Boris Zbarsky wrote:
Anne van Kesteren wrote:
I suppose. Though it would create some path in the XMLHttpRequest
specification we can't really test and it's not clear to me why we
should do that.
Because if it's low-cost, and lets people reuse the specification, it's
a win. Whoever comes up with another embedding of XHR would write the
tests for it, of course. Otherwise their specification wouldn't have
that whole test suite thing needed to exit CR.
Also, I forgot to mention that the definition of origin in HTML5
depends on Window and we really want to use the HTML5 definition of
origin.
Again, it seems like this could be handled by requiring that anyone
using XHR not in a Window specify how one gets an origin from the object
that the XHR constructor is on (or from the current scope object,
whichever XHR is using).
Or would that not work for some reason?
I fully agree with Boris. Also, remember that we'll have to remove any
normative references to the Window spec by the time we go to rec one way
or another no matter what, since the Window spec is unlikely to be in
Rec before XHR.
/ Jonas
