split out CFDatetimeCoder, deprecate use_cftime as kwarg (#9901)

* split out CFDatetimeCoder into coders, deprecate use_cftime as keyword argument * add whats-new.rst entry * Apply suggestions from code review Co-authored-by: Deepak Cherian <dcherian@users.noreply.github.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix warning * fix docstrings * try fix typing * Apply suggestions from code review Co-authored-by: Spencer Clark <spencerkclark@gmail.com> * Apply suggestions from code review Co-authored-by: Spencer Clark <spencerkclark@gmail.com> * Update xarray/conventions.py * Update deprecated directive * fix docstrings/whats-new.rst after merge * fix whats-new.rst * update warnings/errors --------- Co-authored-by: Deepak Cherian <dcherian@users.noreply.github.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Spencer Clark <spencerkclark@gmail.com>
pydata · Jan 4, 2025 · 1486bea · 1486bea
1 parent ff7d157
commit 1486bea
Show file tree

Hide file tree

Showing 12 changed files with 190 additions and 52 deletions.
diff --git a/doc/api.rst b/doc/api.rst
@@ -1096,6 +1096,17 @@ DataTree methods
 ..    Missing:
 ..    ``open_mfdatatree``
 
+Encoding/Decoding
+=================
+
+Coder objects
+-------------
+
+.. autosummary::
+   :toctree: generated/
+
+   coders.CFDatetimeCoder
+
 Coordinates objects
 ===================
 

diff --git a/doc/whats-new.rst b/doc/whats-new.rst
@@ -22,15 +22,19 @@ v2025.01.1 (unreleased)
 
 New Features
 ~~~~~~~~~~~~
-
+- Split out :py:class:`coders.CFDatetimeCoder` as public API in ``xr.coders``, make ``decode_times`` keyword argument
+  consume :py:class:`coders.CFDatetimeCoder` (:pull:`9901`).
+  By `Kai Mühlbauer <https://github.com/kmuehlbauer>`_.
 
 Breaking changes
 ~~~~~~~~~~~~~~~~
 
 
 Deprecations
 ~~~~~~~~~~~~
-
+- Time decoding related kwarg ``use_cftime`` is deprecated. Use keyword argument
+  ``decode_times=CFDatetimeCoder(use_cftime=True)`` in :py:func:`~xarray.open_dataset`, :py:func:`~xarray.open_dataarray`, :py:func:`~xarray.open_datatree`, :py:func:`~xarray.open_groups`, :py:func:`~xarray.open_zarr` and :py:func:`~xarray.decode_cf` instead (:pull:`9901`).
+  By `Kai Mühlbauer <https://github.com/kmuehlbauer>`_.
 
 Bug fixes
 ~~~~~~~~~
@@ -70,6 +74,7 @@ New Features
   iso8601-parser (:pull:`9885`).
   By `Kai Mühlbauer <https://github.com/kmuehlbauer>`_.
 
