W3C home > Mailing lists > Public > www-style@w3.org > April 2013

Re: [CSSWG][css-overflow] First Public WD of CSS Overflow Module Level 3

From: Brad Kemper <brad.kemper@gmail.com>
Date: Sat, 27 Apr 2013 14:58:05 -0700
Message-Id: <419A8B95-B7E8-470F-AEC6-56CA0A1F8265@gmail.com>
Cc: Håkon Wium Lie <howcome@opera.com>, "L. David Baron" <dbaron@dbaron.org>, "www-style@w3.org" <www-style@w3.org>
To: Alan Gresley <alan@css-class.com>
On Apr 27, 2013, at 6:57 AM, Alan Gresley <alan@css-class.com> wrote:

> On 26/04/2013 5:13 AM, Brad Kemper wrote:
>> 
>> On Apr 25, 2013, at 11:37 AM, Håkon Wium Lie <howcome@opera.com>
>> wrote:
>> 
>>> Also sprach Brad Kemper:
>>> 
>>>> I mentioned it in another thread full of other points, but how
>>>> do you feel about making 'overflow-x' and 'overflow-y' be based
>>>> on writing mode, so that 'overflow-x' overflowed to the left for
>>>> vertical-rl writing?
>>> 
>>> +1
>>> 
>>> This matches Opera's intent, if not implementation (as Presto
>>> doesn't support vertical writing).
>>> 
>>>> If there are compatibility concerns, this behavior could be
>>>> switched on with a separate property, such as
>>>> 'overflow-behavior: logical | physical', or something like that.
>>> 
>>> I'd like to avoid such switches as they create more
>>> interdependencies. It's better to have names/values where the
>>> meaning is self-evident. For example, 'text-indent' depends on the
>>> writing direction, while 'margin-left' doesn't.
>> 
>> So, do you think compatibility with existing overflow-x and
>> overflow-y using vertical writing mode in the wild is not a big
>> concern? Or do you think there should be two new, better-named
>> properties, such as 'overflow-inline' and 'overflow-block'?
> 
> Brad. Can you give some examples?
> 
> I have previously mentioned about a logical block progression [1 & 2] and where does visible overflow go.

Not sure what examples you want. But basically, I think it is well established that if a box is set to contain vertical writing, the block progression direction in that box is horizontal. That is, new blocks in that box are added in a direction that is perpendicular to the writing direction, same as with horizontal writing.

So the main question is merely "can we make the "y" in 'overflow-y' equal to the logical Y of the writing mode (i.e., the block progression direction)", so that it is horizontally overflowed when the writing mode is vertical? And, of course, "can we make the "x" in 'overflow-x' equal to the logical X of the writing mode (i.e., the inline progression direction)", so that it is horizontally overflowed when the writing mode is vertical? And if not, because of existing content, what are the alternatives?

As I see it, one alternative is a property that switches from physical X and Y (horizontal and vertical, respectively) to logical X and Y (inline- and block-progression, respectively).

The other alternative is have two new properties that are similar to 'overflow-x' and 'overflow-y', which take their initial values from overflow-x' and 'overflow-y', but which use logical direction property names, like 'overflow-inline' and 'overflow-block'. I don't like that alternative as much.

As for the initial scroll positions, I would hope that they would always be at the start and before edges.

> 
> Håkon. This is a major issue with non page documents (or paging effects) in CJK. There would be plenty of hacks to render ths test case [3] in a logical manner where you arrived at the beginning (01) of the document and not it's end (30).
> 
> 
> Alan
> 
> 1. http://lists.w3.org/Archives/Public/www-style/2012Jun/0540.html
> 2. http://lists.w3.org/Archives/Public/www-style/2010Oct/0645.html
> 3. http://css-class.com/test/css/bidi/kanji-test2.htm
> 
> 
> 
> 
> -- 
> Alan Gresley
> http://css-3d.org/
> http://css-class.com/
Received on Saturday, 27 April 2013 21:58:35 UTC

This archive was generated by hypermail 2.3.1 : Saturday, 27 April 2013 21:58:35 UTC