SPEC 0: drop py310 and support py313#6195
Conversation
benchmarks/asv.conf.json
Outdated
| ], | ||
| "44fae030": [ | ||
| "ENV_PARENT={conf_dir}/.asv/env/nox310", | ||
| "PY_VER=3.10 nox --envdir={env_parent} --session=tests --install-only --no-error-on-external-run --verbose" |
There was a problem hiding this comment.
In this case I would prefer that we keep the old one too, as we occasionally have need to benchmark further back in history.
benchmarks/asv.conf.json
Outdated
| ], | ||
| "delegated_env_commands": { | ||
| "c8a663a0": [ | ||
| "ENV_PARENT={conf_dir}/.asv/env/nox313", |
There was a problem hiding this comment.
This is awkward - we don't know what the new commit hash will be once this is merged. I didn't think of that 🙄
There was a problem hiding this comment.
Discussed with @ESadek-MO: if you just provide one of the commit hashes that is on your branch, and then we'll merge this one with a merge-commit (not a squash).
There was a problem hiding this comment.
Almost, however there are now duplicate 3.13, which is what is causing the benchmark failure. The sequence should be 3.10, 3.11, 3.12, 3.13 - unlike other things we are not describing CI coverage, but support for benchmarking older commits if requested.
|
Looks like python 3.13 is causing a lot of test failures. Most of these seem to be in cml files where I'm seeing |
|
Looks like this change might be the cause of the failing tests, python/cpython#107947 . |
|
hi folks, we are ready to roll out support for ESMValCore and Python 3.13 in ESMValGroup/ESMValCore#2566 - @bouweandela brought this issue to my attention, do we understand that |
The test failures are all caused by a change in XML character representation, which stops us comparing against known-good-outputs. This has made 5% of our tests 'blind' until we can work on a solution. But we have NO evidence of any Iris functionality breaking, and 95% of our tests still work and they show Iris functioning fine. Your own passing tests (ESMValGroup/ESMValCore#2566 (comment)) back this up even more. So the Iris 'project' cannot support Python 3.13 until we can get our CI working with 3.13; we have no immediate plans for finishing this work as we need space to finish some other things. But it looks like the Iris software supports 3.13 just fine already 👍. This happens with all new Python versions - our users have success with it before we even get a chance to test against it. |
wonderful, exactly the answer I was looking for, both wrt clarity and content, cheers, Martin! 🍺 |
# Conflicts: # docs/src/whatsnew/latest.rst # requirements/locks/py311-linux-64.lock # requirements/locks/py312-linux-64.lock # requirements/locks/py313-linux-64.lock
cea9fde to
7957983
Compare
7957983 to
217b162
Compare
for more information, see https://pre-commit.ci
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #6195 +/- ##
==========================================
- Coverage 89.88% 89.87% -0.02%
==========================================
Files 90 90
Lines 23447 23450 +3
Branches 4361 4361
==========================================
Hits 21076 21076
- Misses 1644 1646 +2
- Partials 727 728 +1 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Current review status
|
|
We have a remaining problem with how our expected checks are set up - they are specific to Python versions:
This has previously affected Stratify, but considering the lower traffic that wasn't such a problem there. @HGWright can you offer any extra insight? |
Decided as a team to allow repository admins to override the expected checks. |
⏱️ Performance Benchmark Report: 601d471Performance shiftsFull benchmark resultsGenerated by GHA run |
* upstream/main: (98 commits) [pre-commit.ci] pre-commit autoupdate (SciTools#6335) SPEC 0: drop py310 and support py313 (SciTools#6195) Better benchmarking Python version handling (SciTools#6329) Move loading and combine code into their own submodules. (SciTools#6321) Bump scitools/workflows from 2025.02.1 to 2025.02.2 (SciTools#6327) replaced reference from build to python build (SciTools#6324) [pre-commit.ci] pre-commit autoupdate (SciTools#6315) Cache Dask arrays created from `NetCDFDataProxy`s to speed up loading files with multiple variables (SciTools#6252) Bump scitools/workflows from 2025.02.0 to 2025.02.1 (SciTools#6313) [pre-commit.ci] pre-commit autoupdate (SciTools#6310) Bump scitools/workflows from 2025.01.5 to 2025.02.0 (SciTools#6306) Updated environment lockfiles (SciTools#6301) Improve speed of loading small NetCDF files (SciTools#6229) [pre-commit.ci] pre-commit autoupdate (SciTools#6298) Use cube chunks for weights in aggregations with smart weights (SciTools#6288) Updated environment lockfiles (SciTools#6296) Bump scitools/workflows from 2025.01.4 to 2025.01.5 (SciTools#6300) Bump scitools/workflows from 2025.01.3 to 2025.01.4 (SciTools#6295) Lazy rectilinear interpolator (SciTools#6084) Revert "Fix broken link. (SciTools#6246)" (SciTools#6297) ...

🚀 Pull Request
Description
closes #6177