RLS: 1.3.3

[simonjayhawkins]

Tracking issue for the 1.3.3 release. Currently scheduled for September 05, 2021.

https://github.com/pandas-dev/pandas/milestone/89

List of open regressions: https://github.com/pandas-dev/pandas/issues?q=is%3Aopen+is%3Aissue+label%3ARegression

performance regressions between 1.2.5 and 1.3.x https://simonjayhawkins.github.io/fantastic-dollop/#regressions?sort=3&dir=desc&branch=1.3.x

cc @pandas-dev/pandas-core @pandas-dev/pandas-triage

[simonjayhawkins]

xref question at #43181 (comment)

---

There is a lengthy discussion in numpy/numpy#19630 about proving 3.10 wheels on PyPI for py3.10

numpy have published aarch64 and manylinux2014 wheels for py3.10rc1 with their latest release, numpy 1.21.2

we are now also building 3.10 wheels, MacPython/pandas-wheels#152 for nightly [see note 1] and for the latest test of 1.3.x these are successfully built (https://github.com/MacPython/pandas-wheels/runs/3495117390)

These wheels are built and tested with numpy 1.21.2

so when we release 1.3.3, these wheels will be built and the release scripts will publish them.

The question is should these be uploaded to PyPI or not? (if not would just delete the downloaded wheel before running twine, nbd)

---

on ci, actions-310-dev we currently use numpy 1.22 dev

pip install -i https://pypi.anaconda.org/scipy-wheels-nightly/simple numpy

maybe we should change ci to use the numpy 1.21.2 wheels (and backport) prior to 1.3.3 release if we are going to upload the pandas py3.10 wheels?

---

[1] note that we have yet actually published 3.10 wheels on nightly, but that's a different issue. MacPython/pandas-wheels#153

[simonjayhawkins]

should mention that no urgency here. 1.3.3 was originally scheduled for September 05, 2021 but quite happy to push that out a week as we have more PRs in the pipeline than the number of issues fixed so far on 1.3.3.

for the patch releases the schedule is probably like 3.3.4.5.6 weeks so that they are spaced out over the 6 months with the first issues reported getting fixed soonest.

we are 3 weeks since 1.3.2, so I am quite happy to give it another week for 1.3.3

[jreback]

sounds good to push it out a week

and yes should try to publish 310 wheels

[simonjayhawkins]

planning to release 1.3.3 on Sunday. please use the blocker label on any issues/PRs if necessary (next release in approx 5 weeks)

1.3.4 milestone created https://github.com/pandas-dev/pandas/milestone/90 and issues without PRs and PRs that need more work can be moved across.

[simonjayhawkins]

pre-release checklist

• wheels building test wheel build status of 1.3.x branch MacPython/pandas-wheels#135
• release readiness passing https://github.com/simonjayhawkins/pandas-release/actions?query=workflow%3A%22Tag+Release%22
• release notes checked https://pandas.pydata.org/pandas-docs/dev/whatsnew/v1.3.3.html
• ci checks https://github.com/pandas-dev/pandas/commits/1.3.x
• no manual backports outstanding https://github.com/pandas-dev/pandas/pulls?q=is%3Apr+label%3A%22Still+Needs+Manual+Backport%22+is%3Aclosed
• no open issues/PRs (except this one) https://github.com/pandas-dev/pandas/milestone/89

[simonjayhawkins]

• no open issues/PRs (except this one) https://github.com/pandas-dev/pandas/milestone/89

#42946 will be closed after successful upload of wheels to PyPI

starting release shortly.

[simonjayhawkins]

1.3.3 milestone now closed off.

• tag locally
• build sdist
• pip test
• conda test
• build docs
• push tag https://github.com/pandas-dev/pandas/tags
• open PR to build wheels RLS v1.3.3 MacPython/pandas-wheels#154
• upload docs https://pandas.pydata.org/pandas-docs/version/1.3.3/
• wheels in staging area https://anaconda.org/multibuild-wheels-staging/pandas/files
• github release - https://github.com/pandas-dev/pandas/releases
• upload wheels to PyPI https://pypi.org/project/pandas/
• change link for stable docs https://pandas.pydata.org/docs/
• start v1.3.4 release notes DOC: Start v1.3.4 release notes #43530
• conda-forge bot picked up on release https://github.com/conda-forge/pandas-feedstock/pulls
• available on conda-forge https://anaconda.org/conda-forge/pandas/files
• announce