AppDoctor
Java Python Shell JavaScript C++ CSS Other
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
apptest
dynclass
instrument_server
media
static
tools
.gitignore
3rd-targets-list.txt
LICENSE.txt
README.md
WEBUI
analyze.py
andchecker.keystore
android_test.conf
apk.py
apktool.jar
app.py
apps_100.csv
apps_open.csv
buglib.py
check.sh
check_reach.py
cleanapp.sh
cleanavd.sh
cleanlog.sh
collect.sh
collect_log.sh
config.py
configs_exp.csv
controller.py
count_time.py
crawl.py
device.py
executor.py
flags.py
hints.csv
instrumenter.py
jobs.py
launch.py
launch.sh
libinfo.py
list.sh
log.py
parser.py
query.sh
queryex.sh
rep.sh
replay.py
reproduce.py
run.sh
settings.py
sort.sh
stop.sh
test.py
web_ui.py
web_ui_reproduce_bug
web_ui_scan_logs

README.md

Overview

AppDoctor can automatically test Android applications for problems. It can run test sessions on multiple hosts and analyze logs from the sessions to find and verify bugs. It can also be used as a scripting tool for developers to write automatic test scripts.

Usage

First:

  • generate a keystore
  • edit the config files to prepare for a test

On the testing hosts:

  • upload AppDoctor to ~/andchecker
  • ensure that Android NDK and SDK are available in the PATH
  • ensure that JDK is in the PATH
  • run tools/get.sh to get external tools

On the controlling host:

  • use scripts to start/stop test sessions
  • use analyze script to analyze logs

Config Files

  • andchecker.keystore: java key storage, used to sign apks
  • android_test.conf: test configuration file
  • apps.csv: applications to test
  • configs.csv: device configurations to test
  • hints.csv: hints for text boxes and other widgets
  • hosts.txt: hosts to run tests with
  • settings.py: some constants for the scripts

Scripts

  • run.sh: start a test session on specified hosts
  • check.sh: start a session on local host
  • collect.sh: collect logs from hosts
  • list.sh: check for a clean environment
  • stop.sh: stop the test session on all the hosts
  • analyze.py: analyze logs of test sessions
  • reproduce.py: the underlying building block. Can be used to test applications, reproduce bugs or start a console session to run a step-by-step debug session.
  • controller.py: the test controller running on each host

Directories

  • dynclass/: compile and run custom test commands
  • media/: media files to be transferred to the device for testing
  • tools/: external tools
  • instrument_server/: instrumentation application, run on the device