1 Release published by 1 person

• v2025.01.2

  published Jan 31, 2025

44 Pull requests merged by 22 people

• More precisely type pipe methods

  #10038 merged Feb 19, 2025

• Flexible coordinate transform

  #9543 merged Feb 14, 2025

• Add types stubs to optional dependencies

  #10048 merged Feb 13, 2025

• Add Coordinates.from_xindex method (+ refactor API doc)

  #10000 merged Feb 12, 2025

• run CI on python=3.13

  #9681 merged Feb 12, 2025

• implement map_over_datasets kwargs

  #10012 merged Feb 10, 2025

• Upgrade mypy to 1.15

  #10041 merged Feb 10, 2025

• Start splitting up dataset.py

  #10039 merged Feb 10, 2025

• Fix dataarray drop attrs

  #10030 merged Feb 9, 2025

• use mean of min/max years as offset in calculation of datetime64 mean

  #10035 merged Feb 7, 2025

• DOC: Fix 404 on Cubed's documentation

  #10029 merged Feb 6, 2025

• Use resolution-dependent default units for lazy time encoding

  #10017 merged Feb 5, 2025

• map_over_datasets: fix error message for wrong result type

  #10016 merged Feb 5, 2025

• Duck array ops for all and any

  #9883 merged Feb 4, 2025

• spelling fix: possibilites -> possibilities

  #10023 merged Feb 4, 2025

• Update pre-commit hooks

  #10021 merged Feb 3, 2025

• add new section in whats-new.rst

  #10011 merged Jan 31, 2025

• release notes: 2025.01.2

  #10007 merged Jan 31, 2025

• Add FAQ answer about API stability & backwards compatibility

  #9855 merged Jan 30, 2025

• Preserve order of variables in combine_by_coords

  #9070 merged Jan 30, 2025

• remove outdated type-ignores

  #10006 merged Jan 30, 2025

• cast numpy scalars to arrays in NamedArray.from_array

  #10008 merged Jan 30, 2025

• Use freq="D" instead of freq="d" in pd.timedelta_range

  #10004 merged Jan 30, 2025

• Fix the push method when the limit parameter is bigger than the chunk…

  #9940 merged Jan 29, 2025

• Migrate Zarr region="auto" tests to a class

  #9990 merged Jan 29, 2025

• Fix infer_freq, check for subdtype "datetime64"/"timedelta64"

  #9999 merged Jan 29, 2025

• Revert "Use flox for grouped first, last (#9986)"

  #10001 merged Jan 29, 2025

• Enable passing a CFTimedeltaCoder to decode_timedelta

  #9966 merged Jan 29, 2025

• fix mean for datetime-like using the respective time resolution unit

  #9977 merged Jan 29, 2025

• Add time_unit argument to CFTimeIndex.to_datetimeindex

  #9965 merged Jan 29, 2025

• Enable DataTree.to_zarr(compute=False)

  #9958 merged Jan 28, 2025

• Remove repetitive that (replace it with the)

  #9994 merged Jan 28, 2025

• Add shxarray to the xarray ecosystem list

  #9995 merged Jan 28, 2025

• Add shards to valid_encodings to enable sharded Zarr writing

  #9948 merged Jan 27, 2025

• Use flox for grouped first, last

  #9986 merged Jan 27, 2025

• Bump the actions group with 2 updates

  #9989 merged Jan 27, 2025

• Fix some typing

  #9988 merged Jan 27, 2025

• Remove unnecessary a article

  #9980 merged Jan 23, 2025

• Fix test_doc_example on big-endian systems

  #9949 merged Jan 23, 2025

• fix weighted polyfit for arrays with more than 2 dimensions

  #9974 merged Jan 23, 2025

• Use zarr-fixture to prevent thread leakage errors

  #9967 merged Jan 22, 2025

• remove dask-expr from CI runs, fix related tests

  #9971 merged Jan 22, 2025

• Update time coding tests to assert exact equality

  #9961 merged Jan 22, 2025

• cast type to PDDatetimeUnitOptions

  #9963 merged Jan 20, 2025

11 Pull requests opened by 10 people

• Provide Variable initialization from `np.full` for Datasets

  #9976 opened Jan 23, 2025

• include accessors in `__dir__`

  #9985 opened Jan 25, 2025

• Refactoring/fixing zarr-pyhton v3 incompatibilities in xarray datatrees

  #10020 opened Feb 3, 2025

• deprecate cftime_range() in favor of date_range(use_cftime=True)

  #10024 opened Feb 4, 2025

• Ensure KeyError raised for zarr datasets missing dim names

  #10025 opened Feb 5, 2025

• Add typos check to pre-commit hooks

  #10040 opened Feb 9, 2025

• map_over_datasets: skip empty nodes

  #10042 opened Feb 10, 2025

• Fix issues when partially coding datetimes/timedeltas

  #10050 opened Feb 14, 2025

• [DNM] GeoTransformIndex prototype

  #10055 opened Feb 17, 2025

• Default to phony_dims="access" in h5netcdf-backend

  #10058 opened Feb 18, 2025

• New defaults for `concat`, `merge`, `combine_*`

  #10062 opened Feb 19, 2025

26 Issues closed by 5 people

• Improve type signatures of `pipe` methods to enable type checkers to flag erroneous usages

  #9997 closed Feb 19, 2025

• Add python 3.13 to CI

  #9664 closed Feb 12, 2025

• allow `kwargs` in `map_over_datasets`?

