Pass ... on to merge

Fixes #41

And rewrite test

DESCRIPTION

@@ -11,6 +11,7 @@ Description: This implements the data table back-end for 'dplyr' so that you
 License: GPL (>= 2)
 Imports:
  dplyr (>= 0.5.0),
+ ellipsis,
  tidyselect,
  data.table (>= 1.9.6),
  rlang

NEWS.md

@@ -1,5 +1,7 @@
 # dtplyr 0.0.3.9000
 
+* Joins now pass `...` on to data table's merge method (#41).
+
 * Convert from lazyeval to tidy eval (@christophsax).
 
 * `ungroup()` now copies it's input (@christophsax, #54).
@@ -20,7 +22,6 @@
  data.table 1.9.6. Avoids copy and allows joins by different keys (#20, #21,
  @christophsax).
 
-
 # dtplyr 0.0.2
 
 - This is a compatibility release. It makes dtplyr compatible with

R/joins.R

@@ -5,7 +5,10 @@
 #'
 #' @inheritParams dplyr::join
 #' @param x,y tbls to join
-#' @param ... Included for compatibility with generic; otherwise ignored.
+#' @param ... For `inner_join()`, `left_join()`, `right_join()` and
+#' `full_join()` passed on to data.table [merge()] method. For
+#' `semi_join()` and `anti_join()` included only for compatibility with
+#' generic and must be empty.
 #' @examples
 #' library(dplyr, warn.conflicts = FALSE)
 #'
@@ -31,51 +34,56 @@
 #' @name join.tbl_dt
 NULL
 
-join_using_merge <- function(x, y, by, copy, suffix, 
+join_using_merge <- function(x, y, ..., by, copy, suffix,
  all.x = FALSE, all.y = FALSE){
  by <- dplyr::common_by(by, x, y)
  y <- dplyr::auto_copy(x, y, copy = copy)
  out <- merge(
- x, y, 
- by.x = by$x, by.y = by$y, 
+ x, y,
+ by.x = by$x, by.y = by$y,
  all.x = all.x, all.y = all.y,
- suffixes = suffix, 
- allow.cartesian = TRUE
+ suffixes = suffix,
+ allow.cartesian = TRUE,
+ ...
  )
- grouped_dt(out, groups(x)) 
+ grouped_dt(out, groups(x))
 }
 
 #' @rdname join.tbl_dt
-inner_join.data.table <- function(x, y, by = NULL, copy = FALSE, 
- suffix = c(".x", ".y"), ...){
- join_using_merge(x, y, by = by, copy = copy, suffix = suffix)
+inner_join.data.table <- function(x, y, ..., by = NULL, copy = FALSE,
+ suffix = c(".x", ".y")){
+ join_using_merge(x, y, by = by, copy = copy, suffix = suffix, ...)
 }
 
 #' @rdname join.tbl_dt
-left_join.data.table <- function(x, y, by = NULL, copy = FALSE, 
- suffix = c(".x", ".y"), ...){
- join_using_merge(x, y, by = by, copy = copy, suffix = suffix, all.x = TRUE)
+left_join.data.table <- function(x, y, ..., by = NULL, copy = FALSE,
+ suffix = c(".x", ".y")){
+ join_using_merge(x, y, by = by, copy = copy, suffix = suffix, all.x = TRUE, ...)
 }
 
 #' @rdname join.tbl_dt
-right_join.data.table <- function(x, y, by = NULL, copy = FALSE, 
- suffix = c(".x", ".y"), ...){
- join_using_merge(x, y, by = by, copy = copy, suffix = suffix, all.y = TRUE)
+right_join.data.table <- function(x, y, ..., by = NULL, copy = FALSE,
+ suffix = c(".x", ".y")){
+ join_using_merge(x, y, by = by, copy = copy, suffix = suffix, all.y = TRUE, ...)
 }
 
 #' @rdname join.tbl_dt
-full_join.data.table <- function(x, y, by = NULL, copy = FALSE, 
- suffix = c(".x", ".y"), ...){
- join_using_merge(x, y, 
- by = by, 
- copy = copy, 
- suffix = suffix, 
- all.x = TRUE, all.y = TRUE
+full_join.data.table <- function(x, y, ..., by = NULL, copy = FALSE,
+ suffix = c(".x", ".y")){
+ join_using_merge(x, y,
+ by = by,
+ copy = copy,
+ suffix = suffix,
+ all.x = TRUE,
+ all.y = TRUE,
+ ...
  )
 }
 
 #' @rdname join.tbl_dt
 semi_join.data.table <- function(x, y, by = NULL, copy = FALSE, ...) {
+ ellipsis::check_dots_empty()
+
  by <- dplyr::common_by(by, x, y)
  y <- dplyr::auto_copy(x, y, copy = copy)
  on <- set_names(by$y, by$x)
@@ -87,6 +95,8 @@ semi_join.data.table <- function(x, y, by = NULL, copy = FALSE, ...) {
 
 #' @rdname join.tbl_dt
 anti_join.data.table <- function(x, y, by = NULL, copy = FALSE, ...) {
+ ellipsis::check_dots_empty()
+
  by <- dplyr::common_by(by, x, y)
  y <- dplyr::auto_copy(x, y, copy = copy)
  on <- set_names(by$y, by$x)

tests/testthat/test-joins.R

@@ -2,34 +2,21 @@ context("joins")
 
 
 test_that("joining data tables returns data tables (#470) and does not modify them (#659)", {
- a <- data.table(x = c(1, 1, 2, 3), y = 4:1)
- b <- data.table(x = c(1, 2, 2, 4), z = 1:4)
-
- test_join <- function(join_fun, ak, bk) {
- data.table::setkeyv(a, ak)
- data.table::setkeyv(b, bk)
- ac <- data.table::copy(a)
- bc <- data.table::copy(b)
-
- out <- join_fun(a, b, "x")
- expect_is(out, "data.table")
- expect_equal(a, ac)
- expect_equal(b, bc)
- }
-
- for (ak in names(a)) {
- for (bk in names(b)) {
- test_join(left_join, ak, bk)
- test_join(semi_join, ak, bk)
- test_join(right_join, ak, bk)
- test_join(full_join, ak, bk)
- test_join(inner_join, ak, bk)
- test_join(anti_join, ak, bk)
- }
+ x <- data.table(x = c(1, 1, 2, 3), y = 4:1)
+ y <- data.table(x = c(1, 2, 2, 4), z = 1:4)
+
+ join_funs <- list(left_join, semi_join, anti_join)
+ for (join_fun in join_funs) {
+ xprime <- data.table::copy(x)
+ yprime <- data.table::copy(y)
+
+ out <- join_fun(xprime, yprime, "x")
+ expect_s3_class(out, "data.table")
+ expect_equal(xprime, x)
+ expect_equal(yprime, y)
  }
 })
 
-
 test_that("joining data tables returns same result as dplyr", {
  a_dt <- data.table(x = c(1, 1, 2, 3), y = 4:1)
  b_dt <- data.table(x = c(1, 2, 2, 4), z = 1:4)