What happened?

The xarray.plot.scatter method has changed its behavior in 2022.10.0. Code that used to work now doesn't.

The issue seems to be related to default values for the size and hue of markers. Instead of using s and c arguments, xarray tries to infer DataArrays to use, but picks non-sensical values.

What did you expect to happen?

A scatter plot with default size and color for markers. Now xarray has inferred that the size is somehow related to the j dimension, and the hue to bnds.

Note that the calculations required to draw the figure with those defaults take a huge amount of time. In the example below, I've subsetted the file so the code runs in a short time. Without subsetting, it runs forever.

Minimal Complete Verifiable Example

Python import xarray as xr from matplotlib import pyplot as plt url = "https://pavics.ouranos.ca/twitcher/ows/proxy/thredds/dodsC/birdhouse/testdata/xclim/cmip6/sic_SImon_CCCma-CanESM5_ssp245_r13i1p2f1_2020.nc" ds = xr.open_dataset(url) t = ds.isel(i=slice(0,10), j=slice(0,11)) t.plot.scatter(x="longitude", y="latitude", s=1) plt.show()

MVCE confirmation

• [X] Minimal example — the example is as focused as reasonably possible to demonstrate the underlying issue in xarray.
• [X] Complete example — the example is self-contained, including all data and the text of any traceback.
• [X] Verifiable example — the example copy & pastes into an IPython prompt or Binder notebook, returning the result.
• [X] New issue — a search of GitHub Issues suggests this is not a duplicate.

Relevant log output

No response

Anything else we need to know?

Environment

What happened: Slicing into a DataArray along time with a CFTimeIndex fails since upgrade to pandas 1.3

What you expected to happen: The usual.

Minimal Complete Verifiable Example:

python import xarray as xr t = xr.cftime_range('2000-01-01', '2030-12-31', freq='D', calendar='noleap') ref = xr.DataArray(range(len(t)), dims=('time',), coords={'time': t}) ref.sel(time=slice(None, "2015-01-01"))

```python

TypeError Traceback (most recent call last) <ipython-input-5-3afe7d577940> in <module> ----> 1 ref.sel(time=slice(None, "2015-01-01"))

~/.conda/envs/xclim/lib/python3.8/site-packages/xarray/core/dataarray.py in sel(self, indexers, method, tolerance, drop, **indexers_kwargs) 1269 Dimensions without coordinates: points 1270 """ -> 1271 ds = self._to_temp_dataset().sel( 1272 indexers=indexers, 1273 drop=drop,

~/.conda/envs/xclim/lib/python3.8/site-packages/xarray/core/dataset.py in sel(self, indexers, method, tolerance, drop, **indexers_kwargs) 2363 """ 2364 indexers = either_dict_or_kwargs(indexers, indexers_kwargs, "sel") -> 2365 pos_indexers, new_indexes = remap_label_indexers( 2366 self, indexers=indexers, method=method, tolerance=tolerance 2367 )

~/.conda/envs/xclim/lib/python3.8/site-packages/xarray/core/coordinates.py in remap_label_indexers(obj, indexers, method, tolerance, **indexers_kwargs) 419 } 420 --> 421 pos_indexers, new_indexes = indexing.remap_label_indexers( 422 obj, v_indexers, method=method, tolerance=tolerance 423 )

~/.conda/envs/xclim/lib/python3.8/site-packages/xarray/core/indexing.py in remap_label_indexers(data_obj, indexers, method, tolerance) 272 coords_dtype = data_obj.coords[dim].dtype 273 label = maybe_cast_to_coords_dtype(label, coords_dtype) --> 274 idxr, new_idx = convert_label_indexer(index, label, dim, method, tolerance) 275 pos_indexers[dim] = idxr 276 if new_idx is not None:

~/.conda/envs/xclim/lib/python3.8/site-packages/xarray/core/indexing.py in convert_label_indexer(index, label, index_name, method, tolerance) 119 "cannot use method argument if any indexers are slice objects" 120 ) --> 121 indexer = index.slice_indexer( 122 _sanitize_slice_element(label.start), 123 _sanitize_slice_element(label.stop),

~/.conda/envs/xclim/lib/python3.8/site-packages/pandas/core/indexes/base.py in slice_indexer(self, start, end, step, kind) 5684 slice(1, 3, None) 5685 """ -> 5686 start_slice, end_slice = self.slice_locs(start, end, step=step) 5687 5688 # return a slice

~/.conda/envs/xclim/lib/python3.8/site-packages/pandas/core/indexes/base.py in slice_locs(self, start, end, step, kind) 5892 end_slice = None 5893 if end is not None: -> 5894 end_slice = self.get_slice_bound(end, "right") 5895 if end_slice is None: 5896 end_slice = len(self)

~/.conda/envs/xclim/lib/python3.8/site-packages/pandas/core/indexes/base.py in get_slice_bound(self, label, side, kind) 5796 # For datetime indices label may be a string that has to be converted 5797 # to datetime boundary according to its resolution. -> 5798 label = self._maybe_cast_slice_bound(label, side) 5799 5800 # we need to look up the label

TypeError: _maybe_cast_slice_bound() missing 1 required positional argument: 'kind' ```

Anything else we need to know?: A quick diagnostic suggests that convert_label_indexer should call pandas.Index.slice_indexer with a kind argument.

Environment:

The new link is where development now occurs.

A CFPeriodIndex supporting non-standard calendars would be useful to facilitate climate analyses. The use case for me would be to find the start and end date of a resampling group. This is useful to spot missing values in a resampled time series, or to create time_bnds arrays in a netCDF file.

