Josh Errickson

Thanks - I do something similar right now but its helpful to see another example! Where this falls down for me is needing to do `clean = FALSE`. This could...

I think this is actually a combination of issues between gruvbox and use-package. As @Remillard shows, loading gruvbox through use-package produces the error. However, you can also replicate this error...

gruvbox has addressed the issue - https://github.com/greduan/emacs-theme-gruvbox/issues/192#issuecomment-1207629588 With version gruvbox-theme-20220808.302, I now can load it without error via use-package.

I looked into `survival::strata` for the IES project; the actual code required to using our own `strata` (or, perhaps, a different name to avoid overloading? After deprecating `strata` for a...

I don't know for sure, but my guess is that we'd have to export it. I'll try and play around with it later today

It needs to be exported.

I would argue that we either need to 1. name it something besides `strata`, or 3. if we use `strata`, copy over survival's version completely (and keep it up to...

We could also go more verbose. `matchWithin` or some such. I think if we use `block`, we’d run into the same issue with flexida eventually, though that is far more...

I just pushed up a branch, [remove_survival_strata](https://github.com/markmfredrickson/optmatch/commit/fa1b7c45ce18d399aa236902133016123035553b), that removes our dependence on `survival::strata` in favor of our own `optmatch::strata`. It passes all tests involving the use of `strata` currently. The...

Above you have an example that fails in survival's strata. ``` > a b survival::strata(a, b) [1] b, a, b b, a, b Levels: b, a, b > length(unique(survival::strata(a, b)))...