+
 Breaking changes
 ~~~~~~~~~~~~~~~~
 - Methods including ``dropna``, ``rank``, ``idxmax``, ``idxmin`` require

diff --git a/xarray/__init__.py b/xarray/__init__.py
@@ -1,6 +1,6 @@
 from importlib.metadata import version as _version
 
-from xarray import groupers, testing, tutorial, ufuncs
+from xarray import coders, groupers, testing, tutorial, ufuncs
 from xarray.backends.api import (
     load_dataarray,
     load_dataset,
@@ -66,6 +66,7 @@
 # `mypy --strict` running in projects that import xarray.
 __all__ = (  # noqa: RUF022
     # Sub-packages
+    "coders",
     "groupers",
     "testing",
     "tutorial",

diff --git a/xarray/backends/api.py b/xarray/backends/api.py
@@ -33,6 +33,7 @@
     _normalize_path,
 )
 from xarray.backends.locks import _get_scheduler
+from xarray.coders import CFDatetimeCoder
 from xarray.core import indexing
 from xarray.core.combine import (
     _infer_concat_order_from_positions,
@@ -481,7 +482,10 @@ def open_dataset(
     cache: bool | None = None,
     decode_cf: bool | None = None,
     mask_and_scale: bool | Mapping[str, bool] | None = None,
-    decode_times: bool | Mapping[str, bool] | None = None,
+    decode_times: bool
+    | CFDatetimeCoder
+    | Mapping[str, bool | CFDatetimeCoder]
+    | None = None,
     decode_timedelta: bool | Mapping[str, bool] | None = None,
     use_cftime: bool | Mapping[str, bool] | None = None,
     concat_characters: bool | Mapping[str, bool] | None = None,
@@ -543,9 +547,10 @@ def open_dataset(
         be replaced by NA. Pass a mapping, e.g. ``{"my_variable": False}``,
         to toggle this feature per-variable individually.
         This keyword may not be supported by all the backends.
-    decode_times : bool or dict-like, optional
+    decode_times : bool, CFDatetimeCoder or dict-like, optional
         If True, decode times encoded in the standard NetCDF datetime format
-        into datetime objects. Otherwise, leave them encoded as numbers.
+        into datetime objects. Otherwise, use :py:class:`coders.CFDatetimeCoder` or leave them
+        encoded as numbers.
         Pass a mapping, e.g. ``{"my_variable": False}``,
         to toggle this feature per-variable individually.
         This keyword may not be supported by all the backends.
@@ -569,6 +574,10 @@ def open_dataset(
         raise an error. Pass a mapping, e.g. ``{"my_variable": False}``,
         to toggle this feature per-variable individually.
         This keyword may not be supported by all the backends.
+
+        .. deprecated:: 2025.01.1
+           Please pass a :py:class:`coders.CFDatetimeCoder` instance initialized with ``use_cftime`` to the ``decode_times`` kwarg instead.
+
     concat_characters : bool or dict-like, optional
         If True, concatenate along the last dimension of character arrays to
         form string arrays. Dimensions will only be concatenated over (and
@@ -698,7 +707,10 @@ def open_dataarray(
     cache: bool | None = None,
     decode_cf: bool | None = None,
     mask_and_scale: bool | None = None,
-    decode_times: bool | None = None,
+    decode_times: bool
+    | CFDatetimeCoder
+    | Mapping[str, bool | CFDatetimeCoder]
+    | None = None,
     decode_timedelta: bool | None = None,
     use_cftime: bool | None = None,
     concat_characters: bool | None = None,
@@ -761,9 +773,11 @@ def open_dataarray(
         `missing_value` attribute contains multiple values a warning will be
         issued and all array values matching one of the multiple values will
         be replaced by NA. This keyword may not be supported by all the backends.
-    decode_times : bool, optional
+    decode_times : bool, CFDatetimeCoder or dict-like, optional
         If True, decode times encoded in the standard NetCDF datetime format
-        into datetime objects. Otherwise, leave them encoded as numbers.
+        into datetime objects. Otherwise, use :py:class:`coders.CFDatetimeCoder` or leave them encoded as numbers.
+        Pass a mapping, e.g. ``{"my_variable": False}``,
+        to toggle this feature per-variable individually.
         This keyword may not be supported by all the backends.
     decode_timedelta : bool, optional
         If True, decode variables and coordinates with time units in
@@ -781,6 +795,10 @@ def open_dataarray(
         represented using ``np.datetime64[ns]`` objects.  If False, always
         decode times to ``np.datetime64[ns]`` objects; if this is not possible
         raise an error. This keyword may not be supported by all the backends.
+
+        .. deprecated:: 2025.01.1
+           Please pass a :py:class:`coders.CFDatetimeCoder` instance initialized with ``use_cftime`` to the ``decode_times`` kwarg instead.
+
     concat_characters : bool, optional
         If True, concatenate along the last dimension of character arrays to
         form string arrays. Dimensions will only be concatenated over (and
@@ -903,7 +921,10 @@ def open_datatree(
     cache: bool | None = None,
     decode_cf: bool | None = None,
     mask_and_scale: bool | Mapping[str, bool] | None = None,
-    decode_times: bool | Mapping[str, bool] | None = None,
+    decode_times: bool
