sandwich: Implements network condition on WPR server and browser.

tools/android/loading/trace_recorder.py

Index: tools/android/loading/trace_recorder.py
diff --git a/tools/android/loading/trace_recorder.py b/tools/android/loading/trace_recorder.py
index 6bcdf5afa77d8616eec92e59990fa461cdb4de54..3eaa788c21854e075abc4ba332fad980a0f3c5ac 100755
--- a/tools/android/loading/trace_recorder.py
+++ b/tools/android/loading/trace_recorder.py
@@ -31,13 +31,15 @@ import tracing
 
 
 def MonitorUrl(connection, url, clear_cache=False,
-               categories=tracing.DEFAULT_CATEGORIES):
+               categories=tracing.DEFAULT_CATEGORIES,
+               timeout=devtools_monitor.DEFAULT_TIMEOUT):

[mattcary, 2016/02/19 15:40:17]

As mentioned elsewhere, this timeout should really be a connection configuration
detail. That would avoid plumbing it around (the timeout is really a
characteristic of our particular connection and not something that MonitorUrl
needs to know about).

[gabadie, 2016/02/22 10:05:00]

Please see the other comment about this issue.

   """Monitor a URL via a trace recorder.
 
   Args:
-    connection: A device_monitor.DevToolsConnection instance.
+    connection: A devtools_monitor.DevToolsConnection instance.
     url: url to navigate to as string.
     clear_cache: boolean indicating if cache should be cleared before loading.
+    timeout: Websocket timeout.
 
   Returns:
     loading_trace.LoadingTrace.
@@ -49,7 +51,7 @@ def MonitorUrl(connection, url, clear_cache=False,
   if clear_cache:
     connection.ClearCache()
   connection.SendAndIgnoreResponse('Page.navigate', {'url': url})
-  connection.StartMonitoring()
+  connection.StartMonitoring(timeout=timeout)
   metadata = {'date': datetime.datetime.utcnow().isoformat(),
               'seconds_since_epoch': time.time()}
   return loading_trace.LoadingTrace(url, metadata, page, request, trace)