Android's MonkeyRunner occasionally throws exceptions - python

I am running an automated test using an Android emulator driving an app with a Monkey script written in Python.
The script is copying files onto the emulator, clicks buttons in the app and reacts depending on the activities that the software triggers during its operation. The script is supposed to be running the cycle a few thousand times so I have this in a loop to run the adb tool to copy the files, start the activities and see how the software is reacting by calling the getProperty method on the device with the parameter 'am.current.comp.class'.
So here is a very simplified version of my script:
for target in targets:
androidSDK.copyFile(emulatorName, target, '/mnt/sdcard')
# Runs the component
device.startActivity(component='com.myPackage/com.myPackage.myactivity')
while 1:
if device.getProperty('am.current.comp.class') == 'com.myPackage.anotheractivity':
time.sleep(1) # to allow the scree to display the new activity before I click on it
device.touch(100, 100, 'DOWN_AND_UP')
# Log the result of the operation somewhere
break
time.sleep(0.1)
(androidSDK is a small class I've written that wraps some utility functions to copy and delete files using the adb tool).
On occasions the script crashes with one of a number of exceptions, for instance (I am leaving out the full stack trace)
[com.android.chimpchat.adb.AdbChimpDevice]com.android.ddmlib.ShellCommandUnresponsiveException
or
[com.android.chimpchat.adb.AdbChimpDevice] Unable to get variable: am.current.comp.class
[com.android.chimpchat.adb.AdbChimpDevice]java.net.SocketException: Software caused connectionabort: socket write error
I have read that sometimes the socket connection to the device becomes unstable and may need a restart (adb start-server and adb kill-server come in useful).
The problem I'm having is that the tools are throwing Java exceptions (Monkey runs in Jython), but I am not sure how those can be trapped from within my Python script. I would like to be able to determine the exact cause of the failure inside the script and recover the situation so I can carry on with my iterations (re-establish the connection, for instance? Would for instance re-initialising my device with another call to MonkeyRunner.waitForConnection be enough?).
Any ideas?
Many thanks,
Alberto
EDIT. I thought I'd mention that I have discovered that it is possible to catch Java-specific exceptions in a Jython script, should anyone need this:
from java.net import SocketException
...
try:
...
except(SocketException):
...

It is possible to catch Java-specific exceptions in a Jython script:
from java.net import SocketException
...
try:
...
except(SocketException):
...
(Taken from OP's edit to his question)

This worked for me:
device.shell('exit')# Exit the shell

Related

Steam browser protocol failing silently when run over ssh

I am trying to launch a steam game on my computer through an ssh connection (into a Win10 machine). When run locally, the following python call works.
subprocess.run("start steam://rungameid/[gameid]", shell=True)
However, whenever I run this over an ssh connection—either in an interactive interpreter or by invoking a script on the target machine—my steam client suddenly exits.
I haven't noticed anything in the steam logs except that Steam\logs\connection_log.txt contains logoff and a new session start each time. This is not the case when I run the command locally on my machine. Why is steam aware of the different sources of this command, and why is this causing the steam connection to drop? Can anyone suggest a workaround?
Thanks.
Steam is likely failing to launch the application because Windows services, including OpenSSH server, cannot access the desktop, and, hence, cannot launch GUI applications. Presumably, Steam does not expect to run an application in an environment in which it cannot interact with the desktop, and this is what eventually causes Steam to crash. (Admittedly, this is just a guess—it's hard to be sure exactly what is happening when the crash does not seem to appear in the logs or crash dumps.)
You can see a somewhat more detailed explanation of why starting GUI applications over SSH fails when the server is run as a Windows service in this answer by domih to this question about running GUI applications over SSH on Windows.
domih also suggests some workarounds. If it is an option for you, the simplest one is probably to download and run OpenSSH server manually instead of running the server as a service. You can find the latest release of Win32-OpenSSH/Windows for OpenSSH here.
The other workaround that still seems to work is to use schtasks. The idea is to create a scheduled task that runs your command—the Task Scheduler can access the desktop. Unfortunately, this is only an acceptable solution if you don't mind waiting until the next minute at least; schtasks can only schedule tasks to occur exactly on the minute. Moreover, to be safe to run at any time, code should probably schedule the task for at least one minute into the future, meaning that wait times could be anywhere between 1–2 minutes.
There are also other drawbacks to this approach. For example, it's probably harder to monitor the running process this way. However, it might be an acceptable solution in some circumstances, so I've written some Python code that can be used to run a program with schtasks, along with an example. The code depends on the the shortuuid package; you will need to install it before trying the example.
import subprocess
import tempfile
import shortuuid
import datetime
def run_with_schtasks_soon(s, delay=2):
"""
Run a program with schtasks with a delay of no more than
delay minutes and no less than delay - 1 minutes.
"""
# delay needs to be no less than 2 since, at best, we
# could be calling subprocess at the end of the minute.
assert delay >= 2
task_name = shortuuid.uuid()
temp_file = tempfile.NamedTemporaryFile(mode="w", suffix=".bat", delete=False)
temp_file.write('{}\nschtasks /delete /tn {} /f\ndel "{}"'.format(s, task_name, temp_file.name))
temp_file.close()
run_time = datetime.datetime.now() + datetime.timedelta(minutes=delay)
time_string = run_time.strftime("%H:%M")
# This is locale-specific. You will need to change this to
# match your locale. (locale.setlocale and the "%x" format
# does not seem to work here)
date_string = run_time.strftime("%m/%d/%Y")
return subprocess.run("schtasks /create /tn {} /tr {} /sc once /st {} /sd {}".format(task_name,
temp_file.name,
time_string,
date_string),
shell=True)
if __name__ == "__main__":
# Runs The Witness (if you have it)
run_with_schtasks_soon("start steam://rungameid/210970")

