From e14c3289f65809fbf5dcd4513954e0bbf92a1b2b Mon Sep 17 00:00:00 2001
From: cafreeman <cfreeman@alteryx.com>
Date: Tue, 3 Mar 2015 16:04:55 -0600
Subject: [PATCH 1/5] `selectExpr`

---
 pkg/NAMESPACE                  |  1 +
 pkg/R/DataFrame.R              | 29 +++++++++++++++
 pkg/inst/tests/test_sparkSQL.R | 67 ++++++++++++++++++++--------------
 3 files changed, 69 insertions(+), 28 deletions(-)

diff --git a/pkg/NAMESPACE b/pkg/NAMESPACE
index 26d0690a3cac0..a2831f04c5a26 100644
--- a/pkg/NAMESPACE
+++ b/pkg/NAMESPACE
@@ -100,6 +100,7 @@ exportMethods("columns",
               "schema",
               "sortDF",
               "select",
+              "selectExpr",
               "toRDD",
               "where")
 
diff --git a/pkg/R/DataFrame.R b/pkg/R/DataFrame.R
index 8da89e54f00bf..8daae8d694cf4 100644
--- a/pkg/R/DataFrame.R
+++ b/pkg/R/DataFrame.R
@@ -607,6 +607,35 @@ setMethod("select", signature(x = "DataFrame", col = "Column"),
             dataFrame(sdf)
           })
 
+
+#' SelectExpr
+#'
+#' Select from a DataFrame using a set of SQL expressions.
+#'
+#' @param x A DataFrame to be sorted.
+#' @param expr A string containing a SQL expression
+#' @param ... Additional expressions
+#' @return A DataFrame
+#' @rdname selectExpr
+#' @export
+#' @examples
+#'\dontrun{
+#' sc <- sparkR.init()
+#' sqlCtx <- sparkRSQL.init(sc)
+#' path <- "path/to/file.json"
+#' df <- jsonFile(sqlCtx, path)
+#' selectExpr(df, "col1", "(col2 * 5) as newCol")
+#' }
+setGeneric("selectExpr", function(x, expr, ...) { standardGeneric("selectExpr") })
+
+setMethod("selectExpr",
+          signature(x = "DataFrame", expr = "character"),
+          function(x, expr, ...) {
+            exprList <- list(expr, ...)
+            sdf <- callJMethod(x@sdf, "selectExpr", listToSeq(exprList))
+            dataFrame(sdf)
+          })
+
 #' SortDF 
 #'
 #' Sort a DataFrame by the specified column(s).
diff --git a/pkg/inst/tests/test_sparkSQL.R b/pkg/inst/tests/test_sparkSQL.R
index 26450eca63dd1..68ef8d0dc56a3 100644
--- a/pkg/inst/tests/test_sparkSQL.R
+++ b/pkg/inst/tests/test_sparkSQL.R
@@ -231,6 +231,17 @@ test_that("select with column", {
   expect_true(count(df2) == 3)
 })
 
+test_that("selectExpr() on a DataFrame", {
+  df <- jsonFile(sqlCtx, jsonPath)
+  selected <- selectExpr(df, "age * 2")
+  expect_true(names(selected) == "age * 2")
+  expect_equal(collect(selected), collect(select(df, df$age * 2L)))
+  
+  selected2 <- selectExpr(df, "name as newName", "abs(age) as age")
+  expect_equal(names(selected2), c("newName", "age"))
+  expect_true(count(selected2) == 3)
+})
+
 test_that("column calculation", {
   df <- jsonFile(sqlCtx, jsonPath)
   d <- collect(select(df, alias(df$age + 1, "age2")))
@@ -282,34 +293,34 @@ test_that("filter() on a DataFrame", {
 })
 
 test_that("join() on a DataFrame", {
-df <- jsonFile(sqlCtx, jsonPath)
-
-mockLines2 <- c("{\"name\":\"Michael\", \"test\": \"yes\"}",
-                "{\"name\":\"Andy\",  \"test\": \"no\"}",
-                "{\"name\":\"Justin\", \"test\": \"yes\"}",
-                "{\"name\":\"Bob\", \"test\": \"yes\"}")
-jsonPath2 <- tempfile(pattern="sparkr-test", fileext=".tmp")
-writeLines(mockLines2, jsonPath2)
-df2 <- jsonFile(sqlCtx, jsonPath2)
-
-joined <- join(df, df2)
-expect_equal(names(joined), c("age", "name", "name", "test"))
-expect_true(count(joined) == 12)
-
-joined2 <- join(df, df2, df$name == df2$name)
-expect_equal(names(joined2), c("age", "name", "name", "test"))
-expect_true(count(joined2) == 3)
-
-joined3 <- join(df, df2, df$name == df2$name, "right_outer")
-expect_equal(names(joined3), c("age", "name", "name", "test"))
-expect_true(count(joined3) == 4)
-expect_true(is.na(collect(joined3)$age[4]))
-
-joined4 <- select(join(df, df2, df$name == df2$name, "outer"),
-                  alias(df$age + 5, "newAge"), df$name, df2$test)
-expect_equal(names(joined4), c("newAge", "name", "test"))
-expect_true(count(joined4) == 4)
-expect_true(first(joined4)$newAge == 24)
+  df <- jsonFile(sqlCtx, jsonPath)
+  
+  mockLines2 <- c("{\"name\":\"Michael\", \"test\": \"yes\"}",
+                  "{\"name\":\"Andy\",  \"test\": \"no\"}",
+                  "{\"name\":\"Justin\", \"test\": \"yes\"}",
+                  "{\"name\":\"Bob\", \"test\": \"yes\"}")
+  jsonPath2 <- tempfile(pattern="sparkr-test", fileext=".tmp")
+  writeLines(mockLines2, jsonPath2)
+  df2 <- jsonFile(sqlCtx, jsonPath2)
+  
+  joined <- join(df, df2)
+  expect_equal(names(joined), c("age", "name", "name", "test"))
+  expect_true(count(joined) == 12)
+  
+  joined2 <- join(df, df2, df$name == df2$name)
+  expect_equal(names(joined2), c("age", "name", "name", "test"))
+  expect_true(count(joined2) == 3)
+  
+  joined3 <- join(df, df2, df$name == df2$name, "right_outer")
+  expect_equal(names(joined3), c("age", "name", "name", "test"))
+  expect_true(count(joined3) == 4)
+  expect_true(is.na(collect(joined3)$age[4]))
+  
+  joined4 <- select(join(df, df2, df$name == df2$name, "outer"),
+                    alias(df$age + 5, "newAge"), df$name, df2$test)
+  expect_equal(names(joined4), c("newAge", "name", "test"))
+  expect_true(count(joined4) == 4)
+  expect_true(first(joined4)$newAge == 24)
 })
 
