fix regrid for snakemake

juliettelavoie commented 2 weeks ago

Pull Request Checklist:

[x] This PR addresses an already opened issue (for bug fixes / features)
- This PR fixes #399
- This PR fixes #411
[ ] (If applicable) Documentation has been added / updated (for bug fixes / features).
[x] (If applicable) Tests have been added.
[x] This PR does not seem to break the templates.
[x] CHANGELOG.rst has been updated (with summary of main changes).
- [x] Link to issue (:issue:number) and pull request (:pull:number) has been added.

What kind of change does this PR introduce?

write regrid weights in file with domain name

Does this PR introduce a breaking change?

no

Other information:

Previously, we would only do one regrid at the time. With snakemake, they can be done synchronously. This means we need the name of the domain in the name of the file, or different task try to write on the same file.

PVI: @oumoukal @aulemahal

RondeauG commented 2 weeks ago

Could you also include the fix to #411? The fix was given here: https://github.com/Ouranosinc/xscen/commit/f12895a084d8d3521e8c721c7c2a37955fa59c07

RondeauG commented 2 weeks ago

Could you also add a <0.51 to xclim, to help with #399 and future breaking changes?

juliettelavoie commented 2 weeks ago

I pinned xarray <2024.06.0. I think the failing tests in adjust were related to https://github.com/pydata/xarray/issues/9138.

juliettelavoie commented 2 weeks ago

@Zeitsperre do you have an idea what is the issue for py310-covervalls ?

juliettelavoie commented 2 weeks ago

@Zeitsperre We managed to fix the issue in the coveralls tests by pinning h5py and netcdf4. Would still love your input if you know the cause of the problem!

Ouranosinc / xscen