From 6f787a12564d9983964cd447efba9936845323a9 Mon Sep 17 00:00:00 2001
From: Neal Richardson <neal.p.richardson@gmail.com>
Date: Sun, 10 May 2020 11:02:13 -0700
Subject: [PATCH] Updates for compatibility with dplyr 1.0

---
 r/NEWS.md                     |  2 ++
 r/R/dplyr.R                   | 13 +++++++++----
 r/tests/testthat/test-dplyr.R |  1 +
 3 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/r/NEWS.md b/r/NEWS.md
index 7623cf87d13..231b3d376ba 100644
--- a/r/NEWS.md
+++ b/r/NEWS.md
@@ -19,6 +19,8 @@
 
 # arrow 0.17.0.9000
 
+* Updates for compatibility with `dplyr` 1.0
+
 # arrow 0.17.0
 
 ## Feather v2
diff --git a/r/R/dplyr.R b/r/R/dplyr.R
index 8c68b0f45e0..bf5d3c688c0 100644
--- a/r/R/dplyr.R
+++ b/r/R/dplyr.R
@@ -264,7 +264,7 @@ restore_dplyr_features <- function(df, query) {
   }
   # Preserve groupings, if present
   if (length(query$group_by_vars)) {
-    df <- dplyr::grouped_df(df, dplyr::groups(query))
+    df <- dplyr::grouped_df(df, dplyr::group_vars(query))
   }
   df
 }
@@ -294,9 +294,15 @@ summarise.arrow_dplyr_query <- function(.data, ...) {
 }
 summarise.Dataset <- summarise.Table <- summarise.RecordBatch <- summarise.arrow_dplyr_query
 
-group_by.arrow_dplyr_query <- function(.data, ..., add = FALSE) {
+group_by.arrow_dplyr_query <- function(.data, ..., .add = FALSE, add = .add) {
   .data <- arrow_dplyr_query(.data)
-  .data$group_by_vars <- dplyr::group_by_prepare(.data, ..., add = add)$group_names
+  if (".add" %in% names(formals(dplyr::group_by))) {
+    # dplyr >= 1.0
+    gv <- dplyr::group_by_prepare(.data, ..., .add = .add)$group_names
+  } else {
+    gv <- dplyr::group_by_prepare(.data, ..., add = add)$group_names
+  }
+  .data$group_by_vars <- gv
   .data
 }
 group_by.Dataset <- group_by.Table <- group_by.RecordBatch <- group_by.arrow_dplyr_query
@@ -324,7 +330,6 @@ mutate.arrow_dplyr_query <- function(.data, ...) {
   dplyr::mutate(dplyr::collect(.data), ...)
 }
 mutate.Dataset <- mutate.Table <- mutate.RecordBatch <- mutate.arrow_dplyr_query
-# transmute() "just works" because it calls mutate() internally
 # TODO: add transmute() that does what summarise() does (select only the vars we need)
 
 arrange.arrow_dplyr_query <- function(.data, ...) {
diff --git a/r/tests/testthat/test-dplyr.R b/r/tests/testthat/test-dplyr.R
index e531fb30cf6..3ae915e867a 100644
--- a/r/tests/testthat/test-dplyr.R
+++ b/r/tests/testthat/test-dplyr.R
@@ -228,6 +228,7 @@ test_that("mutate", {
 })
 
 test_that("transmute", {
+  skip("TODO: reimplement transmute (with dplyr 1.0, it no longer just works via mutate)")
   expect_dplyr_equal(
     input %>%
       select(int, chr) %>%