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/3029#issuecomment-506600574,https://api.github.com/repos/pydata/xarray/issues/3029,506600574,MDEyOklzc3VlQ29tbWVudDUwNjYwMDU3NA==,1217238,2019-06-28T04:59:46Z,2019-06-28T04:59:46Z,MEMBER,"This is working on Azure, which has a longer build timeout.","{""total_count"": 1, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 1, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,457080809
https://github.com/pydata/xarray/pull/3029#issuecomment-505967580,https://api.github.com/repos/pydata/xarray/issues/3029,505967580,MDEyOklzc3VlQ29tbWVudDUwNTk2NzU4MA==,5635139,2019-06-26T17:19:43Z,2019-06-26T17:19:43Z,MEMBER,"> It probably makes sense to use a separate pip install afterwards to install pandas

To confirm, does this mean running a separate `pip install git+https://github.com/pandas-dev/pandas.git` after the install? We already have a separate pip section in the conda environment (I generally don't use conda so am not well-versed in this)","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,457080809
https://github.com/pydata/xarray/pull/3029#issuecomment-504013893,https://api.github.com/repos/pydata/xarray/issues/3029,504013893,MDEyOklzc3VlQ29tbWVudDUwNDAxMzg5Mw==,1217238,2019-06-20T12:53:14Z,2019-06-20T12:53:14Z,MEMBER,"It probably makes sense to use a separate `pip install` afterwards to install pandas. That lets us leverage `conda` for all the binary dependencies, and on the whole should go much faster.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,457080809
https://github.com/pydata/xarray/pull/3029#issuecomment-502865632,https://api.github.com/repos/pydata/xarray/issues/3029,502865632,MDEyOklzc3VlQ29tbWVudDUwMjg2NTYzMg==,5635139,2019-06-17T21:54:49Z,2019-06-17T21:54:49Z,MEMBER,"I had this passing a couple of times (e.g. https://travis-ci.org/max-sixty/xarray/builds/546932430), I think this those didn't fix some fundamental problem rather
- the end of the conda install stalls for around ten minutes
- travis will error the build if it's silent for ten minutes
- those builds were _slightly_ less than ten minutes

I'll pause here for the moment; let me know if anyone has any bright ideas","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,457080809
https://github.com/pydata/xarray/pull/3029#issuecomment-502846459,https://api.github.com/repos/pydata/xarray/issues/3029,502846459,MDEyOklzc3VlQ29tbWVudDUwMjg0NjQ1OQ==,5635139,2019-06-17T20:52:10Z,2019-06-17T20:52:10Z,MEMBER,"This is how dask installs the 'nightlies' (I'm guessing these are similar to the master branch): https://github.com/dask/dask/blob/master/continuous_integration/travis/install.sh

...maybe we need to do this; ideally we could make a smaller tweak...","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,457080809
https://github.com/pydata/xarray/pull/3029#issuecomment-502835872,https://api.github.com/repos/pydata/xarray/issues/3029,502835872,MDEyOklzc3VlQ29tbWVudDUwMjgzNTg3Mg==,5635139,2019-06-17T20:20:42Z,2019-06-17T20:20:42Z,MEMBER,We're not alone: https://github.com/conda/conda/issues/6986,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,457080809