GH-49534: [R] Implement dplyr recode_values(), replace_values(), and replace_when()

[thisisnic]

Rationale for this change

Implement new dplyr functions

What changes are included in this PR?

Implement them

Are these changes tested?

Yeah

Are there any user-facing changes?

Moar functions

AI Use

Code generated using Claude, with plenty of input from me. I've gone through it in detail and refactored lots, but it needs a last pass before it's ready for review.

• GitHub Issue: [R] Implement dplyr recode_values(), replace_values(), and replace_when() #49534

[jonkeane]

The tests look good, and so long as CI doesn't bonk I'm good with shipping this. I do think we should keep the validation error when being given a vectorized .default though.

[jonkeane]

I'm not totally sure I think we should make this change. The original message is clearer to me about what needs to happen. I also don't mind that it's a validation_error since that is what it is.

[thisisnic]

Good shout on the contents of the error message, will update to be more recommendationy!

I'm not sure about the type of error though - my reasoning here was that it's a feature which is supported in dplyr but not arrow, which is where we typically use arrow_not_supported, whereas validation_error is more for things that fail in both. I guess it's tricky when size isn't either length 1 or the same length as the input, as then it's wrong in both, but in the bindings for str_sub/substr we use arrow_not_supported.

[jonkeane]

Ok, that's fair. I'm ok with arrow_not_supported + the clearer message about number of arguments

[jonkeane]

@github-actions crossbow submit -g r

[github-actions]

Revision: 825940b

Submitted crossbow builds: ursacomputing/crossbow @ actions-40f5a339fc

Task	Status
r-binary-packages
r-recheck-most
test-r-alpine-linux-cran
test-r-arrow-backwards-compatibility
test-r-depsource-system
test-r-dev-duckdb
test-r-devdocs
test-r-extra-packages
test-r-fedora-clang
test-r-gcc-11
test-r-gcc-12
test-r-install-local
test-r-install-local-minsizerel
test-r-linux-as-cran
test-r-linux-rchk
test-r-linux-sanitizers
test-r-linux-valgrind
test-r-m1-san
test-r-macos-as-cran
test-r-offline-maximal
test-r-ubuntu-22.04
test-r-versions

[thisisnic]

@github-actions crossbow submit -g r

[github-actions]

Revision: bbb4113

Submitted crossbow builds: ursacomputing/crossbow @ actions-29b827672d

Task	Status
r-binary-packages
r-recheck-most
test-r-alpine-linux-cran
test-r-arrow-backwards-compatibility
test-r-depsource-system
test-r-dev-duckdb
test-r-devdocs
test-r-extra-packages
test-r-fedora-clang
test-r-gcc-11
test-r-gcc-12
test-r-install-local
test-r-install-local-minsizerel
test-r-linux-as-cran
test-r-linux-rchk
test-r-linux-sanitizers
test-r-linux-valgrind
test-r-m1-san
test-r-macos-as-cran
test-r-offline-maximal
test-r-ubuntu-22.04
test-r-versions

[thisisnic]

CI failure appears to be unrelated (duckddb installation) so I'll merge

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 3 benchmarking runs that have been run so far on merge-commit bb4e492.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 4 possible false positives for unstable benchmarks that are known to sometimes produce them.