fix: support xarray v2025.x by aaronspring · Pull Request #883 · pangeo-data/climpred

aaronspring · 2026-01-16T19:12:49Z

Summary

This PR removes xarray version pins to support v2025.x releases and fixes related compatibility issues.

Closes #880

Changes Made

Dependencies

Removed pin from xarray in
Removed xarray version pins from CI requirements files

Test Infrastructure

Fixed pytest-lazy-fixtures import conflict (wrong package name was being imported)
Removed duplicate from dependency-groups

Analysis

Breaking changes in xarray v2025.x:

v2025.09.1: returns - Verified climpred uses , not affected
v2025.11.0: Operations preserve attributes by default - May need doctest updates
v2025.09.0: pandas 3.0 resampling changes - May need filterwarnings

Testing

Basic xarray 2025.1.2 operations verified working locally
Full CI test suite running on this branch
Local testing limited by netCDF4/Python 3.14 segfault on macOS

References

Original investigation: Dependency and workflow adjustments - Testing #880
xarray Changelog: https://docs.xarray.dev/en/stable/whats-new.html

- Remove xarray version pin (<2025.3.0) from dependencies - Fix pytest-lazy-fixtures import conflict - Prepare for xarray v2025 breaking changes

- Remove <2025.3.0 pin from minimum-tests.yml and maximum-tests.yml - Allow testing with latest xarray versions

for more information, see https://pre-commit.ci

codecov · 2026-01-16T19:37:10Z

Codecov Report

❌ Patch coverage is 90.90909% with 2 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (updates-2025@60bf4ec). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
src/climpred/bootstrap.py	71.42%	2 Missing ⚠️

Additional details and impacted files

@@               Coverage Diff               @@
##             updates-2025     #883   +/-   ##
===============================================
  Coverage                ?   91.51%           
===============================================
  Files                   ?       59           
  Lines                   ?     6273           
  Branches                ?        0           
===============================================
  Hits                    ?     5741           
  Misses                  ?      532           
  Partials                ?        0

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

- Coordinate ordering changed in xarray v2025 - Updated expected outputs in utils.py and classes.py to match new ordering

for more information, see https://pre-commit.ci

- PerfectModelEnsemble.verify: fix coordinate ordering (skill before lead) - PerfectModelEnsemble.bootstrap: update data values for new random state

- Add explicit np.random.seed(42) for reproducibility (attempt)

for more information, see https://pre-commit.ci

Signed-off-by: Trevor James Smith <10819524+Zeitsperre@users.noreply.github.com>

for more information, see https://pre-commit.ci

Signed-off-by: Trevor James Smith <10819524+Zeitsperre@users.noreply.github.com>

…to fix-xarray-v2025

Signed-off-by: Trevor James Smith <10819524+Zeitsperre@users.noreply.github.com>

for more information, see https://pre-commit.ci

aaronspring · 2026-01-21T19:17:12Z

#884 may fix src/climpred/tests/test_bootstrap.py::test_bootstrap_PM_lazy_result in py3.12 https://github.com/pangeo-data/climpred/actions/runs/21221428782/job/61056950903?pr=883

aaronspring · 2026-01-21T19:18:15Z

how about deprecating all below python 3.11 the same way as Xarray?

Zeitsperre · 2026-01-21T21:42:07Z

how about deprecating all below python 3.11 the same way as Xarray?

If you think that's fine, we could do that both here and in xskillscore. I'd much rather have something that is functional/reliable than something that is inconsistent across older/newer Python versions.

Avoid xarray bug with partial chunk dicts like {'iteration': 'auto'} by only chunking non-dask arrays and object-dtype variables.

…an-centered data The test was failing for with because: - The test data is mean-centered (anomalies) - The climatology reference = mean of observations ≈ 0 (a constant) - pearson_r = covariance / (std_forecast * std_climatology) - std(climatology) = 0 when data is mean-centered, causing division by zero -> NaN This is a mathematical limitation, not a bug. The fix excludes from the CI variance check when using metric, while other metrics (, ) still check all skills.

aaronspring · 2026-01-22T15:59:08Z

all green ticks. AWESOME @Zeitsperre

aaronspring · 2026-01-22T16:01:20Z

is uv.lock needed? did that spill over from my uv migration attempt?

Zeitsperre · 2026-01-22T16:03:04Z

is uv.lock needed? did that spill over from my uv migration attempt?

I think it's an artifact, actually. We could look into trying that out later, as you mentioned. I'd love to get this and the bot updates in, then we can release and make more significant changes. Thrilled to see this working as well!

aaronspring · 2026-01-22T16:05:36Z

deleted the AI generated md files. anything missing? anything to do here? @Zeitsperre

Zeitsperre · 2026-01-22T16:08:05Z

I'll give it a final review to be certain!

Signed-off-by: Trevor James Smith <10819524+Zeitsperre@users.noreply.github.com>

Zeitsperre · 2026-01-22T17:08:55Z

I'm going to do some final finagling here for the xclim/xsdba. Let's make this the final version supporting Python3.9 to save ourselves from headaches in the future.

