Davis Vaughan

```r filter( .data, ..., .missing = NULL, .how = c("keep", "drop") ) ``` This signature may work, where `.missing` can still take on one of `c("keep", "drop", "error")`, but would...

It works correctly if you extract the function out into its own named function, so it is probably something weird with our anonymous function handling. ``` r library(dplyr) mtcars mpg...

Yea I see this if I step through the anonymous function after putting a `browser()` in there ```r debug at #5: approx(dens$mpg, dens$y, xout = mpg, na.rm = FALSE, rule...

I think we need to teach `node_walk_replace()` about `$`, like how it knows about `~` and `function`. If we see a `$` then we should not attempt a node replacement....

I'd be open to `.by_row` as a boolean argument to `mutate()` and `reframe()`. It does feel better than re-adding suffixed variants since we worked so hard to back away from...

Yea we don't intend for matrices to be passed through as conditions. We probably need an additional check here for `dim` https://github.com/tidyverse/dplyr/blob/16b472fb2afc50a87502c2b4ed803e2f5f82b9d6/R/vec-case-when.R#L54

This is a consequence of `across()` inlining anonymous functions, so what you end up with is an expression that looks like this: ```r mtcars |> mutate(mpg = glue("Oh hai {x}"),...

That news bullet is saying two things: - Filtering by a matrix is an error - Filtering by a data frame (often created by `across()`) is a warning with a...

Part of the confusion may be that `across()` returns a data frame but `!` on that data frame turns it into a matrix ``` r df # A tibble: 3...

Your 1.0.7 example titled `# also works, but gives a different answer` actually didn't work in 1.0.7. It only used the first column of the matrix and was the reason...