On 2/11/06, Anne van Kesteren <[EMAIL PROTECTED]> wrote: > Quoting Jim Ley <[EMAIL PROTECTED]>: > > D you mean the bug in Opera 9 that means changing the size of the > > select selects an entry? Surely that's just a bug in Opera 9 preview > > (but not 8.5), changing the size should have no effect on what's > > selected. > > I agree (with the part after the comma of the last sentence). Given that per > <http://whatwg.org/specs/web-forms/current-work/#select-check-default> the > option should be selected, it should remain selected when changing some factor > that has nothing to do with that.
but in your test case, they are not selected, and then become selected in Opera 9.0 when you give them an invalid attribute - was Opera 9.0 not the behaviour you were looking to standardise too? (I assumed that as the Opera 8.5 behaviour was absurd with -1 making it about 40 high) For me ignoring the invalid DOM change makes more sense that the inconsistent proposals you're advocating. The cited part of the spec linked to above certainly does not say what should happen when a DOM changes to a single select in any case, it is still undefined even in the size=1 case. Cheers, Jim.
