Brian Van Klaveren - in order to test this, will I need an environment with python 3.8 and treecorr > 4? I'm not exactly sure how to do that within the science pipelines conda env.

There's a few ways to go about this.

1. If you use lsstsw `rebuild` , I can give you an environment reference, then you would run `bin/deploy -r [reference]` and then `rebuild validate_drp`
2. If you use eups workflow ( `eups distrib install ...` followed by several `setup` commands), you would re-run newinstall also with a new environment reference, and then you would install the stack on that environment
3. If you don't mind doing some conda commands, we can clone an environment of a working stack, do a forced upgrade of only treecorr, and then you can do some development in that way.

I can help you out with any of these.

(also - we only need to test out for treecorr 4, we don't need to test for python 3.8 at this time)

A cloned conda env on lsst-dev would be most comfortable to me. (Option 3)

Hi Jeff - I created a cloned environment but after some investigation myself I think I actually just fixed the issue. Would you mind double checking?

Confirmed that the new version runs fine, and yields the same result as the most recent previous run of `validate_drp`.