picked the wrong branch when creating the PR, but because I had based the change on the 3.13 branch, GitHub decided I wanted to merge all of the commits since 3.13 branched off into main. (Ok, I guess that is a rebase, I just didn't realize it was one when I clicked the button, heh.)
I realize there will be a squash merge, but tidying the commits makes reviews much easier. Is force pushing before the PR has been reviewed still frowned upon? Or force pushing to draft PRs? It seems wasteful to make reviewers review known buggy commits for problems that CI catches if I can instead fix those problems before anyone ever looks at it. |
Yep, that'll reliably do it and ping every codeowner too 🙃 We'd disable that option if possible.
My main problem with force pushes is the notification takes you to an empty diff, so it costs an extra click to go to the main PR page, and it's not obvious to see what changed from the last time I checked.
I feel like marking a PR as draft is enough of a signal to say it's not ready for review yet, but an explicit comment can help. I'm not sure how often people review commit-by-commit compared to the whole diff, so I wouldn't worry about small new commits that fix earlier ones. |
Drastically improve the performance of pyrepl when pasting large amounts of text.
CC: @pablogsal