``` import xarray as xr import pandas as pd

cftime = xr.cftime_range(start='2000-01-01', periods=361, freq='D', calendar='360_day') pdtime = pd.date_range(start='2000-01-01', periods=361, freq='D')

cf_da = xr.DataArray(range(361), coords={'time': cftime}, dims='time') pd_da = xr.DataArray(range(361), coords={'time': pdtime}, dims='time')

cf_c = cf_da.resample(time='M').count()

pd_c = pd_da.resample(time='M').count()

cf_p = cf_c.indexes['time'].to_period()

pd_p = pd_c.indexes['time'].to_period()

cf_expected_days_in_group = cf_p.end_time - cf_p.start_time + pd.offsets.Day(1)

pd_expected_days_in_group = pd_p.end_time - pd_p.start_time + pd.offsets.Day(1) ```

Depends on #2191

MCVE Code Sample

```python

Your code here

import xarray as xr
da = xr.DataArray([0, 0], dims="x", attrs={'units':'K'})
out = da.quantile(.9, dim='x', keep_attrs=True)
out.attrs
returns OrderedDict() ```

Expected Output

OrderedDict([('units', 'K')])

Output of xr.show_versions()

• [x] Closes #3252
• [x] Tests added
• [x] Passes isort -rc . && black . && mypy . && flake8
• [x] Fully documented, including whats-new.rst for all changes and api.rst for new API

Replaces #3262 (I think).

• [x] Closes #3641
• [x] Tests added
• [x] Passes black . && mypy . && flake8
• [x] Fully documented, including whats-new.rst for all changes and api.rst for new API

Related to #3349

As suggested by @spencerkclark, index values are computed as a delta with respect to 1970-01-01.

At the moment, this fails if dates fall outside of the range for nanoseconds timedeltas [ 1678 AD, 2262 AD]. Is this something we can fix ?

• [x] Closes #3641
• [x] Tests added
• [x] Passes black . && mypy . && flake8
• [ ] Fully documented, including whats-new.rst for all changes and api.rst for new API

This is likely only safe with NumPy>=1.17 though.

• [x] Closes #3304
• [x] Tests added
• [x] Passes black . && mypy . && flake8
• [x] Fully documented, including whats-new.rst for all changes and api.rst for new API

Note that I've set the default to True (if keep_attrs is None). This sounded reasonable since quantiles share the same units and properties as the original array, but I can switch it to False if that's the usual default.

Dataset and DataArray objects have a quantile method, but not GroupBy. This would be useful for climatological analyses.

MCVE Code Sample

```python import xarray as xr

PyDAP access without subsetting - everything's fine

url = 'http://remotetest.unidata.ucar.edu/thredds/dodsC/testdods/coads_climatology.nc' ds = xr.open_dataset(url, engine='pydap', decode_times=False) ds.TIME.units # yields 'hour since 0000-01-01 00:00:00'

PyDAP access with subsetting - variable attributes are global...

dss = xr.open_dataset(url+'?SST[0:1:11][0:1:0][0:1:0]', engine='pydap', decode_times=False) print(dss.SST.attrs) # all good so far print(dss.TIME.attrs) # oh oh... nothing print(dss.attrs['TIME.units']) ```

Problem Description

Opening a subsetted dataset with PydapDataStore creates global attributes instead of variable attributes.

Expected Output

All the global TIME.* attributes should be attributes of the TIME DataArray.

Output of xr.show_versions()

• [x] Tests added
• [x] Fully documented, including whats-new.rst for all changes and api.rst for new API

Fixes #3018

Note that I've added an unrelated test that exposes an issue with grouping when there is only one element per group.