- From: Ian Hickson <ian@hixie.ch>
- Date: Mon, 7 Feb 2005 23:20:49 +0000 (UTC)
On Mon, 7 Feb 2005, [ISO-8859-1] Olav Junker Kj?r wrote: > You write: > > The model used by the validity DOM attribute is not very clean, and > > may deserve to be replaced with something that has been thought > > through with more care > > Perhaps the bit field interface could be left out. Yeah, I'm considering it. My main reason not to just do it is that it would require _massive_ changes to the draft (everything is currently defined in terms of the bit masks, and the object is then defined in terms of that). > A better solution would perhaps be to add an attribute, "isValid" to > ValidationState. So you could write: > > if (field1.validity.isValid) ... > > Only problem is that the naming seems a bit strange. "validityState" > would be clearer than "validity", but also a bit more cumbersome. That's pretty much the line of thought I had on the matter. > Anyway, it seems a bit redundant to support both the bit field and > property interfaces. I agree. Hence the comment in the status section. :-) -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Monday, 7 February 2005 15:20:49 UTC