Tomasz Kalinowski

I ended up working around this by first capturing the call w/ `exprs()`, (which supports injection with `!!!`), then passing that to `match.call()`. Something like this (simplified from the actual...

Thank You @lionel-! Swapping out `exprs()` and `eval()` with `quos()` and `eval_tidy()` seems to allow for resolving arguments in each of their respective environments, as well as splicing with `!!!`...

I see the same failure in reticulate: https://github.com/rstudio/reticulate/actions/runs/8631772754/job/23660886607?pr=1582

Do we have to go through r-hub? Would it make sense to pull `kalibera/rchk:latest` directly?

Just passing by; I'm starting to look into writing a SQLite extension and came across this PR. Seems like it's a nice improvement and ready to merge. Did this just...

Not too much, I'm just starting to look into it. The example seems fine at first impression, was just curious about this PR status. Should this PR be closed?

I'm really not trying to rush anything; just as I'm looking through the docs I wanted to check if the example extension will soon be stale because of this PR....

:+1: other alternatives I've tried using modern tidyverse functions feel clunky: a) `do.call(recode, list(.x = val, named_new_vals))` or b) `factor(val) %>% forcats::fct_recode(named_new_vals) %>% as.character()`

In case anyone else runs into this before it is fixed: A quick workaround is to add this chunk to the document right after the global chunk. ```` ```{r, include...

There is no need to use conda, a standard venv works fine. In 2.15, tensorflow knew to go look for the NVIDIA binaries installed with `pip`. With TF 2.16, you...