The E2E-tests can be found in the e2e-tests subdirectory. To run the E2e-tests take a look at the commands in the package.json.
WebdriverIO is the framework surrounding the E2E-tests. It provides services to instantiate a local selenium server, connect to browserstack and to select the correct chromedriver. Additionally, it can be configured to run multiple E2E-tests in parallel.
Browserstack provides you with a large set of available operating systems, browser, and browser versions.
To see and run the E2E tests on Browserstack you need a Browserstack account, with access to our organization.
You will need to set E2E_BROWSERSTACK_KEY and E2E_BROWSERSTACK_USER as environment variables, which you can find
when logging in to Browserstack and selecting either the Automate or AppAutomate tab.
If you don't have this access you can still execute the tests locally with yarn test:web or yarn test:native.
For native browserstack tests you also have to set the environment variable E2E_CONFIG to either android or ios.
You also have to set the E2E_BROWSERSTACK_APP variable, which corresponds to the id of a *.apk or *.ipa, which is
already uploaded to browserstack.
To get such id, you either upload an *.apk or *.ipa to browserstack or use an already existing one (Checkout other
jobs. Using the same *.apk will not influence the other test). The id looks something like
this: bs://8ad23ac842b0ed12dafff0461bc3a0b98484234f.
The E2E-tests project implements everything using TypeScript. The E2E test folder contains two important subdirectories:
When writing new tests create helper classes for each Page you are using in the tests to keep the tests uniform and
readable. This is also called the page object pattern.
The tests itself are then created in the test folder and have to end on *.e2e.ts.
Selenium standalone requires Java 8.0+. To run the web E2E-tests you first have to start the web app, and then execute the e2e tests by running these two commands:
yarn workspace e2e prepare:web:start
yarn workspace e2e test:web
Note: Running tests local on Mac you have to enable Safari web-driver support. Run the following command in your terminal:
/usr/bin/safaridriver --enable
This will launch an automated Chrome session where you can follow the test execution. If you don't have Chrome
installed, you can append --firefox or any other browser of your choice to the command. Chromium does not seem to work
currently.
To run a single spec file you can run
yarn workspace e2e test:web --spec regexp
Check the documents here.
Selectors are used to select an element in the app programmatically.
However, during native development most of the common selectors are not available.
Therefore, you should use the accessibility identifier for this.
Add the accessibility-id to a React component using testID('Example-Component'). You can query this component
with $('~Example-Component') in your test.
For more complex queries you should add/use a custom Selector using predicate strings for iOS and UiSelectors for Android.
https://stackoverflow.com/a/54995267
The Appium method hideKeyboard() is known to be unstable when used on iPhone devices, as listed in Appium’s currently known open issues. Using this method for an iOS device may cause the Appium script to hang. Appium identifies that the problem is because "There is no automation hook for hiding the keyboard. Rather than using this method just think about how a user would hide the keyboard in your app, and tell Appium to do that instead.