> Hm, I guess for <br ubi> it would mean it is treated as U+FFFC.
> This probably makes sense, if ubi were allowed on that element.
I do not want it to work that way. We can disallow ubi (and dir) on <br>,
but I would be happier if it were part of the definition of
unicode-bidi:isolate.
Aharon
On Tue, Sep 28, 2010 at 7:08 AM, fantasai <fantasai.lists@inkedblade.net>wrote:
> On 09/27/2010 04:14 AM, Aharon (Vladimir) Lanin wrote:
>
>> The spec says "uninterrupted by a forced (bidi class B) line break or
>> block boundary".
>>
>> According to the proposal (sections 3.1 and 3.3), <br> (by default) and
>> <div>...</div> form UBA paragraph breaks, i.e. those bidi class B line
>> breaks and block boundaries.
>>
>> But unicode-bidi:isolate says that when it's applied to an element, it
>> acts as U+FFFC, an ON in its surrounding paragraph. So, is <br ubi> or a
>> <div ubi /> a B or an ON? It is quite easy to think that it is in fact
>> an ON. However, during the f2f, we explicitly said ubi does not have any
>> effect on non-inline elements.
>>
>
> Hm, I guess for <br ubi> it would mean it is treated as U+FFFC.
> This probably makes sense, if ubi were allowed on that element.
>
> But <div ubi/> is a block element; it creates block boundaries
> before and after it, and so it is treated as a hard bidi paragraph
> break.
>
> ~fantasai
>