Hi Mirko,
We offer some basic DOM navigation functions within WebDriver as it
stands: just call "findElement" or "findElements" on an existing
WebElement to access nodes underneath it. We currently only model
actual element objects because this is what people interact with when
using the browser. One of the design goals of WebDriver is to follow
as closely as possible what a user would do when interacting with the
browser.
The interesting tension in developing a framework like WebDriver (and,
I assume Qooxdoo) is between what's technically possible, and which
features hit the sweet-spot between complexity (either of usage or
implemenation) and capability. "Access to the DOM' is already there,
to some extent, but I'd love to hear what feature is missing and how
it can help our users do their testing. Your thoughts and suggestions
would be greatly appreciated.
Regards,
Simon