id,node_id,number,title,user,state,locked,assignee,milestone,comments,created_at,updated_at,closed_at,author_association,active_lock_reason,draft,pull_request,body,reactions,performed_via_github_app,state_reason,repo,type
1722614979,I_kwDOAMm_X85mrQTD,7870,Name collision with Pulsar Timing package 'PINT' ,3092444,closed,0,,,4,2023-05-23T18:54:18Z,2023-05-26T16:19:37Z,2023-05-26T16:19:37Z,CONTRIBUTOR,,,,"### What is your issue?

In the astrophysics community of [pulsar timers](https://en.wikipedia.org/wiki/Pulsar_timing_array), there is an analysis package called `PINT`. PINT is widely used in that community. As you can see on their [github](https://github.com/nanograv/PINT), they have been aware of the name collision and on pip/conda the package is available as `pint-pulsar`. This has not been a problem so far, because most if not all astrophysicists use the great [astropy](https://www.astropy.org/) to keep track of units where necessary.

However, Bayesian modeling through PyMC is becoming more and more popular, meaning that arviz and xarray are now getting installed alongside pint-pulsar, giving obvious issues.

A very simple workaround would be to change line 37 in https://github.com/pydata/xarray/blob/main/xarray/core/pycompat.py to something like:

`except (ImportError, AttributeError):`

This means that `pint-pulsar` would get imported through `mod`), and the `AttributeError` gets caught, and all should be well. It fits the design of duck-typing, since the package doesn't Quack like pint should. Would xarray be willing to accommodate the pulsar timing community this way? As you are all aware, changing the name of a package that is integral in projects with many dependencies is kind of painful.

EDIT: fixed typo","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/7870/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue