Merge remote-tracking branch 'upstream/main' into feature/fix-interp-docs

Fixes merge with pydata#6637

Author: Louis Stenger

.gitignore

@@ -1,5 +1,7 @@
 *.py[cod]
 __pycache__
+.env
+.venv
 
 # example caches from Hypothesis
 .hypothesis/

.pre-commit-config.yaml

@@ -46,7 +46,7 @@ repos:
   #     - id: velin
   #       args: ["--write", "--compact"]
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v0.950
+    rev: v0.960
     hooks:
       - id: mypy
         # Copied from setup.cfg

HOW_TO_RELEASE.md

@@ -111,4 +111,4 @@ upstream        https://github.com/pydata/xarray (push)
 
 As of 2022.03.0, we utilize the [CALVER](https://calver.org/) version system.
 Specifically, we have adopted the pattern `YYYY.MM.X`, where `YYYY` is a 4-digit
-year (e.g. `2022`), `MM` is a 2-digit zero-padded month (e.g. `01` for January), and `X` is the release number (starting at zero at the start of each month and incremented once for each additional release).
+year (e.g. `2022`), `0M` is a 2-digit zero-padded month (e.g. `01` for January), and `X` is the release number (starting at zero at the start of each month and incremented once for each additional release).

xarray/backends/api.py

@@ -56,6 +56,8 @@
         T_NetcdfEngine,
         Literal["pydap", "pynio", "pseudonetcdf", "cfgrib", "zarr"],
         Type[BackendEntrypoint],
+        str,  # no nice typing support for custom backends
+        None,
     ]
     T_Chunks = Union[int, dict[Any, Any], Literal["auto"], None]
     T_NetcdfTypes = Literal[
@@ -392,7 +394,8 @@ def open_dataset(
         scipy.io.netcdf (only netCDF3 supported). Byte-strings or file-like
         objects are opened by scipy.io.netcdf (netCDF3) or h5py (netCDF4/HDF).
     engine : {"netcdf4", "scipy", "pydap", "h5netcdf", "pynio", "cfgrib", \
-        "pseudonetcdf", "zarr"} or subclass of xarray.backends.BackendEntrypoint, optional
+        "pseudonetcdf", "zarr", None}, installed backend \
+        or subclass of xarray.backends.BackendEntrypoint, optional
         Engine to use when reading files. If not provided, the default engine
         is chosen based on available dependencies, with a preference for
         "netcdf4". A custom backend class (a subclass of ``BackendEntrypoint``)
@@ -579,7 +582,8 @@ def open_dataarray(
         scipy.io.netcdf (only netCDF3 supported). Byte-strings or file-like
         objects are opened by scipy.io.netcdf (netCDF3) or h5py (netCDF4/HDF).
     engine : {"netcdf4", "scipy", "pydap", "h5netcdf", "pynio", "cfgrib", \
-        "pseudonetcdf", "zarr"}, optional
+        "pseudonetcdf", "zarr", None}, installed backend \
+        or subclass of xarray.backends.BackendEntrypoint, optional
         Engine to use when reading files. If not provided, the default engine
         is chosen based on available dependencies, with a preference for
         "netcdf4".
@@ -804,8 +808,9 @@ def open_mfdataset(
         If provided, call this function on each dataset prior to concatenation.
         You can find the file-name from which each dataset was loaded in
         ``ds.encoding["source"]``.
-    engine : {"netcdf4", "scipy", "pydap", "h5netcdf", "pynio", "cfgrib", "zarr"}, \
-        optional
+    engine : {"netcdf4", "scipy", "pydap", "h5netcdf", "pynio", "cfgrib", \
+        "pseudonetcdf", "zarr", None}, installed backend \
+        or subclass of xarray.backends.BackendEntrypoint, optional
         Engine to use when reading files. If not provided, the default engine
         is chosen based on available dependencies, with a preference for
         "netcdf4".

xarray/core/alignment.py

@@ -16,6 +16,7 @@
     Tuple,
     Type,
     TypeVar,
+    cast,
 )
 
 import numpy as np
@@ -30,7 +31,7 @@
 if TYPE_CHECKING:
     from .dataarray import DataArray
     from .dataset import Dataset
-    from .types import JoinOptions
+    from .types import JoinOptions, T_DataArray, T_DataArrayOrSet, T_Dataset
 
 DataAlignable = TypeVar("DataAlignable", bound=DataWithCoords)
 
@@ -559,7 +560,7 @@ def align(self) -> None:
 def align(
     *objects: DataAlignable,
     join: JoinOptions = "inner",
-    copy=True,
+    copy: bool = True,
     indexes=None,
     exclude=frozenset(),
     fill_value=dtypes.NA,
@@ -592,7 +593,7 @@ def align(
           those of the first object with that dimension. Indexes for the same
           dimension must have the same size in all objects.
 
-    copy : bool, optional
+    copy : bool, default: True
         If ``copy=True``, data in the return values is always copied. If
         ``copy=False`` and reindexing is unnecessary, or can be performed with
         only slice operations, then the output may share memory with the input.
@@ -609,7 +610,7 @@ def align(
 
     Returns
     -------
-    aligned : DataArray or Dataset
+    aligned : tuple of DataArray or Dataset
         Tuple of objects with the same type as `*objects` with aligned
         coordinates.
 
@@ -935,7 +936,9 @@ def _get_broadcast_dims_map_common_coords(args, exclude):
     return dims_map, common_coords
 
 
-def _broadcast_helper(arg, exclude, dims_map, common_coords):
+def _broadcast_helper(
+    arg: T_DataArrayOrSet, exclude, dims_map, common_coords
+) -> T_DataArrayOrSet:
 
     from .dataarray import DataArray
     from .dataset import Dataset
@@ -950,22 +953,25 @@ def _set_dims(var):
 
         return var.set_dims(var_dims_map)
 
-    def _broadcast_array(array):
+    def _broadcast_array(array: T_DataArray) -> T_DataArray:
         data = _set_dims(array.variable)
         coords = dict(array.coords)
         coords.update(common_coords)
-        return DataArray(data, coords, data.dims, name=array.name, attrs=array.attrs)
+        return array.__class__(
+            data, coords, data.dims, name=array.name, attrs=array.attrs
+        )
 
-    def _broadcast_dataset(ds):
+    def _broadcast_dataset(ds: T_Dataset) -> T_Dataset:
         data_vars = {k: _set_dims(ds.variables[k]) for k in ds.data_vars}
         coords = dict(ds.coords)
         coords.update(common_coords)
-        return Dataset(data_vars, coords, ds.attrs)
+        return ds.__class__(data_vars, coords, ds.attrs)
 
+    # remove casts once https://github.com/python/mypy/issues/12800 is resolved
     if isinstance(arg, DataArray):
-        return _broadcast_array(arg)
+        return cast("T_DataArrayOrSet", _broadcast_array(arg))
     elif isinstance(arg, Dataset):
-        return _broadcast_dataset(arg)
+        return cast("T_DataArrayOrSet", _broadcast_dataset(arg))
     else:
         raise ValueError("all input must be Dataset or DataArray objects")