Re: Issue 101 Resolution

On Tue, Oct 12, 2010 at 10:39:53AM +1100, Alan Gresley wrote:
> Anton Prowse wrote:
> >On 11/10/2010 16:06, Alan Gresley wrote:

> >>Rule 3 doesn't specify which float is
> >>place lower and no where else in 9.5 does it specify that alternative
> >>floated boxes can be placed lower.
> >
> >Rule 5 covers this.
> 
> 
> No it doesn't. Rule 5 only mentions the outer top edge.
> 
>   | The outer top of a floating box may not be higher than the
>   | outer top of any block or floated box generated by an element
>   | earlier in the source document.
> 
> 
> There is nothing about the top edge of a float being lower than the
> bottom edge of a float that is earlier in the source. The only rule
> that mentions something like this is in rule 2 but that doesn't
> apply with interaction between left floats and right floats.

I believe Anton Prowse's point is that in effect it does "specify which float
is place[d] lower" (to satisfy rule 3), because only one of the two options for
a given pair would satisfy rule 5.

So in some sense it's not quite true that rule 3 is the only rule where
"something left[-floated] taking an interest in what is right[-floated]", in
that rule 5 gives at least vertical constraints between left and right floats.

However, I still don't understand why rule 3's exceptionalness is important.
I still don't see how the float rules are "all about LTR inline progression"
(though I agree that they're all about (i.e. they assume) horizontal inline
progression), and I don't see why "This has to be fined tuned since the mirror
computations must happen for RTL inline progression".

pjrm.

Received on Tuesday, 12 October 2010 02:10:09 UTC