Add new map_vec() to purrr_mappers vector

[Bisaloo]

No description provided.

[MichaelChirico]

thanks! could you add yourself to the NEWS item and add a test for this?

[IndrajeetPatil]

Oh, purrr 1.0.0 is already on CRAN. Cool!

Thanks.

[codecov-commenter]

Codecov Report

Merging #1866 (b3d6a0e) into main (67576ef) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##             main    #1866   +/-   ##
=======================================
  Coverage   98.86%   98.86%           
=======================================
  Files         112      112           
  Lines        4839     4839           
=======================================
  Hits         4784     4784           
  Misses         55       55           

Impacted Files	Coverage Δ
R/unnecessary_lambda_linter.R	100.00% <ø> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[MichaelChirico]

Should map2_vec and pmap_vec also be included?

[Bisaloo]

Should map2_vec and pmap_vec also be included?

Then all map2() and pmap() variants should be added, which is currently not the case. But since mapply() is covered, I guess it makes sense to put them as well.

[MichaelChirico]

ah, nvm, I didn't read the docs enough... actually I'm surprised mapply is there, maybe it shouldn't be. the logic would be different for the >1-arg function case

[Bisaloo]

But if I understand correctly, it clashes with

lintr/R/unnecessary_lambda_linter.R

Line 34 in 7657984

# a. it's a one-variable function [TODO(michaelchirico): is this necessary?]

[MichaelChirico]

I guess I'd you're using mapply(function(x) x, 1:10) that is indeed a lint but I still assume the logic is off more generally. anyway, that's for a follow-up issue

[Bisaloo]

On the same note, outer() behaves differently than the others as well

[Bisaloo]

thanks! could you add yourself to the NEWS item and add a test for this?

Should I just expand on the existing news bullet rather than adding a new one since there hasn't been a release since this linter was created?

lintr/NEWS.md

Lines 98 to 100 in 67576ef

	* `unnecessary_lambda_linter()`: detect unnecessary lambdas (anonymous functions), e.g.
	`lapply(x, function(xi) sum(xi))` can be `lapply(x, sum)` and `purrr::map(x, ~quantile(.x, 0.75, na.rm = TRUE))`
	can be `purrr::map(x, quantile, 0.75, na.rm = TRUE)`. Naming `probs = 0.75` can further improve readability (#1531, @MichaelChirico).

[MichaelChirico]

yes exactly, I would just add to the bullet with your name and this PR number

[MichaelChirico]

thanks!

[DavisVaughan]

@MichaelChirico is this recommending you use map_vec(x, foo, y = y) instead?

We actually recommend using the anonymous function form rather than using ... to pass named arguments through to foo() now. Using the anonymous function form makes it a little easier to see which arguments belong to which function (i.e. does y belong to map_vec() or foo()?), and tends to give slightly better error messages. See the new documentation of ... at https://purrr.tidyverse.org/reference/map.html for more.

[MichaelChirico]

Thanks Davis! Given that recommendation I think we should exclude purrr mappers by default; filed #1871 to handle that before release.