SimpleApp.R

# SparkR [1] version of the Spark quickstart SimplApp example [2] 

library(SparkR)
sc <- sparkR.init(master="local")

filter <- function(rdd, pred) { flatMap(rdd, function(x) { if (pred(x)) { list(x) } else { list() } })}

logFile <- "spark-0.9.0-incubating-bin-hadoop2/readme.md"

logData <- textFile(sc, logFile)

numAs <- length(filter(logData, function(s) { grepl("a", s) }))
numBs <- length(filter(logData, function(s) { grepl("b", s) }))

cat(sprintf("Lines with a: %d, lines with b: %d",numAs, numBs))


# [1] SparkR: http://amplab-extras.github.io/SparkR-pkg/
# [2] Spark quickstart SimpleApp example: https://spark.incubator.apache.org/docs/latest/quick-start.html