An extension library for adding Selenium 3.0 draft and Mobile JSON Wire Protocol Specification draft functionality to the Python language bindings, for use with the mobile testing framework Appium.
There are two ways to install and use the Appium Python client.
-
Install from PyPi, as 'Appium-Python-Client'.
pip install Appium-Python-Client
-
Install from source, via PyPi. From 'Appium-Python-Client', download and unarchive the source tarball (Appium-Python-Client-X.X.tar.gz).
tar -xvf Appium-Python-Client-X.X.tar.gz cd Appium-Python-Client-X.X python setup.py install
-
Install from source via GitHub.
git clone [email protected]:appium/python-client.git cd python-client python setup.py install
The Appium Python Client is fully compliant with the Selenium 3.0 specification draft, with some helpers to make mobile testing in Python easier. The majority of the usage remains as it has been for Selenium 2 (WebDriver), and as the official Selenium Python bindings begins to implement the new specification that implementation will be used underneath, so test code can be written that is utilizable with both bindings.
To use the new functionality now, and to use the superset of functions, instead of
including the Selenium webdriver
module in your test code, use that from
Appium instead.
from appium import webdriver
From there much of your test code will work with no change.
As a base for the following code examples, the following sets up the UnitTest environment:
# Android environment
import unittest
from appium import webdriver
desired_caps = {}
desired_caps['platformName'] = 'Android'
desired_caps['platformVersion'] = '4.2'
desired_caps['deviceName'] = 'Android Emulator'
desired_caps['app'] = PATH('../../../apps/selendroid-test-app.apk')
self.driver = webdriver.Remote('http://localhost:4723/wd/hub', desired_caps)
# iOS environment
import unittest
from appium import webdriver
desired_caps = {}
desired_caps['platformName'] = 'iOS'
desired_caps['platformVersion'] = '7.1'
desired_caps['deviceName'] = 'iPhone Simulator'
desired_caps['app'] = PATH('../../apps/UICatalog.app.zip')
self.driver = webdriver.Remote('http://localhost:4723/wd/hub', desired_caps)
The methods that do change are...
For mobile testing the Selnium methods for switching between windows was previously commandeered for switching between native applications and webview contexts. Methods explicitly for this have been added to the Selenium 3 specification, so moving forward these 'context' methods are to be used.
To get the current context, rather than calling driver.current_window_handle
you
use
current = driver.current_context
The available contexts are not retrieved using driver.window_handles
but with
driver.contexts
Finally, to switch to a new context, rather than driver.switch_to.window(name)
,
use the comparable context method
context_name = "WEBVIEW_1"
driver.switch_to.context(context_name)
This allows elements in iOS applications to be found using recursive element
search using the UIAutomation library. Adds the methods driver.find_element_by_ios_uiautomation
and driver.find_elements_by_ios_uiautomation
.
el = self.driver.find_element_by_ios_uiautomation('.elements()[0]')
self.assertEqual('UICatalog', el.get_attribute('name'))
els = self.driver.find_elements_by_ios_uiautomation('.elements()')
self.assertIsInstance(els, list)
This allows elements in an Android application to be found using recursive element
search using the UIAutomator library. Adds the methods driver.find_element_by_android_uiautomator
and driver.find_elements_by_android_uiautomator
.
el = self.driver.find_element_by_android_uiautomator('new UiSelector().description("Animation")')
self.assertIsNotNone(el)
els = self.driver.find_elements_by_android_uiautomator('new UiSelector().clickable(true)')
self.assertIsInstance(els, list)
Allows for elements to be found using the "Accessibility ID". The methods take a
string representing the accessibility id or label attached to a given element, e.g., for iOS the accessibility identifier and for Android the content-description. Adds the methods
driver.find_element_by_accessibility_id
and find_elements_by_accessibility_id
.
