Github user sun-rui commented on a diff in the pull request:
https://github.com/apache/spark/pull/12836#discussion_r66721611
--- Diff: R/pkg/inst/worker/worker.R ---
@@ -79,75 +127,72 @@ if (numBroadcastVars > 0) {
# Timing broadcast
broadcastElap <- elapsedSecs()
+# Initial input timing
+inputElap <- broadcastElap
# If -1: read as normal RDD; if >= 0, treat as pairwise RDD and treat the
int
# as number of partitions to create.
numPartitions <- SparkR:::readInt(inputCon)
-isDataFrame <- as.logical(SparkR:::readInt(inputCon))
+# 0 - RDD mode, 1 - dapply mode, 2 - gapply mode
+mode <- SparkR:::readInt(inputCon)
-# If isDataFrame, then read column names
-if (isDataFrame) {
+if (mode > 0) {
colNames <- SparkR:::readObject(inputCon)
}
isEmpty <- SparkR:::readInt(inputCon)
+computeInputElapsDiff <- 0
+outputComputeElapsDiff <- 0
if (isEmpty != 0) {
-
if (numPartitions == -1) {
if (deserializer == "byte") {
# Now read as many characters as described in funcLen
data <- SparkR:::readDeserialize(inputCon)
} else if (deserializer == "string") {
data <- as.list(readLines(inputCon))
- } else if (deserializer == "row") {
+ } else if (deserializer == "row" && mode == 2) {
+ dataWithKeys <- SparkR:::readMultipleObjectsWithKeys(inputCon)
+ keys <- dataWithKeys$keys
+ data <- dataWithKeys$data
+ } else if (deserializer == "row"){
--- End diff --
style: (deserializer == "row") {
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
