Repository: spark
Updated Branches:
refs/heads/branch-1.5 881baf100 -> 4f75ce2e1
[SPARK-8844] [SPARKR] head/collect is broken in SparkR.
This is a WIP patch for SPARK-8844 for collecting reviews.
This bug is about reading an empty DataFrame. in readCol(),
lapply(1:numRows, function(x) {
does not take into consideration the case where numRows = 0.
Will add unit test case.
Author: Sun Rui <[email protected]>
Closes #7419 from sun-rui/SPARK-8844.
(cherry picked from commit 5f9ce738fe6bab3f0caffad0df1d3876178cf469)
Signed-off-by: Shivaram Venkataraman <[email protected]>
Project: http://git-wip-us.apache.org/repos/asf/spark/repo
Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/4f75ce2e
Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/4f75ce2e
Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/4f75ce2e
Branch: refs/heads/branch-1.5
Commit: 4f75ce2e193c813f4e3ad067749b6e7b4f0ee135
Parents: 881baf1
Author: Sun Rui <[email protected]>
Authored: Sun Aug 16 00:30:02 2015 -0700
Committer: Shivaram Venkataraman <[email protected]>
Committed: Sun Aug 16 00:30:10 2015 -0700
----------------------------------------------------------------------
R/pkg/R/deserialize.R | 16 ++++++++++------
R/pkg/inst/tests/test_sparkSQL.R | 20 ++++++++++++++++++++
2 files changed, 30 insertions(+), 6 deletions(-)
----------------------------------------------------------------------
http://git-wip-us.apache.org/repos/asf/spark/blob/4f75ce2e/R/pkg/R/deserialize.R
----------------------------------------------------------------------
diff --git a/R/pkg/R/deserialize.R b/R/pkg/R/deserialize.R
index 6d364f7..33bf13e 100644
--- a/R/pkg/R/deserialize.R
+++ b/R/pkg/R/deserialize.R
@@ -176,10 +176,14 @@ readRow <- function(inputCon) {
# Take a single column as Array[Byte] and deserialize it into an atomic vector
readCol <- function(inputCon, numRows) {
- # sapply can not work with POSIXlt
- do.call(c, lapply(1:numRows, function(x) {
- value <- readObject(inputCon)
- # Replace NULL with NA so we can coerce to vectors
- if (is.null(value)) NA else value
- }))
+ if (numRows > 0) {
+ # sapply can not work with POSIXlt
+ do.call(c, lapply(1:numRows, function(x) {
+ value <- readObject(inputCon)
+ # Replace NULL with NA so we can coerce to vectors
+ if (is.null(value)) NA else value
+ }))
+ } else {
+ vector()
+ }
}
http://git-wip-us.apache.org/repos/asf/spark/blob/4f75ce2e/R/pkg/inst/tests/test_sparkSQL.R
----------------------------------------------------------------------
diff --git a/R/pkg/inst/tests/test_sparkSQL.R b/R/pkg/inst/tests/test_sparkSQL.R
index e6d3b21..c77f633 100644
--- a/R/pkg/inst/tests/test_sparkSQL.R
+++ b/R/pkg/inst/tests/test_sparkSQL.R
@@ -408,6 +408,14 @@ test_that("collect() returns a data.frame", {
expect_equal(names(rdf)[1], "age")
expect_equal(nrow(rdf), 3)
expect_equal(ncol(rdf), 2)
+
+ # collect() returns data correctly from a DataFrame with 0 row
+ df0 <- limit(df, 0)
+ rdf <- collect(df0)
+ expect_true(is.data.frame(rdf))
+ expect_equal(names(rdf)[1], "age")
+ expect_equal(nrow(rdf), 0)
+ expect_equal(ncol(rdf), 2)
})
test_that("limit() returns DataFrame with the correct number of rows", {
@@ -492,6 +500,18 @@ test_that("head() and first() return the correct data", {
testFirst <- first(df)
expect_equal(nrow(testFirst), 1)
+
+ # head() and first() return the correct data on
+ # a DataFrame with 0 row
+ df0 <- limit(df, 0)
+
+ testHead <- head(df0)
+ expect_equal(nrow(testHead), 0)
+ expect_equal(ncol(testHead), 2)
+
+ testFirst <- first(df0)
+ expect_equal(nrow(testFirst), 0)
+ expect_equal(ncol(testFirst), 2)
})
test_that("distinct() and unique on DataFrames", {
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
