drbridgewater

    // deep joins are easy using scoobi.
    // for instance if you have a function that computes your most frequent queries in hadoop and you want to 
    // return only the search session objects (which could have tons of event data other than searches) 
    // that contain at least one of those queries you can do it  like this.
    
    val sessions_w_pop_queries = listOfPopQueries.join(sessions).filter {
      case (pop_queries, session) => session.justSearchPages.flatMap { _.keyword }.toSet.intersect(pop_queries).nonEmpty
    }.map { case (pop_queries, session) => session }

drbridgewater

K <- 500
N <- 10000

# sort the normal random variable this help to see clearly that stream
# sampling does not care which observations come first
data <- sort(rnorm(N))

res <- numeric(0)
for (i in 1:K) {
    res <- c(res, data[i])

drbridgewater

pkgs <- c(
"data.table",

"ggplot2",
"scales",
"googleVis",
"igraph",
"gplots",
"ROCR",
"knitr",

drbridgewater

for(x <- 1 to 2 ; y <- 3 to 4 ; z<- 5 to 6 ) println(x,y,z)

(1,3,5)
(1,3,6)
(1,4,5)
(1,4,6)
(2,3,5)
(2,3,6)
(2,4,5)
(2,4,6)