Fixes the issue noticed in: https://github.com/pydata/xarray/pull/4175#issuecomment-649135372

Let's see if this passes CI...

• [x] Passes isort -rc . && black . && mypy . && flake8

See below for the new language, to clarify that documentation is only necessary for "user visible changes."

I added "including notable bug fixes" to indicate that minor bug fixes may not be worth noting (I was thinking of test-suite only fixes in this category) but perhaps that is too confusing.

cc @pydata/xarray for opinions!

• [ ] Closes #xxxx
• [ ] Tests added
• [ ] Passes isort -rc . && black . && mypy . && flake8
• [ ] Fully documented, including whats-new.rst for user visible changes (including notable bug fixes) and api.rst for new API

We don't support creating a Panel with to_pandas() with any version of pandas at present, so this test was previous broken if pandas < 0.25 was installed.

Fixes https://github.com/pydata/xarray/issues/3206

Example usage:

In [3]: import pandas as pd
   ...: import numpy as np
   ...: import xarray
   ...: df = pd.DataFrame({
   ...:     'w': range(10),
   ...:     'x': list('abcdefghij'),
   ...:     'y': np.arange(0, 100, 10),
   ...:     'z': np.ones(10),
   ...: }).set_index(['w', 'x', 'y'])
   ...:

In [4]: ds = xarray.Dataset.from_dataframe(df, sparse=True)

In [5]: ds.z.data
Out[5]: <COO: shape=(10, 10, 10), dtype=float64, nnz=10, fill_value=nan>

• [x] Closes #3206, Closes #2139
• [x] Tests added
• [x] Passes black . && mypy . && flake8
• [x] Fully documented, including whats-new.rst for all changes and api.rst for new API

I've added internal consistency checks to the uses of assert_equal in our test suite, so this shouldn't happen again.

• [x] Closes #2856, closes #2854
• [x] Tests added
• [x] Fully documented, including whats-new.rst for all changes and api.rst for new API

Example usage:

```

import xarray times = xarray.cftime_range('2000', periods=30, freq='MS') da = xarray.DataArray(range(30), [('time', times)]) da.resample(time='1AS').mean() <xarray.DataArray (time: 3)> array([ 5.5, 17.5, 26.5]) Coordinates: * time (time) object 2001-01-01 00:00:00 ... 2003-01-01 00:00:00 ```

This frees users from needing to worry about this.

Using __del__ turned up to be easier than using weak references.

• [x] Closes #2560
• [x] Closes #2614
• [x] Tests added
• [x] Fully documented, including whats-new.rst for all changes and api.rst for new AP

It doesn't always make sense to lint these files fully.

I'm not entirely sure this is a good idea. The advantage is that it ensures that all our indexing code is entirely explicit: everything that reaches a backend must be an ExplicitIndexer. The downside is that it removes a bit of internal flexibility: we can't just use tuples in place of basic indexers anymore. On the whole, I think this is probably worth it but I would appreciate feedback.

@fujiisoup can you take a look?

• [x] Tests added / passed
• [x] Passes git diff upstream/master **/*py | flake8 --diff

I was finding myself writting plt.figure(figsize=(x, y)) way too often. This will be a convenient shortcut.

Still needs tests.

• Renamed the Coordinate class from xarray's low level API to IndexVariable. xref https://github.com/pydata/xarray/pull/947#issuecomment-238549129
• Deprecated supplying coords as a dictionary to the DataArray constructor without also supplying an explicit dims argument. The old behavior encouraged relying on the iteration order of dictionaries, which is a bad practice (fixes #727).
• Removed a number of methods deprecated since v0.7.0 or earlier: load_data, vars, drop_vars, dump, dumps and the variables keyword argument alias to Dataset.

xref #475

Example usage:

``` In [1]: da = xray.DataArray(np.arange(56).reshape((7, 8)), ...: coords={'x': list('abcdefg'), ...: 'y': 10 * np.arange(8)}, ...: dims=['x', 'y']) ...:

In [2]: da Out[2]: <xray.DataArray (x: 7, y: 8)> array([[ 0, 1, 2, 3, 4, 5, 6, 7], [ 8, 9, 10, 11, 12, 13, 14, 15], [16, 17, 18, 19, 20, 21, 22, 23], [24, 25, 26, 27, 28, 29, 30, 31], [32, 33, 34, 35, 36, 37, 38, 39], [40, 41, 42, 43, 44, 45, 46, 47], [48, 49, 50, 51, 52, 53, 54, 55]]) Coordinates: * y (y) int64 0 10 20 30 40 50 60 70 * x (x) |S1 'a' 'b' 'c' 'd' 'e' 'f' 'g'

we can index by position along each dimension

In [3]: da.isel_points(x=[0, 1, 6], y=[0, 1, 0], dim='points') Out[3]: <xray.DataArray (points: 3)> array([ 0, 9, 48]) Coordinates: y (points) int64 0 10 0 x (points) |S1 'a' 'b' 'g' * points (points) int64 0 1 2

or equivalently by label

In [4]: da.sel_points(x=['a', 'b', 'g'], y=[0, 10, 0], dim='points') Out[4]: <xray.DataArray (points: 3)> array([ 0, 9, 48]) Coordinates: y (points) int64 0 10 0 x (points) |S1 'a' 'b' 'g' * points (points) int64 0 1 2 Bug fixes ```

cc @jhamman

Fixes #133

@DamienIrving am I missing anything obvious here?