el = self.driver.find_element_by_accessibility_id('Animation')
self.assertIsNotNone(el)
els = self.driver.find_elements_by_accessibility_id('Animation')
self.assertIsInstance(els, list)
In order to accomodate mobile touch actions, and touch actions involving multiple pointers, the Selenium 3.0 draft specifies "touch gestures" and "multi actions", which build upon the touch actions.
move_to: note that use keyword arguments if no element
The API is built around TouchAction
objects, which are chains of one or more actions to be performed in a sequence. The actions are:
The perform
method sends the chain to the server in order to be enacted. It also empties the action chain, so the object can be reused. It will be at the end of all single action chains, but is unused when writing multi-action chains.
The tap
method stands alone, being unable to be chained with other methods. If you need a tap
-like action that starts a longer chain, use press
.
It can take either an element with an optional x-y offset, or absolute x-y coordinates for the tap, and an optional count.
el = self.driver.find_element_by_accessibility_id('Animation')
action = TouchAction(self.driver)
action.tap(el).perform()
el = self.driver.find_element_by_accessibility_id('Bouncing Balls')
self.assertIsNotNone(el)
In addition to chains of actions performed with in a single gesture, it is also possible to perform multiple chains at the same time, to simulate multi-finger actions. This is done through building a MultiAction
object that comprises a number of individual TouchAction
objects, one for each "finger".
Given two lists next to each other, we can scroll them independently but simultaneously:
els = self.driver.find_elements_by_class_name('listView')
a1 = TouchAction()
a1.press(els[0]) \
.move_to(x=10, y=0).move_to(x=10, y=-75).move_to(x=10, y=-600).release()
a2 = TouchAction()
a2.press(els[1]) \
.move_to(x=10, y=10).move_to(x=10, y=-300).move_to(x=10, y=-600).release()
ma = MultiAction(self.driver, els[0])
ma.add(a1, a2)
ma.perform();
There are a small number of operations that mobile testers need to do quite a bit that can be relatively complicated to build using the Touch and Multi-touch Action API. For these we provide some convenience methods in the Appium client.
This method, on the WebDriver object, allows for tapping with multiple fingers, simply by passing in an array of x-y coordinates to tap.
el = self.driver.find_element_by_name('Touch Paint')
action.tap(el).perform()
# set up array of two coordinates
positions = []
positions.append((100, 200))
positions.append((100, 400))
self.driver.tap(positions)
Swipe from one point to another point.
Zoom in on an element, doing a pinch out operation.
Zoom out on an element, doing a pinch in operation.
There are times when you want, in your tests, to manage the running application, such as installing or removing an application, etc.
The method driver.background_app
sends the running application to the background
for the specified amount of time, in seconds. After that time, the application is
brought back to the foreground.
driver.background_app(1)
sleep(2)
el = driver.find_element_by_name('Animation')
assertIsNotNone(el)
To check if an application is currently installed on the device, use the device.is_app_installed
method. This method takes the bundle id of the application and return True
or
False
.
assertFalse(self.driver.is_app_installed('sdfsdf'))
assertTrue(self.driver.is_app_installed('com.example.android.apis'))
To install an uninstalled application on the device, use device.install_app
,
sending in the path to the application file or archive.
assertFalse(driver.is_app_installed('io.selendroid.testapp'))
driver.install_app('/Users/isaac/code/python-client/test/apps/selendroid-test-app.apk')
assertTrue(driver.is_app_installed('io.selendroid.testapp'))
If you need to remove an application from the device, use device.remove_app
,
passing in the application id.
assertTrue(driver.is_app_installed('com.example.android.apis'))
driver.remove_app('com.example.android.apis')
assertFalse(driver.is_app_installed('com.example.android.apis'))
To launch the application specified in the desired capabilities, call driver.launch_app
.