Where do messages go when using pythoncom.PumpMessages()?

I would like to know where does the function pythoncom.PumpMessage() stores the message when it comes into play. I was going through a site and just saw a Python Script for Key logger, I copied the code and used it on my computer but I don't feel safe, after deleting that code I think it's still running in the background and copying my keypresses. Is it so, please help.
If the python script was sending messages to anyone, you'd probably be able to tell by looking at the code. Somewhere there would be a block of code sending some file to a server.
As for PumpMessages(), it
Pumps all messages for the current thread until a WM_QUIT message.
According to this documentation. You can find other answered questions on this with a Google search as well. Pythoncom isn't sending any information to an external source in and of itself, however your script might be. You'd have to check your code for a block doing that.
As for it running after you delete the script, highly unlikely. If you didn't manually or programmatically (in the script) attach the Python script to some daemon, it isn't manifesting itself anywhere. You may be over-estimating the power of a basic Python keylogger. I wouldn't worry.
I've tried keyloggers before with pythoncom, and I've never had a problem. You can also look in the pythoncom library to find the PumpMessages() function and see if that is sending anything.

Python-Twisted Reactor Starting too Early

I have an application that uses PyQt4 and python-twisted to maintain a connection to another program. I am using "qt4reactor.py" as found here. This is all packaged up using py2exe. The application works wonderfully for 99% of users, but one user has reported that networking is failing completely on his Windows system. No other users report the issue, and I cannot replicate it on my own Windows VM. The user reports no abnormal configuration.
The debugging logs show that the reactor.connectTCP() call is executing immediately, even though the reactor hasn't been started yet! There's no mistaking run order because this is a single-threaded process with 60 sec of computation and multiple log messages between this line and when the reactor is supposed to start.
There's a lot of code, so I am only putting in pseudo-code, hoping that there is a general solution for this issue. I will link to the actual code below it.
import qt4reactor
qt4reactor.install()
# Start setting up main window
# ...
from twisted.internet import reactor
# Separate listener for detecting/processing multiple instances
self.InstanceListener = ListenerFactory(...)
reactor.listenTCP(LISTEN_PORT, self.InstanceListener)
# The active/main connection
self.NetworkingFactory = ClientFactory(...)
reactor.connectTCP(ACTIVE_IP, ACTIVE_PORT, self.NetworkingFactory)
# Finish setting up main window
# ...
from twisted.internet import reactor
reactor.runReturn()
The code is nested throughout the Armory project files. ArmoryQt.py (containing the above code) and armoryengine.py (containing the ReconnectingClientFactory subclass used for this connection).
So, the reactor.connectTCP() call executes immediately. The client code executes the send command and then immediately connectionLost() gets called. It does not appear to try to reconnect. It also doesn't throw any errors other than connectionLost(). Even more mysteriously, it receives messages from the remote node later on, and this app even processes them! But it believes it's not connected (and handshake never finished, so the remote node shouldn't be sending messages, but might be a bug/oversight in that program).
What on earth is going on!? How could the reactor get started before I tell it to start? I searched the code and found no other code that (I believe) could start the reactor.
The API that you're looking for is twisted.internet.reactor.callWhenRunning.
However, it wouldn't hurt to have less than 60 seconds of computation at startup, either :). Perhaps you should spread that out, or delegate it to a thread, if it's relatively independent?

