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/2801#issuecomment-469841163,https://api.github.com/repos/pydata/xarray/issues/2801,469841163,MDEyOklzc3VlQ29tbWVudDQ2OTg0MTE2Mw==,1217238,2019-03-05T20:16:52Z,2019-03-05T20:16:52Z,MEMBER,"> 1. `x[x > 2]` silently resolves the dask graph. It definitely shouldn't.

Yes, agreed.

> There needs to be some discussion on what needs to happen to indices on the NaN-sized dimension

The simple answer is to not allow undefined indices on NaN sized dimensions for now (""hard crash""). I don't think silently dropping indexes is a good idea.

Long term (after https://github.com/pydata/xarray/issues/1603 is complete), I can imagine supporting some sort of ""lazy index"" class.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,417356439
https://github.com/pydata/xarray/issues/2801#issuecomment-469731181,https://api.github.com/repos/pydata/xarray/issues/2801,469731181,MDEyOklzc3VlQ29tbWVudDQ2OTczMTE4MQ==,6815844,2019-03-05T15:47:18Z,2019-03-05T15:47:18Z,MEMBER,"> 1. x[x > 2] silently resolves the dask graph.

Currently, xarray's indexing does not work for dask-indexers.
At least, there is a TODO 
https://github.com/pydata/xarray/blob/c33dab237a0aca316b7f6d4ba10857829a895a25/xarray/core/variable.py#L555-L560
but I am not sure if it is the only one thing to prevent this.

> 2. This crashes:
>>> xarray.DataArray(a.data[a.data > 2]).compute()

I am not sure whether it is our issue or that in upstream.
```python
a.data[(a < 2).data]
```
does not clash, indicating dask can not take xr.DataArray as an index (note, `a < 2` is an xr.DataArray).","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,417356439