Closing that application is initiated by driver.close_app
el = driver.find_element_by_name('Animation')
assertIsNotNone(el)
driver.close_app();
try:
driver.find_element_by_name('Animation')
except Exception as e:
pass # should not exist
driver.launch_app()
el = driver.find_element_by_name('Animation')
assertIsNotNone(el)
To reset the running application, use driver.reset
.
el = driver.find_element_by_name('App')
el.click()
driver.reset()
sleep(5)
el = driver.find_element_by_name('App')
assertIsNotNone(el)
The driver.start_activity
method opens arbitrary activities on a device.
If the activity is not part of the application under test, it will also
launch the activity's application.
driver.start_activity('com.foo.app', '.MyActivity')
The property method driver.app_strings
returns the application strings from
the application on the device.
strings = driver.app_strings
The driver.keyevent
method sends a keycode to the device. The keycodes can be
found here.
Android only.
# sending 'Home' key event
driver.press_keycode(3)
To hide the keyboard from view in iOS, use driver.hide_keyboard
. If a key name
is sent, the keyboard key with that name will be pressed. If no arguments are
passed in, the keyboard will be hidden by tapping on the screen outside the text
field, thus removing focus from it.
# get focus on text field, so keyboard comes up
el = driver.find_element_by_class_name('android.widget.TextView')
el.set_value('Testing')
el = driver.find_element_by_class_name('keyboard')
assertTrue(el.is_displayed())
driver.hide_keyboard('Done')
assertFalse(el.is_displayed())
# get focus on text field, so keyboard comes up
el = driver.find_element_by_class_name('android.widget.TextView')
el.set_value('Testing')
el = driver.find_element_by__name('keyboard')
assertTrue(el.is_displayed())
driver.hide_keyboard()
assertFalse(el.is_displayed())
The property method driver.current_activity
returns the name of the current
activity running on the device.
activity = driver.current_activity
assertEquals('.ApiDemos', activity)
Sometimes one needs to directly set the value of an element on the device. To do
this, the method driver.set_value
or element.set_value
is invoked.
el = driver.find_element_by_class_name('android.widget.EditText')
driver.set_value(el, 'Testing')
text = el.get_attribute('text')
assertEqual('Testing', text)
el.set_value('More testing')
text = el.get_attribute('text')
assertEqual('More testing', text)
To retrieve the contents of a file from the device, use driver.pull_file
, which
returns the contents of the specified file encoded in Base64.
# pulling the strings file for our application
data = driver.pull_file('data/local/tmp/strings.json')
strings = json.loads(data.decode('base64', 'strict'))
assertEqual('You can\'t wipe my data, you are a monkey!', strings[u'monkey_wipe_data'])
To put a file onto the device at a particular place, use the driver.push_file
method, which takes the path and the data, encoded as Base64, to be written to the file.
path = 'data/local/tmp/test_push_file.txt'
data = 'This is the contents of the file to push to the device.'
driver.push_file(path, data.encode('base64'))
data_ret = driver.pull_file('data/local/tmp/test_push_file.txt').decode('base64')
self.assertEqual(data, data_ret)
There is functionality in the Android emulator to instrument certain activities.
For information on this, see the Appium docs. To end this coverage
and retrieve the data, use driver.end_test_coverage
, passing in the intent
that is being instrumentalized, and the path to the coverage.ec
file on the
device.
coverage_ec_file = driver.end_test_coverage(intent='android.intent.action.MAIN', path='')
To lock the device for a certain amount of time, on iOS, use driver.lock
. The
argument is the number of seconds to wait before unlocking.
To shake the device, use driver.shake
.
Settings are a new concept introduced by appium. They are currently not a part of the Mobile JSON Wire Protocol, or the Webdriver spec.
Settings are a way to specify the behavior of the appium server.
Settings are:
Mutable, they can be changed during a session Only relevant during the session they are applied. They are reset for each new session. Control the way the appium server behaves during test automation. They do not apply to controlling the app or device under test.
See the docs for more information.
To get settings:
settings = driver.get_settings()
To set settings:
driver.update_settings({"some setting": "the value"})