  #10009 closed Feb 10, 2025

• DataArray().drop_attrs(deep=False) doesn't do anything

  #10027 closed Feb 9, 2025

• Taking the mean of a long-spanning `np.datetime64` array produces the wrong value

  #10019 closed Feb 7, 2025

• Inconsistent File Hashes When Resaving NetCDF Files with Chunks

  #10028 closed Feb 6, 2025

• Behaviour from `Dataset.broadcast_like` is strange and inconsistent with how arithmetic ops on Datasets actually broadcast

  #10031 closed Feb 6, 2025

• xarray mean and numpy mean yield very different result since v2025.1.0

  #10026 closed Feb 5, 2025

• Some operations broken when using the SciPy engine

  #9824 closed Jan 31, 2025

• Release 2025.01.2

  #10002 closed Jan 31, 2025

• Add FAQ answer about API stability / backwards compatibility?

  #9854 closed Jan 30, 2025

• Dataset combine_by_coords unexpected behavior

  #8828 closed Jan 30, 2025

• Doctest failing with numpy 2.1.3

  #10005 closed Jan 30, 2025

• ⚠️ Nightly upstream-dev CI failed ⚠️

  #10003 closed Jan 30, 2025

• Push method with limit bigger than the chunksize of Dask array is not correctly working

  #9939 closed Jan 29, 2025

• infer_freq not working (yet) with datetime64[s]

  #9998 closed Jan 29, 2025

• ⚠️ Nightly upstream-dev CI failed ⚠️

  #9993 closed Jan 29, 2025

• Averaging timestamps with non-nanosecond precision

  #9975 closed Jan 29, 2025

• Issue on page /generated/xarray.DataArray.differentiate.html

  #9996 closed Jan 29, 2025

• New shxarray accessor for working with spherical harmonic data

  #9992 closed Jan 28, 2025

• Unable to write to sharded Zarr V3 because `shards` not in `valid_encodings`

  #9947 closed Jan 27, 2025

• value error reading era6 netcdf file

  #9991 closed Jan 27, 2025

• `dataset.nbytes` not approximating correctly the size?

  #9979 closed Jan 27, 2025

• Slow looping & indexing: need direct access to underlying zarr store

  #9973 closed Jan 24, 2025

• Weighted polyfit is broken for arrays with more than two dimensions

  #9972 closed Jan 23, 2025

• dask-expr got merged into dask

  #9962 closed Jan 22, 2025

24 Issues opened by 20 people

• Saving an extracted netCDF file is far too time consuming

  #10061 opened Feb 19, 2025

• Broken support for `write_empty_chunks` with Zarr v3

  #10056 opened Feb 17, 2025

• reindex is very slow with small chunksizes

  #10054 opened Feb 16, 2025

• map_over_datasets - What is the required output of func?

  #10053 opened Feb 16, 2025

• DataTree html repr is very slow

  #10052 opened Feb 16, 2025

• phony_dims must be specified when opening HDF5 file without dimension scales

  #10049 opened Feb 14, 2025

• Equality between datetime64[s] and datetime64[ns] is not consistent for coordinates vs data variables

  #10045 opened Feb 12, 2025

• Allow Dataset in numpy array with dtype=object

  #10044 opened Feb 12, 2025

• inconsticency between `xr.open_dataset` and `xr.open_datatree` for `decode_cf` argument

  #10037 opened Feb 7, 2025

• `NamedArray` typing audit

  #10036 opened Feb 7, 2025

• Unable to write into existing Zarr archive due to mismatching dask chunks

  #10034 opened Feb 7, 2025

• Add a new broadcasting mode that inserts size-1 dimensions.

  #10033 opened Feb 6, 2025

• TypeError: Expected a BytesBytesCodec. Got <class 'numcodecs.blosc.Blosc'> instead.

  #10032 opened Feb 6, 2025

• DataArray.isin does not accept sets intuitively

  #10022 opened Feb 4, 2025

• `DataTree`: missing methods

  #10015 opened Jan 31, 2025

• DataTree: selecting via `[["a"]]` suggests `subset`, which is missing

  #10014 opened Jan 31, 2025

• datatree gets dis-aligned in binary op

  #10013 opened Jan 31, 2025

• Improved docs for zarr encoding options.