 unlink(jsonPath)

From 494a4ddef3bc39a8caa848c931033e1a6f6911bf Mon Sep 17 00:00:00 2001
From: cafreeman <cfreeman@alteryx.com>
Date: Tue, 3 Mar 2015 16:06:06 -0600
Subject: [PATCH 2/5] update export

---
 pkg/R/DataFrame.R | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/pkg/R/DataFrame.R b/pkg/R/DataFrame.R
index 8daae8d694cf4..b8b118da0e6d2 100644
--- a/pkg/R/DataFrame.R
+++ b/pkg/R/DataFrame.R
@@ -628,6 +628,8 @@ setMethod("select", signature(x = "DataFrame", col = "Column"),
 #' }
 setGeneric("selectExpr", function(x, expr, ...) { standardGeneric("selectExpr") })
 
+#' @rdname selectExpr
+#' @export
 setMethod("selectExpr",
           signature(x = "DataFrame", expr = "character"),
           function(x, expr, ...) {

From acea1468a9c71ea5d711e0a6845b79d4ad6c8b21 Mon Sep 17 00:00:00 2001
From: cafreeman <cfreeman@alteryx.com>
Date: Tue, 3 Mar 2015 16:20:55 -0600
Subject: [PATCH 3/5] remove extra line

---
 pkg/R/DataFrame.R | 1 -
 1 file changed, 1 deletion(-)

diff --git a/pkg/R/DataFrame.R b/pkg/R/DataFrame.R
index b8b118da0e6d2..09adfce07f5ad 100644
--- a/pkg/R/DataFrame.R
+++ b/pkg/R/DataFrame.R
@@ -607,7 +607,6 @@ setMethod("select", signature(x = "DataFrame", col = "Column"),
             dataFrame(sdf)
           })
 
-
 #' SelectExpr
 #'
 #' Select from a DataFrame using a set of SQL expressions.

From 79186340c2095ba6ec32b06ebb1a3daffd72b486 Mon Sep 17 00:00:00 2001
From: cafreeman <cfreeman@alteryx.com>
Date: Tue, 3 Mar 2015 16:22:31 -0600
Subject: [PATCH 4/5] Fix test

---
 pkg/inst/tests/test_sparkSQL.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pkg/inst/tests/test_sparkSQL.R b/pkg/inst/tests/test_sparkSQL.R
index 68ef8d0dc56a3..539fde06e0108 100644
--- a/pkg/inst/tests/test_sparkSQL.R
+++ b/pkg/inst/tests/test_sparkSQL.R
@@ -234,7 +234,7 @@ test_that("select with column", {
 test_that("selectExpr() on a DataFrame", {
   df <- jsonFile(sqlCtx, jsonPath)
   selected <- selectExpr(df, "age * 2")
-  expect_true(names(selected) == "age * 2")
+  expect_true(names(selected) == "(age * 2)")
   expect_equal(collect(selected), collect(select(df, df$age * 2L)))
   
   selected2 <- selectExpr(df, "name as newName", "abs(age) as age")

From bc901159e30ef3bebb2b8a7a36c9abfb3b1cf230 Mon Sep 17 00:00:00 2001
From: cafreeman <cfreeman@alteryx.com>
Date: Wed, 4 Mar 2015 08:33:00 -0600
Subject: [PATCH 5/5] Fixed docs

---
 pkg/R/DataFrame.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pkg/R/DataFrame.R b/pkg/R/DataFrame.R
index 09adfce07f5ad..704470cd5210c 100644
--- a/pkg/R/DataFrame.R
+++ b/pkg/R/DataFrame.R
@@ -611,7 +611,7 @@ setMethod("select", signature(x = "DataFrame", col = "Column"),
 #'
 #' Select from a DataFrame using a set of SQL expressions.
 #'
-#' @param x A DataFrame to be sorted.
+#' @param x A DataFrame to be selected from.
 #' @param expr A string containing a SQL expression
 #' @param ... Additional expressions
 #' @return A DataFrame