ARROW-13766: [R] Add slice_*() methods

[nealrichardson]

This PR implements slice_head,()slice_tail(), slice_min(), slice_max() and slice_sample(). slice_sample() requires a clever hack using a UDF because the random() C++ function apparently does not work; see ARROW-17974.

[nealrichardson]

As @paleolimbot noted on ARROW-13766, it is possible to do with_ties but it requires doing the top-k calculation and then using that as a filtering join. I'm inclined either not to support it (as done here, before I saw that comment) or to flip the default value of with_ties and perhaps raise a warning the first time the function is called to note the difference. Thoughts?

[github-actions]

https://issues.apache.org/jira/browse/ARROW-13766

[thisisnic]

Thanks, will be excellent to have these functions in 10.0.0 :D

A few bits and pieces I noticed when reviewing are below.

[thisisnic]

There's an error message for extraneous slice_*()-related arguments which dplyr has but we don't have that here:

library(dplyr)
library(arrow)
mtcars %>%
slice_tail(n=3, with_ties=FALSE) %>%
collect()
#> Error in `slice_tail()`:#> ! `...` must be empty.#> ✖ Problematic argument:#> • with_ties = FALSEmtcars %>%
arrow_table() %>%
slice_tail(n=3, with_ties=FALSE) %>%
collect()
#> mpg cyl disp hp drat wt qsec vs am gear carb#> 1 19.7 6 145 175 3.62 2.77 15.5 0 1 5 6#> 2 15.0 8 301 335 3.54 3.57 14.6 0 1 5 8#> 3 21.4 4 121 109 4.11 2.78 18.6 1 1 4 2

[nealrichardson]

Good catch, added the rlang::check_dots_empty call. I left it as rlang:: instead of adding it to the importFrom we use because I thought you had some PRs open that were touching that, can move there before merging. (Also need to re-run the docgen.R anyway.)

[thisisnic]

I had a check, and none of my open PRs are touching that, so good to move whenever.

[nealrichardson]

@thisisnic thanks for your excellent review. I've addressed your feedback and pushed a little more to unlock slice_sample(), PTAL

[nealrichardson]

@thisisnic want to give this one more look?

[thisisnic]

LGTM!

[ursabot]

Benchmark runs are scheduled for baseline = b5b41cc and contender = 80e3986. 80e3986 is a master commit associated with this PR. Results will be available as each benchmark for each run completes.
Conbench compare runs links:
[Finished ⬇️0.0% ⬆️0.0%] ec2-t3-xlarge-us-east-2
[Failed ⬇️0.0% ⬆️15.0%] test-mac-arm
[Failed ⬇️1.92% ⬆️5.75%] ursa-i9-9960x
[Finished ⬇️0.75% ⬆️0.0%] ursa-thinkcentre-m75q
Buildkite builds:
[Finished] 80e39862 ec2-t3-xlarge-us-east-2
[Failed] 80e39862 test-mac-arm
[Failed] 80e39862 ursa-i9-9960x
[Finished] 80e39862 ursa-thinkcentre-m75q
[Finished] b5b41ccf ec2-t3-xlarge-us-east-2
[Failed] b5b41ccf test-mac-arm
[Failed] b5b41ccf ursa-i9-9960x
[Finished] b5b41ccf ursa-thinkcentre-m75q
Supported benchmarks:
ec2-t3-xlarge-us-east-2: Supported benchmark langs: Python, R. Runs only benchmarks with cloud = True
test-mac-arm: Supported benchmark langs: C++, Python, R
ursa-i9-9960x: Supported benchmark langs: Python, R, JavaScript
ursa-thinkcentre-m75q: Supported benchmark langs: C++, Java

[ursabot]

['Python', 'R'] benchmarks have high level of regressions.
ursa-i9-9960x