Marionette currently captures the bounding box of the top-level browsing context's document when no msg.json.id argument is given. According to the WebDriver specification the expected behaviour is to capture the whole browser window, including the chrome, when no reference to a web element is given.
Looking at the spec I can't see this, I do see swathes of prose that need correcting
As dburns points out I was making things up. I've filed https://www.w3.org/Bugs/Public/show_bug.cgi?id=24364 about defining which area we should capture when a reference a web element isn't given, however.
Status: NEW → RESOLVED
Last Resolved: 5 years ago
Resolution: --- → INVALID
You need to log in before you can comment on or make changes to this bug.