html_url,issue_url,id,node_id,user,created_at,updated_at,author_association,body,reactions,performed_via_github_app,issue
https://github.com/pydata/xarray/issues/5664#issuecomment-891602277,https://api.github.com/repos/pydata/xarray/issues/5664,891602277,IC_kwDOAMm_X841JMVl,7017525,2021-08-03T07:21:26Z,2021-08-03T07:46:09Z,NONE,"Thanks,
Here is the ticket on Scipy's bug tracker : https://github.com/scipy/scipy/issues/14531

FWI this is what I ended up doing (which feels like work in both cases, interpolation needed or not):
```python
interpolated_ds =  ds.interp(time=ds.times)
reindexed_ds = ds.reindex(time=ds.times)
ds = interpolated_ds.fillna(reindexed_ds)
```
It is probably not the most efficient in terms of operation but it gets the job done. 

Thanks for your feedback ;) ","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,957918751
https://github.com/pydata/xarray/issues/5664#issuecomment-891597039,https://api.github.com/repos/pydata/xarray/issues/5664,891597039,IC_kwDOAMm_X841JLDv,10194086,2021-08-03T07:13:17Z,2021-08-03T07:13:17Z,MEMBER,"There does not seem a way to force the numpy interpolator for the 1D case:

https://github.com/pydata/xarray/blob/2f8623d4d9929b61ff4ebfd5232c12474420dfae/xarray/core/missing.py#L472

(that could be a work around)","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,957918751
https://github.com/pydata/xarray/issues/5664#issuecomment-891140503,https://api.github.com/repos/pydata/xarray/issues/5664,891140503,IC_kwDOAMm_X841HbmX,2448579,2021-08-02T15:59:10Z,2021-08-02T16:00:14Z,MEMBER,"Thanks @mathisc we use scipy by default and that's what it does...

``` python
In [6]: from scipy.interpolate import interp1d

In [7]: f = sp.interpolate.interp1d([0, 1, 2], [np.nan, 0, 1])

In [8]: f([0, 1, 2])
Out[8]: array([nan, nan,  1.])
```

Can you file an issue over at scipy?

EDIT: I guess we could skip interpolating when output coordinate values are a subset of the input coordinate values.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,957918751