Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

MAINT, REL: prepare for SciPy 1.11.0 "final" #18724

Merged

Conversation

tylerjereddy
Copy link
Contributor

@tylerjereddy tylerjereddy commented Jun 22, 2023

  • I propose we're ready to proceed with the "final" release of SciPy 1.11.0 -- @h-vetinari do you concur as conda-forge release manager? I haven't seen reports of problems apart from the 1 backport patch for MacOS ARM with latest NumPy.
  • remove doc note about SciPy 1.11.0 not being released yet
  • update release note lists one more time
  • python dev.py test -j 12 local testing on Apple silicon with NumPy 1.25.0
  • python dev.py doc local test on this branch
  • make release commit for SciPy 1.11.0 "final" unreleased (switches to released during final release process)

Backports included:

  1. MAINT: NumPy 1.25.0 shims for arm64 #18697
  2. DEP: interpolate: delay interp2d deprecation and update link #18698

Note that there is still a large list of merged backport PRs, but they appear to be targeted for 1.10.x series, and I'm not so sure we'll do 1.10.2 at this point anyway.

tylerjereddy and others added 5 commits June 22, 2023 09:25
* set version to `1.11.0` "final"
unreleased
* a few floating-point error shims to allow our
testsuite to pass for MacOS arm64 with the recently-released
NumPy `1.25.0` -- the issues were showing up in our Cirrus
CI and I could also reproduce them locally on M2 max

* locally I still get a few errors, but those were present
on M2 series prior to NumPy `1.25.0` per scipygh-18308, and I think
this patch brings us back to that basically
* add a general `xfail` to `test_maxiter_worsening()`
based on reviewer feedback--this test has just been causing
way too many problems for us
* update author/list entries

* remove the note about SciPy not being
released yet
@tylerjereddy
Copy link
Contributor Author

For scipy/sparse/linalg/_dsolve/tests/test_linsolve.py::TestSplu::test_threads_parallel, I've seen that failing in some recent PRs against main. If a fix/skip/adjustment is made, I'm happy to backport it.

The Alpine job hasn't flushed yet, rest looks ok.

@h-vetinari
Copy link
Member

@h-vetinari do you concur as conda-forge release manager? I haven't seen reports of problems apart from the 1 backport patch for MacOS ARM with latest NumPy.

I haven't pinged you directly, but as of 1.11.0rc2, conda-forge/scipy-feedstock#235 is fully green. :)

So yes, good to go!

@tylerjereddy
Copy link
Contributor Author

Ok, core dev approval and the 1 remaining CI failure also plagues main, and is additionally restricted to the pre-release job, so I think that won't block me on the 1.11.0 release. It may be good to deal with that for 1.11.1 though, if the issue is on our end.

@tylerjereddy tylerjereddy merged commit 8db01f2 into scipy:maintenance/1.11.x Jun 23, 2023
22 of 23 checks passed
@tylerjereddy tylerjereddy deleted the treddy_1_11_0_final_prep branch June 23, 2023 15:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants