Open jrandolf opened 1 year ago
Why would this be a capability? People are going to be confused when they try drag and drop and it doesn't work natively.
It looks like we would not need anything like this if implementations pass WPT tests for drag and drop: https://wpt.fyi/results/webdriver/tests/bidi/input/perform_actions/pointer_mouse.py?label=master&label=experimental&aligned&q=webdriver Given that this item got a high priority in the roadmap, let's discuss in the next WG meeting if it makes sense to include drag and drop as an explicit milestone into the roadmap.
My expectation is that if this becomes a capability most, if not all, clients would likely have this set to true for all sessions as it will be pretty poor DX to have to figure out why the commands fail to do what we want. WebDriver classic already passes drag and drop if it is a JS driven drag/drop. I raised a issue a few years ago to get the HTML version with draggable
sorted.
And to add using a capability would mean the setting is sticky and a client would have to create a new session to actually flip the value. I agree that this is not a solution that we should favor.
Overall this should have been discussed first (and now I see the needs-discussion label added to the actual referenced issue).
Fixed: https://github.com/w3c/webdriver-bidi/issues/487
Preview | Diff