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

RLS: 1.4.4 #47485

Closed
simonjayhawkins opened this issue Jun 23, 2022 · 11 comments
Closed

RLS: 1.4.4 #47485

simonjayhawkins opened this issue Jun 23, 2022 · 11 comments
Labels
Milestone

Comments

@simonjayhawkins
Copy link
Member

Tracking issue for the 1.4.4 release.

https://github.com/pandas-dev/pandas/milestone/96

Currently scheduled for July 31, 2022.

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

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

@simonjayhawkins simonjayhawkins added this to the 1.4.4 milestone Jun 23, 2022
@simonjayhawkins
Copy link
Member Author

Currently scheduled for July 31, 2022.

1.4.4 does not need to be released before 1.5.0rc0 and not even before 1.5.0.

if 1.5.0rc0 is released before 1.4.4 we need double backports. nbd.

if 1.5.0 is released before 1.4.4 we need double backports and maybe need changes to conda-forge. again nbd.

The management of the 1.4.4 release can be done independently to getting the 1.5 release cycle started.

So I think OK to change the date for this release to the end of this month to ensure all open regressions are considered for backport and review the release date again when the release dates of 1.5.0rc0 and 1.5.0 are firmed up.

@jreback
Copy link
Contributor

jreback commented Aug 4, 2022

@simonjayhawkins yeah we can for sure decouple these releases and push 1.4.4 a month sounds good

@lithomas1
Copy link
Member

Currently scheduled for July 31, 2022.

1.4.4 does not need to be released before 1.5.0rc0 and not even before 1.5.0.

This is just my personal preference, but I would appreciate if we closed out the 1.4.x cycle before starting 1.5.x.
(I don't really feel like backporting the Python 3.11 fixes and reworking the builds for 1.4.x at this point :) )

@jreback
Copy link
Contributor

jreback commented Aug 8, 2022

we don't need 3.11 support at all in 1.4.x

we also want to keep 1.4.x orthogonal

it's possible we need a
1.4.5 even (though maybe not likely)

@simonjayhawkins
Copy link
Member Author

@jorisvandenbossche following up to #45223 (comment)

if we make good progress getting the open PRs merged in the next few days, then there would be more fixes than I would be comfortable with for a final patch release.

So, if that was the case, I would propose doing a 1.4.4 about the same time as 1.5.0rc0 and then have a 2 week window to get a 1.4.5 out before the actual 1.5 release.

@simonjayhawkins
Copy link
Member Author

So, if that was the case, I would propose doing a 1.4.4 about the same time as 1.5.0rc0 and then have a 2 week window to get a 1.4.5 out before the actual 1.5 release.

There are at this time 4 open PRs for regression fixes, leaving around 9 open identified 1.3.5 regressions.

There has been good progress over the last few days getting some regressions fixed. A 1.4.4 release anytime around now would (imo) be a good idea with a 1.4.5 release in the next couple of weeks.

So I could do it now (so that the release is before 1.5.0rc0) but this won't give much chance for feedback from others, so I propose we do it tomorrow after the 1.5.0rc0 release (and maybe get some/all of the 4 open PRs merged).

The other alternative is to not do a 1.4.5 and do the 1.4.4 sometime before 1.5.0.

@simonjayhawkins
Copy link
Member Author

I've removed the milestone from the issues milestoned 1.4.4 where the discussion has not progressed enough to implement a fix. That just left #47101. have put a PR, #48313

planning to release 1.4.4 tomorrow, with or without #48313.

@jorisvandenbossche
Copy link
Member

Sounds good!

(to be honest, I don't think #47101 is crucial, since the support for having pd.NA in object dtype is quite limited / buggy in general)

@simonjayhawkins
Copy link
Member Author

(to be honest, I don't think #47101 is crucial, since the support for having pd.NA in object dtype is quite limited / buggy in general)

not merged so closed.

1.4.4 milestone now closed off. releasing tomorrow am

@simonjayhawkins
Copy link
Member Author

simonjayhawkins commented Aug 31, 2022

starting release now

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

4 participants