Fix lambda variable name issues in nested DataFrame functions in R APIs

R/pkg/R/functions.R

@@ -3670,7 +3670,12 @@ unresolved_named_lambda_var <- function(...) {
     "org.apache.spark.sql.Column",
     newJObject(
       "org.apache.spark.sql.catalyst.expressions.UnresolvedNamedLambdaVariable",
-      list(...)
+      lapply(list(...), function(x) {
+        handledCallJStatic(
+          "org.apache.spark.sql.catalyst.expressions.UnresolvedNamedLambdaVariable",
+          "freshVarName",
+          x)
+      })
     )
   )
   column(jc)

R/pkg/tests/fulltests/test_sparkSQL.R

@@ -2161,6 +2161,20 @@ test_that("higher order functions", {
   expect_error(array_transform("xs", function(...) 42))
 })
 
+test("SPARK-34794: lambda variables must be resolved properly in nested higher order functions") {
+  df <- sql("SELECT array(1, 2, 3) as numbers, array('a', 'b', 'c') as letters")
+  ret <- first(select(
+    df,
+    array_transform("numbers", function(number) {
+      array_transform("letters", function(latter) {
+        struct(alias(number, "n"), alias(latter, "l"))
+      })
+    })
+  ))
+
+  expect_equal(1, ret[[1]][[1]][[1]][[1]]$n)
+}
+
 test_that("group by, agg functions", {
   df <- read.json(jsonPath)
   df1 <- agg(df, name = "max", age = "sum")