issue_comments: 542418950

This data as json

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/3161#issuecomment-542418950 https://api.github.com/repos/pydata/xarray/issues/3161 542418950 MDEyOklzc3VlQ29tbWVudDU0MjQxODk1MA== 1217238 2019-10-15T21:46:58Z 2019-10-15T21:46:58Z MEMBER

I don't remember exactly why I added the allow_lazy argument. I think my original concern was backwards compatibility (when we were first adding dask!) with uses of reduce that expected to be applied to NumPy arrays, not dask arrays.

We do something similar in apply_ufunc with dask='forbidden', but rather than automatically coercing to NumPy arrays we raise an error if a dask array is encountered. This seems much more sensible.

At this point, I think we would probably just remove the argument and always default to allow_lazy=True. Or possibly allow_lazy=False should result is an error instead of automatic coercion, and we should expose/document it as a public argument.

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   473000845