Re: [SVGMobile12] Lack of BIDI 'direction' (ISSUE-2058)

On Wed, 29 Oct 2008 17:48:24 +0100, fantasai <fantasai.lists@inkedblade.net> wrote:

> Erik Dahlström wrote:
>> On Wed, 29 Oct 2008 00:21:35 +0100, fantasai <fantasai.lists@inkedblade.net> wrote:
>>> Doug Schepers wrote:
>>>> fantasai wrote (on 10/28/08 7:10 PM):
>>>>> Doug Schepers wrote:
>>>>>> http://dev.w3.org/SVG/profiles/1.2T/publish/text.html#DirectionProperty
>>>>>   # For the 'direction' property to have any effect, the 'unicode-bidi'
>>>>>   # property's value must be embed or bidi-override.
>>>>>
>>>>> This is false. As I explained before, the 'direction' property alone has an
>>>>> effect when set at the paragraph level ("paragraph" being the unit of text
>>>>> the bidi algorithm operates on).
>>
>> I'm guessing this wording was the SVG translation of the following sentence in CSS:
>> "For the 'direction' property to have any effect on inline-level elements, the
>> 'unicode-bidi' property's value must be 'embed' or 'override'."
>>
>> So the question then becomes: what is an "inline-level element" in terms of svg?
>>
>> I'm thinking that this might be the 'tspan' element, since that cannot start a
>> text content block by itself. The 'tspan' element always needs to be enclosed
>> in a 'text content block element'.
>>
>>> I suggest removing the text. The first quoted sentence is very clearly wrong.
>>
>> Is the corresponding sentence in CSS also wrong?
>
> The corresponding sentence in CSS is qualified as describing only
> inline elements, which are effectively invisible to the bidi
> algorithm unless 'unicode-bidi' is set.
>
> The same might be true of tspan elements *if* they are *never*
> responsible for bounding the bidi algorithm's paragraph (i.e. never
> form a "text chunk" in SVG terms). I don't know enough about SVG's
> text model to say if that is true. But 'direction' also applies to
> <text> elements (or should) so even if tspan elements are the
> equivalent of CSS's inline elements, then you'd need to qualify the
> statement to describe only them.

Thanks for the clarification. 

Yes, in SVG 1.2 Tiny a 'tspan' element can never establish a "text chunk" since 'tspan' lacks the 'x' and 'y' attributes that give absolute position adjustments for text (note that these attributes are available in SVG 1.1 Full).

Would the following replacement text satisfy your comment:

"For the 'direction' property to have any effect on an element that does not by itself establish a new 'text chunk' (such as the 'tspan' element), the 'unicode-bidi' property's value must be 'embed' or 'bidi-override'."

Cheers
/Erik

-- 
Erik Dahlstrom, Core Technology Developer, Opera Software
Co-Chair, W3C SVG Working Group
Personal blog: http://my.opera.com/macdev_ed

Received on Thursday, 30 October 2008 09:29:48 UTC