blob: 9e0a01b5f9fb0357ca5eb0f599ccc2e567aef83b (
plain) (
blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
|
package spark
import java.io.PrintWriter
import java.util.StringTokenizer
import scala.collection.Map
import scala.collection.JavaConversions._
import scala.collection.mutable.ArrayBuffer
import scala.io.Source
/**
* An RDD that pipes the contents of each parent partition through an external command
* (printing them one per line) and returns the output as a collection of strings.
*/
class PipedRDD[T: ClassManifest](
parent: RDD[T], command: Seq[String], envVars: Map[String, String])
extends RDD[String](parent.context) {
def this(parent: RDD[T], command: Seq[String]) = this(parent, command, Map())
// Similar to Runtime.exec(), if we are given a single string, split it into words
// using a standard StringTokenizer (i.e. by spaces)
def this(parent: RDD[T], command: String) = this(parent, PipedRDD.tokenize(command))
override def splits = parent.splits
override val dependencies = List(new OneToOneDependency(parent))
override def compute(split: Split): Iterator[String] = {
val pb = new ProcessBuilder(command)
// Add the environmental variables to the process.
val currentEnvVars = pb.environment()
envVars.foreach { case(variable, value) => currentEnvVars.put(variable, value) }
val proc = pb.start()
val env = SparkEnv.get
// Start a thread to print the process's stderr to ours
new Thread("stderr reader for " + command) {
override def run() {
for(line <- Source.fromInputStream(proc.getErrorStream).getLines) {
System.err.println(line)
}
}
}.start()
// Start a thread to feed the process input from our parent's iterator
new Thread("stdin writer for " + command) {
override def run() {
SparkEnv.set(env)
val out = new PrintWriter(proc.getOutputStream)
for(elem <- parent.iterator(split)) {
out.println(elem)
}
out.close()
}
}.start()
// Return an iterator that read lines from the process's stdout
Source.fromInputStream(proc.getInputStream).getLines
}
}
object PipedRDD {
// Split a string into words using a standard StringTokenizer
def tokenize(command: String): Seq[String] = {
val buf = new ArrayBuffer[String]
val tok = new StringTokenizer(command)
while(tok.hasMoreElements)
buf += tok.nextToken()
buf
}
}
|
