getAttribute() always non-null

Germain Garand germain at ebooksfrance.org
Wed Nov 23 16:29:18 GMT 2005


Le Mercredi 23 Novembre 2005 14:03, Maks Orlovich a écrit :
> On Wednesday 23 November 2005 07:47, Dirk Mueller wrote:
> > On Tuesday 22 November 2005 17:47, David Faure wrote:
> > > I agree with both arguments, and I suggest that we return null too,
> > > unless anyone knows of plans from IE or Mozilla to change this behavior
> > > in the near future.
> >
> > We must not return null if we support the attribute on the element type.
>
> Hmm, IE does behave that way, indeed, but Mozilla doesn't. I don't think we
> have a notion of a supported attribute though, do we?
>

.. and that's just yet another IE implementation accident, so "must" really 
depends whether we ever find a critically important webpage that relies on 
that even more silly detail...

btw, I'm still not sure why we want to break at all a proposal that has been 
asserted in DOM 2 and is *re-asserted* in DOM 3 for this lone 
page/application, that is not so critical, and that is supposedly standards 
aware? are there any known others?
what about reporting a bug on Mr. Meyer's S5 tool instead?

Not that I'm particularly attached to that aspect of the standard, but it's 
always a slippery path to take... it's not like Opera backed off, after 
all...

Greetings,
Germain




More information about the kfm-devel mailing list