Merge pull request #5 from eldondev/master

A couple of updates and intelligent defaults.

README.md

@@ -10,9 +10,9 @@ Install via pip from this repo.
 Run
 =====
 "umbra" script should be in bin/.
-Currently we create a browser and navigate to a single page,
+load_url.py takes urls as arguments and puts them onto a rabbitmq queue
-then exit. We are receiving the network events, but don't yet
+dump_queue.py prints resources discovered by the browser and sent over the return queue.
-have anything to do with them. AMQP/RabbitMQ integration is next.
-
 
+On ubuntu, rabbitmq install with `sudo apt-get install rabbitmq-server` should automatically
+be set up for these three scripts to function on localhost ( the default amqp url ).
 

bin/dump_queue.py

@@ -15,7 +15,7 @@ def print_and_ack(body, message):
     print(body['url'])
     message.ack()
 
-with Connection(sys.argv[1]) as conn:
+with Connection(sys.argv[1] if len(sys.argv) > 1 else "amqp://guest:guest@localhost:5672//") as conn:
     with conn.Consumer(requests_queue, callbacks=[print_and_ack]) as consumer:
         while True:
             conn.drain_events()

umbra/umbra.py

@@ -70,7 +70,7 @@ class Umbra:
 
 		with self.browser_lock:
 			self.send_command(method="Network.enable")
-			self.send_command(method="Runtime.evaluate", params={"expression":"document.location = '%s';" % url})
+			self.send_command(method="Page.navigate", params={"url": url})
 
 			# XXX more logic goes here
 			time.sleep(10)