issue_comments: 334025812

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/1601#issuecomment-334025812 https://api.github.com/repos/pydata/xarray/issues/1601 334025812 MDEyOklzc3VlQ29tbWVudDMzNDAyNTgxMg== 1217238 2017-10-04T01:29:21Z 2017-10-04T01:29:21Z MEMBER

Inspecting your dataset, I see: ``` In [7]: ds = xarray.open_dataset('http://oos.soest.hawaii.edu/thredds/dodsC/hioos/tide_mhi', decode_cf=False)

In [8]: ds.info() xarray.Dataset { dimensions: depth = 3 ; latitude = 242 ; longitude = 521 ; time = 8760 ;

variables: ... float64 v(time, depth, latitude, longitude) ; v:units = meter second-1 ; v:long_name = v-velocity component ; v:standard_name = northward_sea_water_velocity ; v:missing_value = -999.989990234 ; v:_FillValue = -999.99 ; ... ```

So missing_value and _FillValue do indeed differ here, at least in their 64-bit floating point representation. Floating point equality is a pain, but unfortunately for fill values we really do need the bit-patterns to match.

https://github.com/pydata/xarray/pull/1602 updates the error message to print the offending values, which should hopefully improve user-experience if this comes up in the future.

 
{
    "total_count": 1,
    "+1": 1,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   262619571