Re: Drop foster-parented stuff on the floor if parent gone

On Mon, Feb 23, 2009 at 2:04 PM, Ian Hickson <ian@hixie.ch> wrote:
> On Mon, 23 Feb 2009, Henri Sivonen wrote:
>>
>> On Feb 23, 2009, at 15:39, Henri Sivonen wrote:
>>
>> > It breaks compat with Firefox and Safari and it isn't consistent with IE8 or
>> > Opera, either.
>>
>> I suggest reverting r2732, since it didn't actually make anything easier
>> for the multi-threaded case (the same pointers need to travel across the
>> thread boundary) and wasn't that bad to begin with.
>
> Done.

While I can live with any of the proposed solutions here (as long as
they don't break the web), I do still think that the current algorithm
is needlessly complicated. Seems simpler to just drop any
foster-parented content on the floor instead of inserting it where it
was parsed.

The argument against this that has been brought up so far was that
this is more complex. However Henris email on Jan 2nd in this thread
indicate that it would actually simplify the algorithm to drop the
content rather than what the spec says now.

Sorry for mentioning the more complex algorithm, I didn't really mean
to suggest for the spec to use it. I think this case is rare enough
that it's unlikely that anyone cares what we do, so we should just do
whatever is simplest.

/ Jonas

Received on Tuesday, 24 February 2009 00:31:59 UTC