Adopt new pandas frequency codes #1534
Assignees
Labels
bug
Something isn't working
enhancement
New feature or request
priority
Immediate priority
standards / conventions
Suggestions on ways forward
Milestone
Comments
aulemahal
added
bug
This was referenced Nov 28, 2023
aulemahal
added a commit
that referenced
this issue
Nov 29, 2023
<!--Please ensure the PR fulfills the following requirements! --> <!-- If this is your first PR, make sure to add your details to the AUTHORS.rst! --> ### Pull Request Checklist: - [x] This PR addresses an already opened issue (for bug fixes / features) - This PR is a temporary fix. #1534 will be fixed by inverting the pins in the next release (0.48). - [ ] Tests for the changes have been added (for bug fixes / features) - [ ] (If applicable) Documentation has been added / updated (for bug fixes / features) - [x] CHANGES.rst has been updated (with summary of main changes) - [ ] Link to issue (:issue:`number`) and pull request (:pull:`number`) has been added ### What kind of change does this PR introduce? * Pins xarray under 2023.11 and pandas under 2.2 (which is not yet released). ### Does this PR introduce a breaking change? No. I expect the "upstream" builds to fail, errors related to frequencies are ok and will be fixed in the near future. ### Other information: The plan of inverting the pins in 0.48 implies that 0.47 will be the last version supporting python 3.8.
This was referenced Dec 5, 2023
Zeitsperre
added a commit
to bird-house/finch
that referenced
this issue
Dec 5, 2023
## Overview Changes: * Added pins to xarray and pandas to prevent issues with breaking changes. * Fixed an error with a cftime index fixing function. * Switched `bump2version` for `bump-my-version` * Do not fail when running pip check in conda environments ## Related Issue / Discussion Ouranosinc/xclim#1534 pydata/xarray#8394
5 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Addressing a Problem?
Pandas has made changes to the frequency string codes, which will begin in version 2.2.
The first step is the introduction of a new set of default codes:
A->YH->hT->minS->s(and so on for smaller periods)And the addition of the "E" suffix to denote a frequency anchored at the end of the period (instead of it being the default when not mentioned) (applies to Q, and M).
All the previous codes are now deprecated, but will be valid inputs for 2.2.
Potential Solution
If we change all the hardcoded codes to the new ones, we will need to pin pandas, so a temporary dynamic solution might be necessary (as done in xarray, but more complex).
I suggest a
ensure_correct_freq()function that is called everywhere a frequency code is sent to xarray/pandas. The function will translate using the old or the new version depending on the installed version of pandas.Additional context
The function could be removed when we pin pandas >= 2.2 and all codes written with the new syntax.
xscen could make use of that function also, if needed.
See xarray issue, PR and links to the Pandas issues : pydata/xarray#8394
Contribution
Code of Conduct
The text was updated successfully, but these errors were encountered: