issues: 168243768

This data as json

id node_id number title user state locked assignee milestone comments created_at updated_at closed_at author_association active_lock_reason draft pull_request body reactions performed_via_github_app state_reason repo type
168243768 MDU6SXNzdWUxNjgyNDM3Njg= 921 Inconsistency between the types of Dataset.dims and DataArray.dims 1217238 closed 0     10 2016-07-29T03:30:08Z 2019-01-25T22:01:47Z 2019-01-25T22:01:46Z MEMBER      

DataArray.dims is currently a tuple, whereas Dataset.dims is a dict. This results in ugly code like this, taken from xarray/core/group.py:

python try: # Dataset expected_size = obj.dims[group_dim] except TypeError: # DataArray expected_size = obj.shape[obj.get_axis_num(group_dim)]

One way to resolve this inconsistency would be switch DataArray.dims to a (frozen) OrderedDict. The downside is that code like x.dims[0] wouldn't work anymore (unless we add some terrible fallback logic). You'd have to write x.dims.keys()[0], which is pretty ugly. On the plus side, x.dims['time'] would always return the size of the time dimension, regardless of whether x is a DataArray or Dataset.

Another option would be to add an attribute dim_shape (or some other name) that serves as an alias to dims on Dataset and an alias to OrderedDict(zip(dims, shape)) on DataArray. This would be fully backwards compatible, but further pollute the namespace on Dataset and DataArray.

 
{
    "url": "https://api.github.com/repos/pydata/xarray/issues/921/reactions",
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   completed 13221727 issue

Links from other tables