- From: Anne van Kesteren <annevk@opera.com>
- Date: Mon, 20 Mar 2006 18:52:00 +0100
- To: "Web APIs WG (public)" <public-webapi@w3.org>
It's probably relevant to read the thread on www-dom http://lists.w3.org/Archives/Public/www-dom/2005JanMar/thread.html#3 started by David Baron. The easiest way to define keypress would be something along the lines of: The keypress event MUST be dispatched as default action of the keydown event and when the key is hold down it MUST be subsequently dispatched according to platform conventions. If the keypress event leads to text input the textInput event MUST be dispatched accordingly. This means that even when using an IME keypress is dispatched which application authors should be aware of. (That is to say, use it for games, not to catch text input.) Other related issues seem to be what actually happens when you hold down a key and two 'a' characters are inserted. Possible insane scenario's include (leaving keypress aside for the moment): keydown, textInput, keyup, keydown, textInput, keyup ... as per http://www.w3.org/mid/1106593442.11640.51.camel@localhost which does not seem to be supported by the current specification according to Björn. Others: keydown, textInput, textInput, keyup keydown, textInput, keydown, textInput, keyup keydown, textInput, keydown, textInput, keyup, keyup ... where the latter two can probably be excluded. I guess there is need for follow-up issues, testcases and actions... Mostly testcases I guess, if there's one thing the world needs. -- Anne van Kesteren <http://annevankesteren.nl/> <http://www.opera.com/>
Received on Monday, 20 March 2006 17:52:15 UTC