id,node_id,number,title,user,state,locked,assignee,milestone,comments,created_at,updated_at,closed_at,author_association,active_lock_reason,draft,pull_request,body,reactions,performed_via_github_app,state_reason,repo,type
474463902,MDU6SXNzdWU0NzQ0NjM5MDI=,3169,Plotting inconsistencies with Cartopy,358378,closed,0,,,4,2019-07-30T09:06:15Z,2022-04-18T15:49:12Z,2022-04-18T15:49:12Z,CONTRIBUTOR,,,,"I'm on xarray 0.11.3, and currently do not have the possibility to test on master.  However, the commit history suggests that this issue still exists.  Please accept my apologies if I missed something.

I'm following the plotting documentation at https://xarray.pydata.org/en/stable/plotting.html#maps

When plotting two maps in the same call (exactly as written in the documentation),

    p = air.isel(time=[0, 1]).plot(transform=ccrs.PlateCarree(), col='time',
                                   subplot_kws={'projection': ccrs.Orthographic(-80, 35)})
    
    for ax in p.axes.flat:
        ax.coastlines()
        ax.gridlines()

everything works fine.

However, when I only want to plot one map,

    p = air.isel(time=[0]).plot(transform=ccrs.PlateCarree(),
                                subplot_kws={'projection': ccrs.Orthographic(-80, 35)})
    
    for ax in p.axes.flat:
        ax.coastlines()
        ax.gridlines()

I get the following error:

    ---------------------------------------------------------------------------
    ValueError                                Traceback (most recent call last)
    <ipython-input-63-2c5a2a533878> in <module>
          1 p = air.isel(time=[0]).plot(transform=ccrs.PlateCarree(),
    ----> 2                             subplot_kws={'projection': ccrs.Orthographic(-80, 35)})
          3 
          4 for ax in p.axes.flat:
          5     ax.coastlines()
    
    /home/jupyterhub/conda/envs/prod/lib/python3.6/site-packages/xarray/plot/plot.py in __call__(self, **kwargs)
        550 
        551     def __call__(self, **kwargs):
    --> 552         return plot(self._da, **kwargs)
        553 
        554     @functools.wraps(hist)
    
    /home/jupyterhub/conda/envs/prod/lib/python3.6/site-packages/xarray/plot/plot.py in plot(darray, row, col, col_wrap, ax, hue, rtol, subplot_kws, **kwargs)
        185     kwargs['ax'] = ax
        186 
    --> 187     return plotfunc(darray, **kwargs)
        188 
        189 
    
    /home/jupyterhub/conda/envs/prod/lib/python3.6/site-packages/xarray/plot/plot.py in newplotfunc(darray, x, y, figsize, size, aspect, ax, row, col, col_wrap, xincrease, yincrease, add_colorbar, add_labels, vmin, vmax, cmap, center, robust, extend, levels, infer_intervals, colors, subplot_kws, cbar_ax, cbar_kwargs, xscale, yscale, xticks, yticks, xlim, ylim, norm, **kwargs)
        852                              vmax=cmap_params['vmax'],
        853                              norm=cmap_params['norm'],
    --> 854                              **kwargs)
        855 
        856         # Label the plot with metadata
    
    /home/jupyterhub/conda/envs/prod/lib/python3.6/site-packages/xarray/plot/plot.py in pcolormesh(x, y, z, ax, infer_intervals, **kwargs)
       1106             y = _infer_interval_breaks(y, axis=0)
       1107 
    -> 1108     primitive = ax.pcolormesh(x, y, z, **kwargs)
       1109 
       1110     # by default, pcolormesh picks ""round"" values for bounds
    
    /home/jupyterhub/conda/envs/prod/lib/python3.6/site-packages/matplotlib/__init__.py in inner(ax, data, *args, **kwargs)
       1808                         ""the Matplotlib list!)"" % (label_namer, func.__name__),
       1809                         RuntimeWarning, stacklevel=2)
    -> 1810             return func(ax, *args, **kwargs)
       1811 
       1812         inner.__doc__ = _add_data_doc(inner.__doc__,
    
    /home/jupyterhub/conda/envs/prod/lib/python3.6/site-packages/matplotlib/axes/_axes.py in pcolormesh(self, alpha, norm, cmap, vmin, vmax, shading, antialiased, *args, **kwargs)
       6011         if (not isinstance(t, mtransforms.Transform) and
       6012             hasattr(t, '_as_mpl_transform')):
    -> 6013             t = t._as_mpl_transform(self.axes)
       6014 
       6015         if t and any(t.contains_branch_seperately(self.transData)):
    
    lib/cartopy/_crs.pyx in cartopy._crs.CRS._as_mpl_transform()
    
    ValueError: Axes should be an instance of GeoAxes, got <class 'matplotlib.axes._subplots.AxesSubplot'>

It is not clear (from the user perspective) why one should work and the other should not.

