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/3526#issuecomment-554463832,https://api.github.com/repos/pydata/xarray/issues/3526,554463832,MDEyOklzc3VlQ29tbWVudDU1NDQ2MzgzMg==,703554,2019-11-15T17:57:42Z,2019-11-15T17:57:42Z,CONTRIBUTOR,"FWIW in the Zarr Python implementation I don't think we do any special encoding or decoding of attribute values. Whatever value is given then gets serialised using the built-in `json.dumps`. This means I believe that if someone provides a `dict` as an attribute value then that will get serialised as a JSON object, and get deserialised back to a `dict`, although this is not something we test for currently.
From the zarr v2 spec point of view I think anything goes in the `.zattrs` file, as long as `.zattrs` is a JSON object at the root.
Hth.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,522519084