UIAutomator2
To benefit from UIAutomator2 test capabilities (like interacting with other apps or system elements), you must start the test server with:
start_test_server_in_background(with_uiautomator: true)
Then you will be able to use the command perform_action('method', {parameters}) which includes the following actions.
| Method | Parameter | Function |
|---|---|---|
| uiautomator_touch_text | Text | Tap on element with text |
| uiautomator_touch_partial_text | Text | Tap on element containing text |
| uiautomator_drag_coordinates | x origin, x destination, y origin, y destination, (optional) steps | Drag on the given coordinates |
| pull_notification | none | Open the notification shutter |
| clear_notifications | none | Clear notifications and close the shutter |
| uiautomator_text_dump | none | Dumps text from elements in the UI hierarchy |
| uiautomator_ui_dump | none | Dumps elements in the UI hierarchy |
| wifi | 'on' or 'off' | Switch device wifi on or off |
| wait_for_idle_sync | timeout in seconds | Wait for idle state |
And the more advanced command uiautomator_execute which takes 4 arguments.
perform_action('uiautomator_execute', 'strategy', 'locator', element index, 'action')
This will return true, and the result of the query if succeeds or an exception if the strategy, or action are invalid. If an element is not found, it will return false.
Where strategy and action can be:
| strategy | description |
|---|---|
| class | class type of ui element |
| res | element resource |
| desc | ui element description |
| descContains | ui element description contains |
| descEndsWith | ui element description ends with |
| descStartWith | ui element description starts with text |
| text | ui element with text |
| textContains | ui element contains text |
| textEndsWith | ui element ends with text |
| textStartWith | ui element starts with text |
| pkg | package name from application of ui element |
| action | description |
|---|---|
| click | tap on element |
| longClick | long tap on element |
| getText | get text from element |
| getContentDescription | get description |
| getClassName | get class name |
| getResourceName | get resource name |
| getVisibleBounds | get coords for visible bounds of element |
| getVisibleCenter | get coords for visible center of element |
| getApplicationPackage | get package name of application |
| getChildCount | number of children elements |
| clear | clear text from element |
| isCheckable | can the element be checked? |
| isChecked | is the element checked? |
| isClickable | is the element clickable? |
| isEnabled | is the element enabled? |
| isFocusable | can the element be focused? |
| isFocused | is the element focused? |
| isLongClickable | can the element be long clicked? |
| isScrollable | is the element scrollable? |
| isSelected | is the element selected? |
For entering text into the input text field can be performed with:
perform_action('uiautomator_set_text', 'strategy', 'locator', element index, 'input text')
input text is the text value which needs to be into the field.
Scrolling to the element can be performed by Horizontally and vertically with the following commands.
For vertical scrolling the command is:
perform_action('uiautomator_vertically_scroll_to_element', 'strategy', 'locator', 'container_strategy', 'container_locator', 'max_scrolls')
For horizontal scrolling the command is:
perform_action('uiautomator_horizontally_scroll_to_element', 'strategy', 'locator', 'container_strategy', 'container_locator', 'max_scrolls')
Where strategy and parentstrategy can be:
| strategy | description |
|---|---|
| class | class type of ui element |
| res | element resource |
| desc | ui element description |
| descContains | ui element description contains |
| descEndsWith | ui element description ends with |
| descStartWith | ui element description starts with text |
| text | ui element with text |
| textContains | ui element contains text |
| textEndsWith | ui element ends with text |
| textStartWith | ui element starts with text |
| pkg | package name from application of ui element |
max_scrolls is a integer number and It is maxiumum number of times the screen can be scrolled.
container_strategy and container_locator If screen layout contains multiple container and the scrolling to be peformed on a specific container then parent strategy and locator should be provided and for the single container the arugments can be ignored.
Additional to the above scrolling method, below Scrolling can be performed in forward or backward direction for both vertical and horizontal.
For vertical scrolling FORWARD/BACKWARD the command is:
perform_action('uiautomator_vertically_scroll_forward_or_backward_to_element', 'direction', 'strategy', 'locator', 'max_scrolls')
perform_action('uiautomator_vertically_scroll_forward_or_backward_to_element', 'FORWARD', 'strategy', 'locator', 'max_scrolls')
perform_action('uiautomator_vertically_scroll_forward_or_backward_to_element', 'BACKWARD', 'strategy', 'locator', 'max_scrolls')
For horizontal scrolling the command is:
perform_action('uiautomator_horizontally_scroll_forward_or_backward_to_element', 'FORWARD', 'strategy', 'locator', 'max_scrolls')
perform_action('uiautomator_horizontally_scroll_forward_or_backward_to_element', 'BACKWARD', 'strategy', 'locator', 'max_scrolls')