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/1985#issuecomment-372852604,https://api.github.com/repos/pydata/xarray/issues/1985,372852604,MDEyOklzc3VlQ29tbWVudDM3Mjg1MjYwNA==,6815844,2018-03-13T23:24:37Z,2018-03-13T23:24:37Z,MEMBER,"I see no problem with your code...
Can you try updating xarray to 0.10.2 (released today)? We updated some logic of lazy indexing.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,304624171
https://github.com/pydata/xarray/issues/1985#issuecomment-372563938,https://api.github.com/repos/pydata/xarray/issues/1985,372563938,MDEyOklzc3VlQ29tbWVudDM3MjU2MzkzOA==,6815844,2018-03-13T06:48:23Z,2018-03-13T06:48:23Z,MEMBER,"Umm. I could not find what is wrong with your code.
Can you find which line loads the data into memory?
If your data is still a dask array, it does not print the entries of the array but instead, it shows something like this,
```python
<xarray.DataArray (x: 3)>
dask.array<shape=(3,), dtype=int64, chunksize=(3,)>
Dimensions without coordinates: x
```","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,304624171
https://github.com/pydata/xarray/issues/1985#issuecomment-372545491,https://api.github.com/repos/pydata/xarray/issues/1985,372545491,MDEyOklzc3VlQ29tbWVudDM3MjU0NTQ5MQ==,6815844,2018-03-13T04:44:52Z,2018-03-13T04:48:56Z,MEMBER,"I notice this line
```python
    # Copy dataset
    ds = ds2cut.copy(deep=True)
```
loads the data into memory.
I think you don't need to copy the dataset here.
If you need to copy the data, it is more efficient to make a copy *after* the indexing.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,304624171
https://github.com/pydata/xarray/issues/1985#issuecomment-372544809,https://api.github.com/repos/pydata/xarray/issues/1985,372544809,MDEyOklzc3VlQ29tbWVudDM3MjU0NDgwOQ==,6815844,2018-03-13T04:39:47Z,2018-03-13T04:39:47Z,MEMBER,"> When I load the sub-dataset after using the indexing routines, does xarray need to read the whole original 4D variable?

I don't think so. We support lazy indexing for any dimensional arrays (but not coordinate variables).
What does your data (especially '4Dvariable.nc') look like?
Is `Xp1` coordinate or sufficiently small?
`ds['Xp1'].values` loads `Xp1` into the memory.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,304624171