  #9987 opened Jan 26, 2025

• DataTree + Zarr-Python 3

  #9984 opened Jan 25, 2025

• Apparent cache refresh bug when a FileExistsError is triggered

  #9982 opened Jan 24, 2025

• High unmanaged memory using kwargs in apply_ufunc with Dask

  #9981 opened Jan 23, 2025

• Using a forward slash in a node names of a `DataTree` causes a `RecursionError`

  #9978 opened Jan 23, 2025

• ValueError raised when opening a 'plain' zarr with xarray+zarr 3

  #9970 opened Jan 21, 2025

• Stacking doesn't work on dims named `""`

  #9969 opened Jan 21, 2025

36 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• (fix): extension array indexers

  #9671 commented on Feb 7, 2025 • 18 new comments

• Avoid unsafe casts from float to unsigned int

  #9964 commented on Jan 30, 2025 • 5 new comments

• Fill gaps limited 7665

  #9402 commented on Feb 5, 2025 • 4 new comments

• ddof vs correction kwargs in std/var

  #8573 commented on Jan 25, 2025 • 1 new comment

• Thread-safety in xarray

  #9836 commented on Feb 12, 2025 • 0 new comments

• Add `nunique`

  #9548 commented on Feb 13, 2025 • 0 new comments

• netCDF encoding and decoding issues.

  #8957 commented on Feb 13, 2025 • 0 new comments

• [Bug]: cannot chunk a DataArray that originated as a coordinate

  #6204 commented on Feb 13, 2025 • 0 new comments

• Stricter defaults for concat, combine, open_mfdataset, merge

  #8778 commented on Feb 14, 2025 • 0 new comments

• Test failure on RISC-V platform

  #9815 commented on Feb 16, 2025 • 0 new comments

• Adding CDL Parser/`open_cdl`?

  #6269 commented on Feb 17, 2025 • 0 new comments

• Allow DataArray.to_series() without invoking sparse.COO.todense()

  #4007 commented on Feb 17, 2025 • 0 new comments

• Add support for Array API in NamedArray

  #8344 commented on Feb 6, 2025 • 0 new comments

• Migrate indexing and broadcasting logic to `xarray.namedarray` (Part 1)

  #8856 commented on Jan 30, 2025 • 0 new comments

• Using oindex and vindex instead of ExplicitIndexer objects.

  #9273 commented on Jan 30, 2025 • 0 new comments

• Use np.real and np.imag when possible

  #9368 commented on Feb 7, 2025 • 0 new comments

• Array conversion methods

  #9823 commented on Feb 12, 2025 • 0 new comments

• Open mfdataset enchancement

  #9955 commented on Feb 16, 2025 • 0 new comments

• weighted polyfit with NaNs appears broken

  #9048 commented on Jan 22, 2025 • 0 new comments

• Creation of empty DataArrays in a Dataset

  #9468 commented on Jan 23, 2025 • 0 new comments

• Matrix multiplication inconsistent for dask and sparse: `dask @ sparse` works, `sparse @ dask` fails

  #9934 commented on Jan 23, 2025 • 0 new comments

• DataTree roundtrip fails on None group lookup

  #9960 commented on Jan 25, 2025 • 0 new comments

• Stateful user-defined accessors

  #3268 commented on Jan 25, 2025 • 0 new comments

• The method to_netcdf does not preserve chunks

  #8385 commented on Jan 27, 2025 • 0 new comments

• ScipyArrayWrapper' object has no attribute 'oindex'

  #8909 commented on Jan 28, 2025 • 0 new comments

• Support proper custom class reflexive operator applied to xarray objects

  #9944 commented on Jan 29, 2025 • 0 new comments

• Could we defer to flox for `GroupBy.first`?

  #9647 commented on Jan 29, 2025 • 0 new comments

• Automatic Dask-Zarr chunk alignment on the to_zarr method

  #9914 commented on Jan 29, 2025 • 0 new comments

• Design for IntervalIndex

  #8005 commented on Feb 3, 2025 • 0 new comments

• Time dtype encoding defaulting to `int64` when writing netcdf or zarr

  #3942 commented on Feb 5, 2025 • 0 new comments

• `Dataset.broadcast_like(other)` should broadcast against like variables in other

  #6549 commented on Feb 6, 2025 • 0 new comments

• "TypeError: Unsupported type for store_like: 'FSMap'" with new version libraries

  #9957 commented on Feb 7, 2025 • 0 new comments

• Deprecate `xr.cftime_range` in favor of `xr.date_range`

  #9886 commented on Feb 8, 2025 • 0 new comments

• map_over_datasets throws error on nodes without datasets

  #9693 commented on Feb 10, 2025 • 0 new comments

• Failure in pandas TestDataFrameToXArray.test_to_xarray_index_types

  #9661 commented on Feb 11, 2025 • 0 new comments

• Pandas Integer Type Doesn't Convert in Dataset

  #9742 commented on Feb 11, 2025 • 0 new comments