First Steps¶

Requirements¶

Clickpecker requires:

libtesseract (tested on 3.04) and libleptonica (>= 1.71)

In Ubuntu:

$ apt-get install tesseract-ocr libtesseract-dev libleptonica-dev

For more information check out tesserocr repo.

Android Debug Bridge (ADB)

Path to ADB binary should be in $PATH variable.

OpenSTF minicap and minitouch

Should be launched on testing devices. For more information check out minicap and minitouch repos.

To make installing minicap/minitouch and obtaining devices easier, run Clickpecker Device Manager server.

Installation¶

Install Clickpecker library from source:

pip install git+https://github.com/VladX09/clickpecker-library.git

Getting Started¶

Warning! This version uses only template matching with pyramid representation for image-based cases (e.g. finding and tapping icons). This method is robust to differences in size and resolution of the template icon and device’s screen. But it has got it’s own drawbacks. Current template matcher hasn’t got any thresholds and it can’t detect if there’s no wanted image on the screen. That’s why some methods (e.g. search, wait_for, scroll_for) aren’t accept images.

The main object of Clickpecker is clickpecker.BasicAPI. To init API, you should have DeviceWrapper object and a dictionary with the default configuration.

from clickpecker import BasicAPI
from clickpecker.models import Device
from clickpecker.helpers import DeviceWrapper
from clickpecker.configurations import default_config

device = Device(adb_id = "device_adb_id",
             device_name = "Testing Device",
             android_version = "8.0",
             minicap_port = 1111,
             minitouch_port = 1112)

device_wrapper = DeviceWrapper(device)
api = BasicAPI(device_wrapper, default_config)

If your device is connected, visible by ADB and has got minicap and minitouch launched on appropriate ports, you can start work with it.

Send ADB command:

api.adb("shell pm clear com.your.app")

Tap Next button on device’s screen:

api.tap("Next")

Wait for OK button appears:

api.wait_for("OK")

Scroll down (swipe from the first relative (X,Y) point to the second one) till Next button appears:

api.scroll_for("Next", (0.5, 0.6), (0.5, 0.2))

Tap icon on device’s screen (via template_matching):

from clickpecker.recognition import tm_engine

icon = tm_engine.load_template("icon.png")
api.tap(icon)

Wait for OK button appears in the specific region. Region is set by it’s relative X and Y bounds:

api.tap(
    "OK",
    config=dict(
        api_tap_timeout=60 * 5, # Wait for 5 minutes
        crop_x_range=(0.1, 0.9),
        crop_y_range=(0.4, 0.8)))

Assertions¶

Clickpecker provides different methods to make assertions about the state of an application under testing. These methods are based on the human way of visual observing.

First of all, text-based UI assertions can be made by clickpecker.BasicAPI.search() method:

assert len(api.search("OK")) != 0  # Assert OK button presents on screen

Also there are methods to assert that device’s screen should (or shouldn’t) change after some actions:

with api.assert_screen_change()
    api.tap("OK")


import time
with api.assert_screen_same()
    time.sleep(50)

In common case, it’s possible to compare current screenshot with the previously saved one:

tag = "State 1"
api.save_current_screen(tag)
# do something
assert api.screen_similar_with(tag)