When you build your browser automation projects via the Webdriver protocol you need to take into consideration the following:
- Selectors for UI elements inside iframes are not validated. It is recommended to avoid iframes when you build your automation project.
- The Webdriver protocol always creates a new browser session. This means that saved cookies or saved browser data are not taken into consideration when you build your automation project.
- If a selector is not found for a browser with multiple tabs opened, the execution constantly switches between them. In this scenario it is recommended to create your automation project on a single browser tab.
- Headless automation does not rely on visual elements. As such, the Webdriver protocol doesn't actually open a visible browser window when it is used in Headless mode. Because of this, Headless automation cannot function with some activities which use hardware events (such as Click or Type Into) in their default state. They must be configured to use the SimulateClick, SimulateType, or SimulateHover properties.
- Activities which interact with images (such as Click Image or Find Image) are not supported because they rely on the browser window itself.
- Events monitoring (such as Click Trigger or Hotkey Trigger) is not supported, because they rely on hardware events.
<WND>tags in the Close Application activity only closes the browser session and not the corresponding webdriver process. It is recommended to use
- When using Chrome, the browser window opened via the Webdriver protocol always opens in the same visual space and with the same size specifications. Moreover, some websites have adaptive page layout and this can lead to invalid selectors when the window size is changed. In this regard, it is recommended to use the default browser window size when you create your automation project.
Updated about a month ago