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/1933#issuecomment-368386873,https://api.github.com/repos/pydata/xarray/issues/1933,368386873,MDEyOklzc3VlQ29tbWVudDM2ODM4Njg3Mw==,1217238,2018-02-26T04:26:25Z,2018-02-26T04:26:25Z,MEMBER,"Presumably netcdf4 will add a dependency on netcdftime so they don't break their API for users?

Anyways, if this already work then yes I'm good with merging this.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,298854863
https://github.com/pydata/xarray/pull/1933#issuecomment-368382269,https://api.github.com/repos/pydata/xarray/issues/1933,368382269,MDEyOklzc3VlQ29tbWVudDM2ODM4MjI2OQ==,2443309,2018-02-26T03:46:52Z,2018-02-26T03:46:52Z,MEMBER,"right now, `conda install -c conda-forge netcdftime` will give you:

```
netcdftime:   1.0.0a2-py36_0      conda-forge
```

I don't andticipate any code changes before the final release. We're still working out the documentation site and whatnot.

`netcdf4` did just merge the removal of `netcdftime` (https://github.com/Unidata/netcdf4-python/pull/756) so this will need to go in prior to any release of netCDF4. ","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,298854863
https://github.com/pydata/xarray/pull/1933#issuecomment-368287163,https://api.github.com/repos/pydata/xarray/issues/1933,368287163,MDEyOklzc3VlQ29tbWVudDM2ODI4NzE2Mw==,1217238,2018-02-25T06:29:14Z,2018-02-25T06:29:14Z,MEMBER,"My understanding is that netcdftime hasn't had an official release yet?

Normally, you need to use some sort of prerelease channel and/or flag to download a package that hasn't had a normal release.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,298854863
https://github.com/pydata/xarray/pull/1933#issuecomment-368203742,https://api.github.com/repos/pydata/xarray/issues/1933,368203742,MDEyOklzc3VlQ29tbWVudDM2ODIwMzc0Mg==,2443309,2018-02-24T06:11:56Z,2018-02-24T06:11:56Z,MEMBER,@shoyer - do you think this is good to merge in its current state?,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,298854863
https://github.com/pydata/xarray/pull/1933#issuecomment-367927308,https://api.github.com/repos/pydata/xarray/issues/1933,367927308,MDEyOklzc3VlQ29tbWVudDM2NzkyNzMwOA==,2443309,2018-02-23T07:01:19Z,2018-02-23T07:01:19Z,MEMBER,@fmaussion - mind reviewing the updated documentation here?,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,298854863
https://github.com/pydata/xarray/pull/1933#issuecomment-367637016,https://api.github.com/repos/pydata/xarray/issues/1933,367637016,MDEyOklzc3VlQ29tbWVudDM2NzYzNzAxNg==,10050469,2018-02-22T10:27:03Z,2018-02-22T10:27:03Z,MEMBER,"> Agreed. Do you think I should include that here?

As you wish! It can wait until the netcdftime documentation is up and running (most important I guess)","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,298854863
https://github.com/pydata/xarray/pull/1933#issuecomment-367556688,https://api.github.com/repos/pydata/xarray/issues/1933,367556688,MDEyOklzc3VlQ29tbWVudDM2NzU1NjY4OA==,2443309,2018-02-22T03:31:58Z,2018-02-22T03:31:58Z,MEMBER,"@fmaussion -

> Is it safe to do so? I.e. can it hide bugs in the current netCDF4 time handling?

Save the one test failure here, I think this is going to be a fairly safe change.  The `netcdftime` module in `netCDF4` is being ported to a stand-alone package (see: #1048, #1920, https://github.com/Unidata/netcdf4-python/pull/756, and #https://github.com/Unidata/netcdftime/issues/20).  The same test suite is being run on the `netcdftime` package and we've been working through a series of integration tests with xarray (mostly painless, i.e. #1929). 

> Also I think after #1920 it would be good to have a bit more info in the documentation about why people should switch to netcdftime or not.

Agreed. Do you think I should include that here?","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,298854863
https://github.com/pydata/xarray/pull/1933#issuecomment-367368044,https://api.github.com/repos/pydata/xarray/issues/1933,367368044,MDEyOklzc3VlQ29tbWVudDM2NzM2ODA0NA==,10050469,2018-02-21T15:42:01Z,2018-02-21T15:42:01Z,MEMBER,"Is it safe to do so? I.e. can it hide bugs in the current netCDF4 time handling?

Also I think after #1920 it would be good to have a bit more info in the documentation about why people should switch to netcdftime or not.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,298854863