2012-08-10 04:10:02 -04:00
|
|
|
import os
|
2012-12-28 12:06:11 -05:00
|
|
|
import sys
|
2012-12-28 01:47:37 -05:00
|
|
|
from subprocess import Popen, PIPE
|
|
|
|
from threading import Thread
|
|
|
|
from py4j.java_gateway import java_import, JavaGateway, GatewayClient
|
2012-08-10 04:10:02 -04:00
|
|
|
|
|
|
|
|
|
|
|
SPARK_HOME = os.environ["SPARK_HOME"]
|
|
|
|
|
|
|
|
|
|
|
|
def launch_gateway():
|
2012-12-28 01:47:37 -05:00
|
|
|
# Launch the Py4j gateway using Spark's run command so that we pick up the
|
|
|
|
# proper classpath and SPARK_MEM settings from spark-env.sh
|
|
|
|
command = [os.path.join(SPARK_HOME, "run"), "py4j.GatewayServer",
|
|
|
|
"--die-on-broken-pipe", "0"]
|
|
|
|
proc = Popen(command, stdout=PIPE, stdin=PIPE)
|
|
|
|
# Determine which ephemeral port the server started on:
|
|
|
|
port = int(proc.stdout.readline())
|
|
|
|
# Create a thread to echo output from the GatewayServer, which is required
|
|
|
|
# for Java log output to show up:
|
|
|
|
class EchoOutputThread(Thread):
|
|
|
|
def __init__(self, stream):
|
|
|
|
Thread.__init__(self)
|
|
|
|
self.daemon = True
|
|
|
|
self.stream = stream
|
|
|
|
|
|
|
|
def run(self):
|
|
|
|
while True:
|
|
|
|
line = self.stream.readline()
|
2012-12-28 12:06:11 -05:00
|
|
|
sys.stderr.write(line)
|
2012-12-28 01:47:37 -05:00
|
|
|
EchoOutputThread(proc.stdout).start()
|
|
|
|
# Connect to the gateway
|
2012-12-29 01:19:12 -05:00
|
|
|
gateway = JavaGateway(GatewayClient(port=port), auto_convert=False)
|
2012-12-28 01:47:37 -05:00
|
|
|
# Import the classes used by PySpark
|
2012-08-10 04:10:02 -04:00
|
|
|
java_import(gateway.jvm, "spark.api.java.*")
|
|
|
|
java_import(gateway.jvm, "spark.api.python.*")
|
|
|
|
java_import(gateway.jvm, "scala.Tuple2")
|
|
|
|
return gateway
|