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/4089#issuecomment-633336577,https://api.github.com/repos/pydata/xarray/issues/4089,633336577,MDEyOklzc3VlQ29tbWVudDYzMzMzNjU3Nw==,12229877,2020-05-25T01:39:49Z,2020-05-25T08:29:16Z,CONTRIBUTOR,"> Also, I don't think there is a hypothesis.extra.xarray module, yet. Any comments on that, @Zac-HD?

Not yet, though I'm interested in collaborations to write one!

Our existing Numpy + Pandas integrations, along with the `st.builds()` and `@st.composite` strategies, make it reasonably straightforward-if-tedious to compose the objects you want in exquisite detail... the real challenge is to design a clean high-level API which lets you control only the parts you care about.  Plausibly we have to let go of that last part to make this work, and just point people at the low-level tools if high-level isn't working.

[This third-party module](https://github.com/uber/hypothesis-gufunc/blob/master/hypothesis_gufunc/extra/xr.py) is unidiomatic in that it doesn't have the API design above, but I believe it works.  rdturnermtl has a history of great features; we eventually got idiomatic high-performance gufunc shapes upstream and I'm confident we'll get Xarray support eventually too... and sooner if there are people who can help design it :smile:

Just \@-mention me again if this comes up, or I won't be notified.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,623751213