GH-49533: [R] Implement dplyr's when_any() and when_all() helpers

r/R/dplyr-funcs-conditional.R

@@ -99,6 +99,28 @@ register_bindings_conditional <- function() {
     out
   })
 
+  register_binding("dplyr::when_any", function(..., na_rm = FALSE, size = NULL) {
+    if (!is.null(size)) {
+      arrow_not_supported("`when_any()` with `size` specified")
+    }
+    args <- list2(...)
+    if (na_rm) {
+      args <- lapply(args, function(x) call_binding("coalesce", x, FALSE))
+    }
+    Reduce("|", args)
+  })
+
+  register_binding("dplyr::when_all", function(..., na_rm = FALSE, size = NULL) {
+    if (!is.null(size)) {
+      arrow_not_supported("`when_all()` with `size` specified")
+    }
+    args <- list2(...)
+    if (na_rm) {
+      args <- lapply(args, function(x) call_binding("coalesce", x, TRUE))
+    }
+    Reduce("&", args)
+  })
+
   register_binding(
     "dplyr::case_when",
     function(..., .default = NULL, .ptype = NULL, .size = NULL) {

r/R/dplyr-funcs-doc.R

@@ -21,7 +21,7 @@
 #'
 #' The `arrow` package contains methods for 38 `dplyr` table functions, many of
 #' which are "verbs" that do transformations to one or more tables.
-#' The package also has mappings of 224 R functions to the corresponding
+#' The package also has mappings of 226 R functions to the corresponding
 #' functions in the Arrow compute library. These allow you to write code inside
 #' of `dplyr` methods that call R functions, including many in packages like
 #' `stringr` and `lubridate`, and they will get translated to Arrow and run
@@ -214,6 +214,8 @@
 #' * [`if_else()`][dplyr::if_else()]
 #' * [`n()`][dplyr::n()]
 #' * [`n_distinct()`][dplyr::n_distinct()]
+#' * [`when_all()`][dplyr::when_all()]
+#' * [`when_any()`][dplyr::when_any()]
 #'
 #' ## hms
 #'

r/tests/testthat/test-dplyr-funcs-conditional.R

@@ -517,3 +517,69 @@ test_that("external objects are found when they're not in the global environment
     tibble(x = c("a", "b"), x2 = c("foo", NA))
   )
 })
+
+test_that("when_any()", {
+  # combines with OR
+  compare_dplyr_binding(
+    .input |>
+      mutate(result = when_any(lgl, false)) |>
+      collect(),
+    tbl
+  )
+
+  # na_rm=TRUE treats NA as FALSE
+  compare_dplyr_binding(
+    .input |>
+      mutate(result = when_any(lgl, false, na_rm = TRUE)) |>
+      collect(),
+    tbl
+  )
+
+  # works in filter()
+  compare_dplyr_binding(
+    .input |>
+      filter(when_any(int > 5, dbl > 3)) |>
+      collect(),
+    tbl
+  )
+
+  # size not supported
+  expect_arrow_eval_error(
+    when_any(lgl, false, size = 10),
+    "`when_any\\(\\)` with `size` specified not supported in Arrow",
+    class = "arrow_not_supported"
+  )
+})
+
+test_that("when_all()", {
+  # combines with AND
+  compare_dplyr_binding(
+    .input |>
+      mutate(result = when_all(lgl, false)) |>
+      collect(),
+    tbl
+  )
+
+  # na_rm=TRUE treats NA as TRUE
+  compare_dplyr_binding(
+    .input |>
+      mutate(result = when_all(lgl, false, na_rm = TRUE)) |>
+      collect(),
+    tbl
+  )
+
+  # works in filter()
+  compare_dplyr_binding(
+    .input |>
+      filter(when_all(int > 5, dbl > 3)) |>
+      collect(),
+    tbl
+  )
+
+  # size not supported
+  expect_arrow_eval_error(
+    when_all(lgl, false, size = 10),
+    "`when_all\\(\\)` with `size` specified not supported in Arrow",
+    class = "arrow_not_supported"
+  )
+})