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/pull/1776#issuecomment-356154207,https://api.github.com/repos/pydata/xarray/issues/1776,356154207,MDEyOklzc3VlQ29tbWVudDM1NjE1NDIwNw==,1217238,2018-01-09T01:48:25Z,2018-01-09T01:48:25Z,MEMBER,Thanks!,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,281423161
https://github.com/pydata/xarray/pull/1776#issuecomment-353140395,https://api.github.com/repos/pydata/xarray/issues/1776,353140395,MDEyOklzc3VlQ29tbWVudDM1MzE0MDM5NQ==,1217238,2017-12-20T18:11:26Z,2017-12-20T18:11:26Z,MEMBER,"So, for example, we should probably not do anything about the ""unlimited dimension"" attribute. I think this may also be a detail of the server-side hosting this netCDF file over DAP.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,281423161
https://github.com/pydata/xarray/pull/1776#issuecomment-353140026,https://api.github.com/repos/pydata/xarray/issues/1776,353140026,MDEyOklzc3VlQ29tbWVudDM1MzE0MDAyNg==,1217238,2017-12-20T18:10:01Z,2017-12-20T18:10:01Z,MEMBER,I will take a look at this more carefully -- thanks for putting together the test case! My general opinion is that we should avoid parsing attributes returned by pydap as much as possible.,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,281423161
https://github.com/pydata/xarray/pull/1776#issuecomment-352972835,https://api.github.com/repos/pydata/xarray/issues/1776,352972835,MDEyOklzc3VlQ29tbWVudDM1Mjk3MjgzNQ==,6815844,2017-12-20T06:00:18Z,2017-12-20T06:00:18Z,MEMBER,"I am not very familiar with *unlimited dimension* of netCDF and how we handle this.
Pydap supports *unlimited dimension* attributes.

For example, 
['bears.nc' example](http://test.opendap.org/opendap/hyrax/data/nc/bears.nc.html) has three finite-sized dimensions (`i`, `j`, `l`) and one unlimited dimension `k`.

We just need to create a three-dimensional array with `encoding['unlimited_dims'] = 'k'`?
or need to add scalar coordinate for `k`?","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,281423161
https://github.com/pydata/xarray/pull/1776#issuecomment-352011792,https://api.github.com/repos/pydata/xarray/issues/1776,352011792,MDEyOklzc3VlQ29tbWVudDM1MjAxMTc5Mg==,6815844,2017-12-15T13:57:21Z,2017-12-15T13:57:21Z,MEMBER,"Now tests are passing and this PR is ready for some review.
I'm quite new to pydap. I would appreciate any comments from experienced members.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,281423161
https://github.com/pydata/xarray/pull/1776#issuecomment-351127359,https://api.github.com/repos/pydata/xarray/issues/1776,351127359,MDEyOklzc3VlQ29tbWVudDM1MTEyNzM1OQ==,1217238,2017-12-12T17:39:52Z,2017-12-12T17:39:52Z,MEMBER,I'd really like to have at least a basic integration with pydap -- these tests that access remote resources are not very reliable. We should probably construct a pydap `DatasetType` object by hand and pass it to `PydapDataStore`.,"{""total_count"": 1, ""+1"": 1, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,281423161