Why does this Python service stop by itself in spite of an infinite loop?

I managed to install a Python script as a service using this recipe at ActiveState: win-services-helper.
In order to get much use out of it, I included the business end of my program by replacing lines 99 - 100:
95 class Test(Service):
96 def start(self):
97 self.runflag=True
98 while self.runflag:
99 # My logging script
100 # goes here
101 def stop(self):
102 self.runflag=False
103 self.log("I'm done")
105
106 instart(Test, 'aTest', 'Python Service Test')
My problem is, the service logs only one entry and then stops by itself. Then I go to services.msc, attempt to start the service again, then Windows displays this message:
The Python Service Test service on Local Computer started and then stopped. Some services stop automatically when they have no work to do, for example, the Performance Logs and Alerts service.
How come my service thinks it has nothing to do when I placed it in an infinite loop? It should stop only when I shut down the computer and start working again the next time the computer is turned on, even without anyone logging in.
NOTE: My logging script uses simple file read/write to csv. It doesn't use any fancy logging modules.
I believe that this message appears too when an error is encountered. It's not unlikely that the user running the service hasn't got the rights to write in the log file. You might try to create that csv file and set up the access rights so that "Everyone" can write to it.
It works now. Thanks everyone for the insights!
The recipe consisted of two scripts:
winservice.py
winservice_test.py
In my previous attempt, I picked out some lines which I thought were useful. I might have missed some of them, hence it didn't work.
I left winservice.py as it is. For the business end of my script, I placed it in a loop structure in winservice_test.py.
After writing the code, I went to the command line, browsed to where the two scripts were stored, then installed the service with
python winservice_test.py
My service can now be accessed through services.msc. By default, it runs as Local system account, which presents problems upon logging out. The service will still continue, but it will grind to a halt because the account doesn't have write access to the log folder.
I modified its Properties so that it runs with my user account (which has write access to the log folder). My service now survives logouts, stops working only upon shutdown, and also starts itself as soon as Windows boots up even without anyone logging in.
You need to see on the entire code when you determine the service stop, and you can user other tools like python daemonize to control your codes.

How can I detect what other copy of Python script is already running

I have a script. It uses GTK. And I need to know if another copy of scrip starts. If it starts window will extend.
Please, tell me the way I can detect it.
You could use a D-Bus service. Your script would start a new service if none is found running in the current session, and otherwise send a D-Bus message to the running instace (that can send "anything", including strings, lists, dicts).
The GTK-based library libunique (missing Python bindings?) uses this approach in its implementation of "unique" applications.
You can use a PID file to determine if the application is already running (just search for "python daemon" on Google to find some working implementations).
If you detected that the program is already running, you can communicate with the running instance using named pipes.
The new copy could search for running copies, fire a SIGUSER signal and trigger a callback in your running process that then handles all the magic.
See the signal library for details and the list of things that can go wrong.
I've done that using several ways depending upon the scenario
In one case my script had to listen on a TCP port. So I'd just see if the port was available it'd mean it is a new copy. This was sufficient for me but in certain cases, if the port is already in use, it might be because some other kind of application is listening on that port. You can use OS calls to find out who is listening on the port or try sending data and checking the response.
In another case I used PID file. Just decide a location and a filename, and everytime your script starts, read that file to get a PID. If that PID is running, it means another copy is already there. Otherwise create that file and write your process ID in it. This is pretty simple. If you are using django then you can simply use django's daemonizer: "from django.utils import daemonize". Otherwise you can use this script: http://www.jejik.com/articles/2007/02/a_simple_unix_linux_daemon_in_python/

Categories