Signed-off-by: Trevor James Smith <10819524+Zeitsperre@users.noreply.github.com>

aaronspring · 2026-01-22T18:59:20Z

great. can be merged but not released before fixing #885

Zeitsperre and others added 9 commits October 21, 2025 09:39

test - try pinning xarray

342c9fc

test - try pinning xarray below v2025.0

54689d2

test - try pinning xarray below v2025.4

c766336

test - try pinning xarray below v2025.3.0

69bf056

fix: support xarray v2025.x

d53d87d

- Remove xarray version pin (<2025.3.0) from dependencies - Fix pytest-lazy-fixtures import conflict - Prepare for xarray v2025 breaking changes

fix: remove xarray version pin from CI requirements

56a40c1

- Remove <2025.3.0 pin from minimum-tests.yml and maximum-tests.yml - Allow testing with latest xarray versions

docs: update PLAN.md with progress

d4b382c

[pre-commit.ci] auto fixes from pre-commit.com hooks

35ca441

for more information, see https://pre-commit.ci

remove uv.lock (not needed for project)

784a853

aaronspring and others added 5 commits January 16, 2026 20:37

fix: update doctest expected outputs for xarray v2025

ba7cc21

- Coordinate ordering changed in xarray v2025 - Updated expected outputs in utils.py and classes.py to match new ordering

[pre-commit.ci] auto fixes from pre-commit.com hooks

0108a36

for more information, see https://pre-commit.ci

fix: update doctest expected outputs for xarray v2025

d49abff

- PerfectModelEnsemble.verify: fix coordinate ordering (skill before lead) - PerfectModelEnsemble.bootstrap: update data values for new random state

fix: update PerfectModelEnsemble.bootstrap doctest for xarray v2025

0d4cb36

- Add explicit np.random.seed(42) for reproducibility (attempt)

fix: update doctest expected outputs for xarray v2025

9cc0f1e

aaronspring force-pushed the fix-xarray-v2025 branch from b7d0b9b to 9cc0f1e Compare January 16, 2026 20:06

pre-commit-ci Bot and others added 7 commits January 16, 2026 20:06

[pre-commit.ci] auto fixes from pre-commit.com hooks

139a705

for more information, see https://pre-commit.ci

add logic for attributes changes in modern xarray

67006e6

Signed-off-by: Trevor James Smith <10819524+Zeitsperre@users.noreply.github.com>

[pre-commit.ci] auto fixes from pre-commit.com hooks

c388a8c

for more information, see https://pre-commit.ci

doctest adjustments

7612cc8

Signed-off-by: Trevor James Smith <10819524+Zeitsperre@users.noreply.github.com>

Merge branch 'fix-xarray-v2025' of github.com:pangeo-data/climpred in…

2e3b973

…to fix-xarray-v2025

doctest fixes and attrs workarounds

55de61a

Signed-off-by: Trevor James Smith <10819524+Zeitsperre@users.noreply.github.com>

[pre-commit.ci] auto fixes from pre-commit.com hooks

3fd4292

for more information, see https://pre-commit.ci

Zeitsperre mentioned this pull request Jan 19, 2026

Dependency and workflow adjustments - Testing #880

Merged

5 tasks

Update lead range in test_bootstrap_PM_lazy_results may fix tests

6869ba1

aaronspring added 2 commits January 22, 2026 16:10

Fix _maybe_auto_chunk to not rechunk dask arrays

892b77b

Avoid xarray bug with partial chunk dicts like {'iteration': 'auto'} by only chunking non-dask arrays and object-dtype variables.

aaronspring and others added 2 commits January 22, 2026 16:10

test_bootstrap_p_climatology no pearsonr

8fa299d

Update CHANGELOG.rst

969136b

Zeitsperre approved these changes Jan 22, 2026

View reviewed changes

aaronspring added 3 commits January 22, 2026 17:03

Delete uv.lock

1e18027

Delete UPDATE.md

887f1a4

Delete PLAN.md

3e6c427

Zeitsperre approved these changes Jan 22, 2026

View reviewed changes

Comment thread pyproject.toml Outdated

Comment thread pyproject.toml Outdated

Comment thread src/climpred/classes.py

Comment thread src/climpred/tests/test_bias_removal.py

Zeitsperre mentioned this pull request Jan 22, 2026

Modern xarray keeps more attributes by default #885

Closed

add lowest supported numpy, specify xclim version for Python3.10+

955e832

Signed-off-by: Trevor James Smith <10819524+Zeitsperre@users.noreply.github.com>

do not install xsdba in Python3.9, pin pandas below v3.0

25e67cb

Signed-off-by: Trevor James Smith <10819524+Zeitsperre@users.noreply.github.com>

Zeitsperre merged commit 1b8857f into updates-2025 Jan 22, 2026
21 checks passed

Conversation

aaronspring commented Jan 16, 2026 • edited by Zeitsperre Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Changes Made

Dependencies

Test Infrastructure

Analysis

Testing

References

Uh oh!

codecov Bot commented Jan 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

aaronspring commented Jan 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

aaronspring commented Jan 21, 2026

Uh oh!

Zeitsperre commented Jan 21, 2026

Uh oh!

aaronspring commented Jan 22, 2026

Uh oh!

aaronspring commented Jan 22, 2026

Uh oh!

Zeitsperre commented Jan 22, 2026

Uh oh!

aaronspring commented Jan 22, 2026

Uh oh!

Zeitsperre commented Jan 22, 2026

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Zeitsperre commented Jan 22, 2026

Uh oh!

aaronspring commented Jan 22, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

aaronspring commented Jan 16, 2026 •

edited by Zeitsperre

Loading

codecov Bot commented Jan 16, 2026 •

edited

Loading

aaronspring commented Jan 21, 2026 •

edited

Loading