issue_comments: 167404601

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/684#issuecomment-167404601 https://api.github.com/repos/pydata/xarray/issues/684 167404601 MDEyOklzc3VlQ29tbWVudDE2NzQwNDYwMQ== 1217238 2015-12-27T12:04:43Z 2015-12-27T12:04:43Z MEMBER

Hmm. We currently use np.ma.masked_where to construct a masked array, which sometimes creates masked arrays with as a scalar mask (mask=False):

```

np.ma.masked_where(np.arange(3) < -10, np.arange(3)) masked_array(data = [0 1 2], mask = False, fill_value = 999999) ```

I'm not entirely sure if these are actually valid masked arrays or not (maybe @jhamman or someone who uses masked arrays more often than me would know?), but I agree that it's not very useful for xray to return them. An alternative would be to use the MaskedArray constructor, which does not have this issue:

```

np.ma.MaskedArray(np.arange(3), np.arange(3) < -10) masked_array(data = [0 1 2], mask = [False False False], fill_value = 999999) ```

This seems like better behavior to me.

If you're interested in putting together a fix, here's the relevant place in xray: https://github.com/xray/xray/blob/1c908bebff4c621c959d1bc8f78b4e4298ee87ce/xray/core/dataarray.py#L993

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   123971294