Of course I can create the axes instance before using xarray's `plot` function, but it would be nicer if I wouldn't have to.","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/3169/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue
219321876,MDU6SXNzdWUyMTkzMjE4NzY=,1352,Saving to netCDF with 0D dimension doesn't work,358378,closed,0,,,9,2017-04-04T17:16:40Z,2021-07-17T18:38:29Z,2021-07-17T18:38:15Z,CONTRIBUTOR,,,,"I'm using xarray 0.9.1.

When I have an array

    <xarray.Dataset>
    Dimensions:       (category: 1, lat: 672, lon: 720, species: 1)
    Coordinates:
      * lat           (lat) float64 30.03 30.09 30.16 30.22 30.28 30.34 30.41 ...
      * category      (category) <U3 'agr'
      * species       (species) <U4 'PM10'
      * lon           (lon) float64 -29.94 -29.81 -29.69 -29.56 -29.44 -29.31 ...
    Data variables:
        TNO-MACC3_EU  (lon, lat) float64 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ...

and try to write to netCDF using `.to_netcdf()`, I get an 

    IndexError: tuple index out of range

I can fix this by manually doing 

    d_['species'] = np.atleast_1d(d_['species'])
    d_['category'] = np.atleast_1d(d_['category'])

before calling `.to_netcdf()`.

I believe it would be good if xarray would do this check on its own and apply the coorection on-the-fly to allow saving to netCDF in this case as well.","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/1352/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue
48301141,MDU6SXNzdWU0ODMwMTE0MQ==,277,Creation of an empty DataArray,358378,closed,0,,987654,11,2014-11-10T19:07:55Z,2020-03-06T12:38:08Z,2020-03-06T12:38:07Z,CONTRIBUTOR,,,,"I'd like to create an empty `DataArray`, i.e., one with only NA values.  The docstring of `DataArray` says that `data=None` is allowed, if a _dataset_ argument is provided.  However, the docstring doesn't say anything about a _dataset_ argument.
1. I think there's a bug in the docstring
2. I'd like to pass _data=None_ and get a `DataArray` with the coords/dims set up properly (as defined by the `coords` and `dims` kwargs), but with a `values` array of NA.
","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/277/reactions"", ""total_count"": 10, ""+1"": 10, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue
46756098,MDU6SXNzdWU0Njc1NjA5OA==,266,Easy iteration over slices of a DataArray,358378,closed,0,,741199,2,2014-10-24T16:20:51Z,2019-01-15T20:09:35Z,2019-01-15T20:09:34Z,CONTRIBUTOR,,,,"The `DataArray` object would benefit from functionality similar to `iris.cube.Cube.slices`. Given an array

```
 [23]: data.coords 
Out[23]: 
Coordinates:
  * sza      (sza) float64 0.0 36.87 53.13 60.0 72.54 75.52 81.37 87.13 88.28
  * vza      (vza) float64 0.0 72.54
  * raa      (raa) float64 0.0 60.0 90.0 120.0 180.0
  * wl       (wl) float64 360.0 380.0 400.0 420.0 440.0
```

it would be nice to be able to do

```
for sl in data.slices([""raa"", ""wl""]):
    # do magic with a DataArray of coordinates (sza, vza)
```
","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/266/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue
213082731,MDU6SXNzdWUyMTMwODI3MzE=,1304,Create `xarray.Dataset` from a `netCDF4.Dataset` object,358378,closed,0,,,5,2017-03-09T16:03:16Z,2017-11-29T16:02:03Z,2017-11-29T16:02:03Z,CONTRIBUTOR,,,,"For some applications, it would be nice if it were possible to create an `xarray.Dataset` from a `netCDF4.Dataset`, so that the following code would work:

    with Dataset(filename) as nc:
        ds = xr.open_dataset(nc)

","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/1304/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue
168848449,MDU6SXNzdWUxNjg4NDg0NDk=,931,How to cite xarray in a research paper,358378,closed,0,2443309,741199,4,2016-08-02T10:13:09Z,2016-08-04T21:17:53Z,2016-08-04T21:17:53Z,CONTRIBUTOR,,,,"It would be helpful if the documentation had an entry (for example, in the FAQ) about how to properly cite xarray for a scientific publication.  I personally like the way e.g. the ipython folks are doing it, they have bibtex code to copy'n'paste (see https://ipython.org/citing.html).

