Updated esmf-related pins

[schlunma]

Description

With the new release of iris-esmf-regrid we can update some pins in the environment.

See also SciTools/iris-esmf-regrid#342 (comment).

---

Checklist

It is the responsibility of the author to make sure the pull request is ready to review. The icons indicate whether the item will be subject to the 🛠 Technical or 🧪 Scientific review.

• 🧪 The new functionality is relevant and scientifically sound
• 🛠 This pull request has a descriptive title and labels
• 🛠 Code is written according to the code quality guidelines
• 🧪 and 🛠 Documentation is available
• 🛠 Unit tests have been added
• 🛠 Changes are backward compatible
• 🛠 Any changed dependencies have been added or removed correctly
• 🛠 The list of authors is up to date
• 🛠 All checks below this pull request were successful

---

To help with the number pull requests:

• 🙏 We kindly ask you to review two other open pull requests in this repository

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.52%. Comparing base (57d1d05) to head (86144a5).
Report is 36 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2445   +/-   ##
=======================================
  Coverage   94.52%   94.52%           
=======================================
  Files         246      246           
  Lines       14036    14036           
=======================================
  Hits        13267    13267           
  Misses        769      769           

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

[valeriupredoi]

Cheers, Manu - small change (if it's >=8.6, it'll never be 8.1 😁 ) but, more importantly, has Bouwe's question been answered?

[valeriupredoi]

👌 🍺

[valeriupredoi]

@schlunma I'd argue this would be nice to have in v2.11 if @chrisbillowsMO is not already too fed up with accepting any more bugfixes 🍺

[valeriupredoi]

OK we have a problem with the environment for Python 3.9 - they require Python >=3.10 so our envs don't solve https://github.com/ESMValGroup/ESMValCore/actions/runs/9350804150/job/25735049553 - we either retire support for Python 3.9 as per @bouweandela 's #2406 now or we don't include this in the release and drop 3.9 the sooner the better - I am leaning towards not dropping 3.9 before the release (and hence, not including this in the release)

[ehogan]

OK we have a problem with the environment for Python 3.9 - they require Python >=3.10 so our envs don't solve https://github.com/ESMValGroup/ESMValCore/actions/runs/9350804150/job/25735049553 - we either retire support for Python 3.9 as per @bouweandela 's #2406 now or we don't include this in the release and drop 3.9 the sooner the better - I am leaning towards not dropping 3.9 before the release (and hence, not including this in the release)

Unfortunately, esmf-related pins were updated in ESMValTool via ESMValGroup/ESMValTool#3643 already, which means it's not possible to solve an environment using the latest version of ESMValTool (on main) and the ESMValCore release branch, since ESMValTool has esmpy >=8.6.0 and the ESMValCore release branch has esmpy !=8.1.0,<8.6.0 😮

So either we should just include this in the release, or I will need to create an ESMValTool release branch from the commit before ESMValGroup/ESMValTool#3643 (from 3 weeks ago!) and cherry pick everything after that into the release branch 😭

What would be best? 🤔

[valeriupredoi]

@ehogan that pin needs to be in the ESMValCore release to allow for uniformity with ESMValTool; having said that, that exact pin creates incompatibility between current ESMValTool and Python=3.12 via NCL not being able to support ESMPy/esmf>=8.6.0, something that will take a while to resolve at upstream (NCL), so if that pin is not absolutely necessary, I'd be very happy to see it dropped - I care more for Python 3.12 support than some sciencey thingie 😆

[bouweandela]

How about you create the ESMValTool release branch from main and then revert the changes introduced in ESMValGroup/ESMValTool#3643 only in the ESMValTool release branch?

[bouweandela]

That would resemble the conda environment that you did all the tests with best, right?

[ehogan]

How about you create the ESMValTool release branch from main and then revert the changes introduced in ESMValGroup/ESMValTool#3643 only in the ESMValTool release branch?

Ah, yes, this sounds good, I will try this 👍

[valeriupredoi]

How about you create the ESMValTool release branch from main and then revert the changes introduced in ESMValGroup/ESMValTool#3643 only in the ESMValTool release branch?

and make sure you don't include the pin in the conda package via recipe.meta 👍

[schlunma]

If the change in iris-esmf-regrid is fully backwards-compatible, we could just relax the pin to esmpy!=8.1.0, but we still don't know this yet 🤷

[valeriupredoi]

If the change in iris-esmf-regrid is fully backwards-compatible, we could just relax the pin to esmpy!=8.1.0, but we still don't know this yet 🤷

aha! Good catch, Manu! I asked the same q myself - I'd go along the "what's the worst it can happen" line, and stash the pin for now, hopefully they'll give us an answer, and by then NCL will have had enough time to support the new ESMPy 🤞