Add Ellipsis typehint to reductions

[headtr1ck]

This PR adds the ellipsis typehint to reductions (only where they behave differently from None to reduce overhead).
Follow up on #7017 (comment)

Additionally I was changing a lot of "one or more dimensions" typehints to str | Iterable[Hashable] (See #6142).
Some code changes were necessary to support this fully. Before several things were not working with actual hashable dimensions that are not strings.

[max-sixty]

Excellent @headtr1ck !

Do we need to run pytest --accept to get the docstrings? It looks like we lost lots...

[headtr1ck]

Could any dev that uses linux rerun the generate_reductions and pytest --doctest-modules xarray/core/_reductions.py --accept?
On windows I still get different results (maybe that should be fixed at some point...)

[headtr1ck]

Turns out that the buildin ellipsis works now with mypy.
Did not test it for older python versions, may require some special casing (Lets see if the tests pass)?

[max-sixty]

Here's the diff from pytest-accept (it is weird that it's slightly different on windows...)

commit 83615a94a6b7c0ae0cf0e0240d7705d9ce6c21e5
Author: Maximilian Roos <[email protected]>
Date:   Sat Sep 17 12:36:45 2022 -0700

    pytest-accept

diff --git a/xarray/core/_reductions.py b/xarray/core/_reductions.py
index a7cf7ec2..d0c2a9d7 100644
--- a/xarray/core/_reductions.py
+++ b/xarray/core/_reductions.py
@@ -97,7 +97,7 @@ def count(
         <xarray.Dataset>
         Dimensions:  ()
         Data variables:
-            da       int32 5
+            da       int64 5
         """
         return self.reduce(
             duck_array_ops.count,
@@ -4400,7 +4400,7 @@ def count(
 
         >>> da.groupby("labels").count()
         <xarray.DataArray (labels: 3)>
-        array([1, 2, 2], dtype=int64)
+        array([1, 2, 2])
         Coordinates:
           * labels   (labels) object 'a' 'b' 'c'
         """
@@ -5485,7 +5485,7 @@ def count(
 
         >>> da.resample(time="3M").count()
         <xarray.DataArray (time: 3)>
-        array([1, 3, 1], dtype=int64)
+        array([1, 3, 1])
         Coordinates:
           * time     (time) datetime64[ns] 2001-01-31 2001-04-30 2001-07-31
         """

[Illviljan]

I wonder if Dims should include ellipsis as well? The few times it's missing might be issues with the functions?

[Illviljan]

@@ -4400,7 +4400,7 @@ def count(
 
         >>> da.groupby("labels").count()
         <xarray.DataArray (labels: 3)>
-        array([1, 2, 2], dtype=int64)
+        array([1, 2, 2])
         Coordinates:
           * labels   (labels) object 'a' 'b' 'c'
         """

Is it just me that this example crashes the second time I run it?

import numpy as np
import pandas as pd
import xarray as xr

da = xr.DataArray(
    np.array([1, 2, 3, 1, 2, np.nan]),
    dims="time",
    coords=dict(
        time=("time", pd.date_range("01-01-2001", freq="M", periods=6)),
        labels=("time", np.array(["a", "b", "c", "c", "b", "a"])),
    ),
)
da.groupby("labels").count()

Traceback (most recent call last):

  File "C:\Users\J.W\anaconda3\envs\xarray-tests\lib\site-packages\spyder_kernels\py3compat.py", line 356, in compat_exec
    exec(code, globals, locals)

  File "g:\program\dropbox\python\xarray_groupby_windows_diff.py", line 34, in <module>
    da.groupby("labels").count()

  File "c:\users\j.w\documents\github\xarray\xarray\core\_reductions.py", line 4384, in count
    return self._flox_reduce(

  File "c:\users\j.w\documents\github\xarray\xarray\core\groupby.py", line 738, in _flox_reduce
    result = xarray_reduce(

  File "C:\Users\J.W\anaconda3\envs\xarray-tests\lib\site-packages\flox\xarray.py", line 240, in xarray_reduce
    ds, *by = xr.broadcast(ds, *by, exclude=exclude_dims)

  File "c:\users\j.w\documents\github\xarray\xarray\core\alignment.py", line 1046, in broadcast
    args = align(*args, join="outer", copy=False, exclude=exclude)

  File "c:\users\j.w\documents\github\xarray\xarray\core\alignment.py", line 765, in align
    aligner.align()

  File "c:\users\j.w\documents\github\xarray\xarray\core\alignment.py", line 549, in align
    self.find_matching_indexes()

  File "c:\users\j.w\documents\github\xarray\xarray\core\alignment.py", line 256, in find_matching_indexes
    obj_indexes, obj_index_vars = self._normalize_indexes(obj.xindexes)

  File "c:\users\j.w\documents\github\xarray\xarray\core\alignment.py", line 205, in _normalize_indexes
    pd_idx = safe_cast_to_index(data)

  File "c:\users\j.w\documents\github\xarray\xarray\core\utils.py", line 140, in safe_cast_to_index
    index = pd.Index(np.asarray(array), **kwargs)

  File "C:\Users\J.W\anaconda3\envs\xarray-tests\lib\site-packages\pandas\core\indexes\base.py", line 483, in __new__
    data = sanitize_array(data, None, dtype=dtype, copy=copy)

  File "C:\Users\J.W\anaconda3\envs\xarray-tests\lib\site-packages\pandas\core\construction.py", line 524, in sanitize_array
    raise ValueError("index must be specified when data is not list-like")

ValueError: index must be specified when data is not list-like

[headtr1ck]

Is it just me that this example crashes the second time I run it?

Could you specify what you mean by "second time I run it"?
Executing the groupby twice?

[Illviljan]

Just running that script file several times without restarting the console. It might be a Spyder bug though since I can't reproduce it in a stand alone ipython console.

This for example works (the first time):

import numpy as np
import pandas as pd
import xarray as xr

da = xr.DataArray(
    np.array([1, 2, 3, 1, 2, np.nan]),
    dims="time",
    coords=dict(
        time=("time", pd.date_range("01-01-2001", freq="M", periods=6)),
        labels=("time", np.array(["a", "b", "c", "c", "b", "a"])),
    ),
)
da.groupby("labels").count()

da = xr.DataArray(
    np.array([1, 2, 3, 1, 2, np.nan]),
    dims="time",
    coords=dict(
        time=("time", pd.date_range("01-01-2001", freq="M", periods=6)),
        labels=("time", np.array(["a", "b", "c", "c", "b", "a"])),
    ),
)
da.groupby("labels").count()

[headtr1ck]

import numpy as np
import pandas as pd
import xarray as xr

da = xr.DataArray(
    np.array([1, 2, 3, 1, 2, np.nan]),
    dims="time",
    coords=dict(
        time=("time", pd.date_range("01-01-2001", freq="M", periods=6)),
        labels=("time", np.array(["a", "b", "c", "c", "b", "a"])),
    ),
)
da.groupby("labels").count()

da = xr.DataArray(
    np.array([1, 2, 3, 1, 2, np.nan]),
    dims="time",
    coords=dict(
        time=("time", pd.date_range("01-01-2001", freq="M", periods=6)),
        labels=("time", np.array(["a", "b", "c", "c", "b", "a"])),
    ),
)
da.groupby("labels").count()

For me this works in a python terminal, python script and jupyter notebook (I don't use spyder but vscode).