This issue is related to #290, but addresses the general problem and not a specific way.
","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/931/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue
101857508,MDU6SXNzdWUxMDE4NTc1MDg=,540,allow custom 'units' when saving datetimes to netCDF,358378,closed,0,,,2,2015-08-19T10:24:22Z,2016-04-08T18:48:31Z,2016-04-08T18:48:31Z,CONTRIBUTOR,,,,"when manually setting the `units` attribute of a DataArray containing datetime types, the error

```
ValueError: Failed hard to prevent overwriting key 'units'
```

is raised.

It would be good to allow custom units, because currently the units seem to be `days since FIRST_TIMESTEP`, which leads to different units for different variables within the same Dataset.
","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/540/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue
126382000,MDExOlB1bGxSZXF1ZXN0NTU4NzU0NDk=,718,Update index.rst,358378,closed,0,,,1,2016-01-13T09:57:40Z,2016-01-13T16:37:56Z,2016-01-13T16:37:56Z,CONTRIBUTOR,,0,pydata/xarray/pulls/718,"fix link to mailing list
","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/718/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,,13221727,pull
97237275,MDExOlB1bGxSZXF1ZXN0NDA4MzkzNjU=,495,ENH: finite difference method `diff`,358378,closed,0,,,12,2015-07-25T16:33:20Z,2015-08-21T18:13:00Z,2015-08-21T18:12:57Z,CONTRIBUTOR,,0,pydata/xarray/pulls/495,"adds `diff` method to both `DataArray` and `Dataset`. closes #490 

Still to be done:
- tests

Possible enhancements:
- allow numeric `axis` instead of `dim` for `DataArray` objects. The signature would then be `DataArray.diff(dim=None, n=1, axis=None)` and I would need to check that exactly one of `dim` and `axis` is `None`. I find it a bit ugly; `DataArray.diff(n=1, dim=None, axis=None)` would be nicer in my view.  But then the signatures for `DataArray.diff` and `Dataset.diff` would be different, which is also not nice.
- allow specifying the new coordinate array explicitly via a `coord` kwarg instead of just taking the coordinate values of the upper bounds.

What do you think?
","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/495/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,,13221727,pull
96844263,MDU6SXNzdWU5Njg0NDI2Mw==,490,add diff method to DataArray class,358378,closed,0,,,2,2015-07-23T15:46:03Z,2015-08-21T18:12:57Z,2015-08-21T18:12:57Z,CONTRIBUTOR,,,,"I think it would be useful if the `DataArray` class had a `diff` method, equivalent to the `pandas.DataFrame.diff` method.
","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/490/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue
101858170,MDU6SXNzdWUxMDE4NTgxNzA=,541,writing DataArray with 2D datetime variable doesn't work,358378,closed,0,,,1,2015-08-19T10:28:32Z,2015-08-19T16:20:37Z,2015-08-19T16:20:37Z,CONTRIBUTOR,,,,"On xray 0.5.1, writing a _Dataset_ with a non-1D-datetime variable to netCDF doesn't work:

```
times = xray.DataArray([[datetime.datetime.now()] * 2])
print times
<xray.DataArray (dim_0: 1, dim_1: 2)>
array([['2015-08-19T14:27:50.911502000+0200',
        '2015-08-19T14:27:50.911502000+0200']], dtype='datetime64[ns]')
Coordinates:
  * dim_0    (dim_0) int64 0
  * dim_1    (dim_1) int64 0 1

times.to_dataset().to_netcdf('test.nc')
```

raises

```
TypeError: Cannot convert input to Timestamp
```
","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/541/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue
46750605,MDU6SXNzdWU0Njc1MDYwNQ==,265,DataArray computations should handle missing values correctly,358378,closed,0,,,3,2014-10-24T15:28:25Z,2015-03-03T07:38:29Z,2014-10-24T16:17:26Z,CONTRIBUTOR,,,,"Maybe I'm just missing something ... but shouldn't

```
xray.DataArray(np.asarray([1., np.nan]), coords=[[1, 2]], dims=[""X""]).mean()
```

return `1` instead of `nan`?  Or how is calculations with missing values supposed to work?

I'm running xray 0.3.0.
","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/265/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue