On Sat, 14 May 2011 00:17:55 +0200, Jonas Sicking <jo...@sicking.cc> wrote:
Yeah, I think we can get away with this.
The one case where it seems that Gecko and the spec differentiate for
the readystatechange event is that gecko dispatches readystatechange
when going to the DONE state even if the network request failed during
a sync-load. The spec only dispatches it for successful loads.
In other words, Gecko always dispatches readystatechange (with
.readyState set to 4) before send() returns.
Hmm, send() either returns and dispatches the event, or it throws an
exception, for synchronous requests.
--
Anne van Kesteren
http://annevankesteren.nl/
