- From: Ian Hickson <ian@hixie.ch>
- Date: Thu, 26 Aug 2004 01:14:49 +0000 (UTC)
On Fri, 13 Aug 2004, Olav Junker [iso-8859-1] Kj?r wrote:
>
> I'm concerned that authors would write things like:
>
> if (inputype.validity==form.ERROR_TOO_LONG) {
> // nag
> } else if (inputype.validity==form.ERROR_PATTERN_MISMATCH) {
> // nag
> }
True.
> I'm not sure what the solution is, maybe an error-object with boolean
> fields for each possible error, so you could write:
>
> if (inputype.validity.isTooLong) {
> // nag
> } else if (inputype.validity.isPatternMismatch) {
> // nag
> }
>
> This seems a bit more scripter-friendly to me.
Good idea.
I've defined it so that |validity| is still the bitfield that it was
before, but now also supports .isFoo property access.
--
Ian Hickson U+1047E )\._.,--....,'``. fL
http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,.
Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Wednesday, 25 August 2004 18:14:49 UTC