+    | CFDatetimeCoder
+    | Mapping[str, bool | CFDatetimeCoder]
+    | None = None,
     decode_timedelta: bool | Mapping[str, bool] | None = None,
     use_cftime: bool | Mapping[str, bool] | None = None,
     concat_characters: bool | Mapping[str, bool] | None = None,
@@ -961,9 +982,9 @@ def open_datatree(
         be replaced by NA. Pass a mapping, e.g. ``{"my_variable": False}``,
         to toggle this feature per-variable individually.
         This keyword may not be supported by all the backends.
-    decode_times : bool or dict-like, optional
+    decode_times : bool, CFDatetimeCoder or dict-like, optional
         If True, decode times encoded in the standard NetCDF datetime format
-        into datetime objects. Otherwise, leave them encoded as numbers.
+        into datetime objects. Otherwise, use :py:class:`coders.CFDatetimeCoder` or leave them encoded as numbers.
         Pass a mapping, e.g. ``{"my_variable": False}``,
         to toggle this feature per-variable individually.
         This keyword may not be supported by all the backends.
@@ -987,6 +1008,10 @@ def open_datatree(
         raise an error. Pass a mapping, e.g. ``{"my_variable": False}``,
         to toggle this feature per-variable individually.
         This keyword may not be supported by all the backends.
+
+        .. deprecated:: 2025.01.1
+           Please pass a :py:class:`coders.CFDatetimeCoder` instance initialized with ``use_cftime`` to the ``decode_times`` kwarg instead.
+
     concat_characters : bool or dict-like, optional
         If True, concatenate along the last dimension of character arrays to
         form string arrays. Dimensions will only be concatenated over (and
@@ -1118,7 +1143,10 @@ def open_groups(
     cache: bool | None = None,
     decode_cf: bool | None = None,
     mask_and_scale: bool | Mapping[str, bool] | None = None,
-    decode_times: bool | Mapping[str, bool] | None = None,
+    decode_times: bool
+    | CFDatetimeCoder
+    | Mapping[str, bool | CFDatetimeCoder]
+    | None = None,
     decode_timedelta: bool | Mapping[str, bool] | None = None,
     use_cftime: bool | Mapping[str, bool] | None = None,
     concat_characters: bool | Mapping[str, bool] | None = None,
@@ -1180,9 +1208,9 @@ def open_groups(
         be replaced by NA. Pass a mapping, e.g. ``{"my_variable": False}``,
         to toggle this feature per-variable individually.
         This keyword may not be supported by all the backends.
-    decode_times : bool or dict-like, optional
+    decode_times : bool, CFDatetimeCoder or dict-like, optional
         If True, decode times encoded in the standard NetCDF datetime format
-        into datetime objects. Otherwise, leave them encoded as numbers.
+        into datetime objects. Otherwise, use :py:class:`coders.CFDatetimeCoder` or leave them encoded as numbers.
         Pass a mapping, e.g. ``{"my_variable": False}``,
         to toggle this feature per-variable individually.
         This keyword may not be supported by all the backends.
@@ -1206,6 +1234,10 @@ def open_groups(
         raise an error. Pass a mapping, e.g. ``{"my_variable": False}``,
         to toggle this feature per-variable individually.
         This keyword may not be supported by all the backends.
+
+        .. deprecated:: 2025.01.1
+           Please pass a :py:class:`coders.CFDatetimeCoder` instance initialized with ``use_cftime`` to the ``decode_times`` kwarg instead.
+
     concat_characters : bool or dict-like, optional
         If True, concatenate along the last dimension of character arrays to
         form string arrays. Dimensions will only be concatenated over (and

diff --git a/xarray/coders.py b/xarray/coders.py
@@ -0,0 +1,10 @@
+"""
+This module provides coder objects that encapsulate the
+"encoding/decoding" process.
+"""
+
+from xarray.coding.times import CFDatetimeCoder
+
+__all__ = [
+    "CFDatetimeCoder",
+]
diff --git a/xarray/coding/times.py b/xarray/coding/times.py
@@ -286,7 +286,10 @@ def _unpack_time_unit_and_ref_date(
 
 
 def _decode_cf_datetime_dtype(
-    data, units: str, calendar: str | None, use_cftime: bool | None
+    data,
+    units: str,
+    calendar: str | None,
+    use_cftime: bool | None,
 ) -> np.dtype:
     # Verify that at least the first and last date can be decoded
     # successfully. Otherwise, tracebacks end up swallowed by
@@ -421,7 +424,10 @@ def _decode_datetime_with_pandas(
 
 
 def decode_cf_datetime(
-    num_dates, units: str, calendar: str | None = None, use_cftime: bool | None = None
+    num_dates,
+    units: str,
+    calendar: str | None = None,
+    use_cftime: bool | None = None,
 ) -> np.ndarray:
     """Given an array of numeric dates in netCDF format, convert it into a
     numpy array of date time objects.
@@ -1093,7 +1099,10 @@ def _lazily_encode_cf_timedelta(
 
 
 class CFDatetimeCoder(VariableCoder):
-    def __init__(self, use_cftime: bool | None = None) -> None:
+    def __init__(
+        self,
+        use_cftime: bool | None = None,
+    ) -> None:
         self.use_cftime = use_cftime
 
     def encode(self, variable: Variable, name: T_Name = None) -> Variable: