TST: add test for converting 0-len Series

[jreback]

are these supported?

In [3]: xray.DataArray.from_series(Series([]))
ValueError: dimensions () must have the same length as the number of data dimensions, ndim=1

In [5]: xray.Dataset.from_dataframe(DataFrame({'A' : []}))
ValueError: dimensions () must have the same length as the number of data dimensions, ndim=1

[shoyer]

These definitely should work.

@MaximilianR did your fix handle Series -> DataArray? Let's leave this open if not...

[max-sixty]

@shoyer Was that ever a problem?

In [36]: xray.DataArray(pd.Series())
Out[36]: 
<xray.DataArray (dim_0: 0)>
array([], dtype=float64)
Coordinates:
  * dim_0    (dim_0) object 

In [37]: xray.DataArray(pd.Series(range(20)))
Out[37]: 
<xray.DataArray (dim_0: 20)>
array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16,
       17, 18, 19])
Coordinates:
  * dim_0    (dim_0) int64 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

[shoyer]

@jreback's first case uses the from_series constructor.

[max-sixty]

Ah yes, that goes through the same path as the empty DataFrame. Alas I didn't add a test, but it works

[max-sixty]

TBC, this does work, even if it's not tested:

In [11]: xr.DataArray.from_series(pd.Series())
Out[11]: 
<xarray.DataArray (index: 0)>
array([], dtype=float64)
Coordinates:
  * index    (index) object 

[shoyer]

@MaximilianR Just updated the title of this issue to match

[stale]

In order to maintain a list of currently relevant issues, we mark issues as stale after a period of inactivity
If this issue remains relevant, please comment here; otherwise it will be marked as closed automatically