Closed seisman closed 1 month ago
Following pydata/xarray#7716 (comment), xarray 2023.04 works with pandas 2.0. Then we must decide whether to merge this PR now (violating the SPEC0 policy) or wait until April 2025
Yeah, I think we'll need to bump the minimum required xarray
version more than needed this time (to 2023.03
) to work with pandas=2.0
, considering that conda-forge did a repodata-patch for this at https://github.com/conda-forge/conda-forge-repodata-patches-feedstock/pull/426#issuecomment-1496389758. Note that SPEC 0 does say the following:
... projects may occasionally want to drop support for core package dependencies earlier than recommended by this SPEC. For instance, if a newer minimum version of a core package is needed by a project due to a critical bug fix, which is not backported to older versions.
I would consider that xarray 2023.03 supporting pandas 2.0 is a critical bug fix :slightly_smiling_face:, so would prefer to get this into PyGMT v0.14.0.
I'm OK with making it to v0.14.0
Need to update
doc/minversions.rst
too.
No need to update this file after https://github.com/GenericMappingTools/pygmt/pull/3380.
Can we also agree on a consistent title for SPEC 0 updates? We've been inconsistent in https://www.pygmt.org/v0.13.0/changes.html:
- SPEC 0: Bump minimum supported version to xarray 2022.09
- SPEC 0: Set minimum supported versions to Python>=3.10, pandas>=1.5 and xarray>=2022.06
- NEP29: Set minimum required version to NumPy 1.23+
I like the middle one using
pandas>=1.5
, but no strong opinions.
Sounds good
Description of proposed changes
Following SPEC0, we should drop pandas 1.5 and xarray 2022.09 support in 2024 Q3, but now we have dependency conflicts.
Following https://github.com/pydata/xarray/issues/7716#issuecomment-1510434058, xarray 2023.04 works with pandas 2.0. Then we must decide whether to merge this PR now (violating the SPEC0 policy) or wait until April 2025.
I prefer to wait until April 2025, so adding this PR to milestone v0.16.0.
Previous pandas version bump at #3043.