Re: [1.2T-LC] text-align incompatible with CSS (ISSUE-2057 ACTION-2200)

Doug Schepers wrote:
> Hi, fantasai-
> 
> The SVG WG agrees with this, and made the following changes, which we
> hope will satisfy you comment.
> 
>> fantasai wrote (on 9/17/08 1:41 PM):
>>> http://www.w3.org/TR/SVGMobile12/text.html#TextAlignProperty
>>>
>>> There are two problems with this definition. First
>>>
>>>   # Note that SVG does not require user-agents to support the
>>>   # following values for this property: "justify", "inside",
>>>   # "outside", <string>, "left", or "right". When an unsupported
>>>   # value is encountered it must be treated as the property
>>>   # had not been specified.
>>>
>>> It should be normatively allowed for the UA to support the
>>> other values of text-align. If the UA also supports CSS then
>>> it's not possible for the values not listed here to be ignored.
> 
> We clarified this to say:
> [[
> When a value not supported by the user agent is encountered, it must be
> treated as the property had not been specified.
> ]]
> 
> This way, UAs that do support CSS may use those values, while others may
> claim conformance to SVG 1.2 Tiny (which doesn't mandate CSS support).

Which section is that sentence in? Also you want to be more careful
with the wording: you don't want something like

   * { text-align: end;
       text-align: right; }

to result in the lacuna value.

>>> Second problem is here:
>>>
>>>   # The values "start" and "end" are dependent on the writing
>>>   # system being used.
>>>   #
>>>   #   * For left to right horizontal (English, French, etc):
>>>   #     start=left and end=right
>>>   #   * For right to left horizontal (Hebrew, Arabic, etc):
>>>   #     start=right and end=left
>>>   #   * For top to bottom vertical (vertical Chinese, etc):
>>>   #     start=up and end=down
>>>
>>> "writing system being used" is not a good basis for alignment.
>>> What about mixed script text? text-align should be dependent
>>> on SVG's equivalent of the containing block's 'direction'
>>> property. (Which afaict doesn't exist in SVG Tiny, but more on
>>> that problem later.)
>>>
>>> Also given "In SVG Tiny 1.2, vertical writing is not supported."
>>> in http://www.w3.org/TR/SVGMobile12/text.html#TextLayout that
>>> last sentence should be removed.
> 
> Because we do not forbid SVG 1.2 Tiny UAs to support vertical text, and
> because we want to keep consistency with our other specs, we decided not
> to remove the last bullet point.
> 
> But we do agree that it could be confusing for readers (though it's
> mentioned in the prefatory material), so we added an explanatory note to
> the bullet point:
> [[
> * For top to bottom vertical (vertical Chinese, etc): start=up and
> end=down (Note: SVG Tiny 1.2 does not mandate the support of vertical text.)
> ]]

s/mandate/include/ and I'm happy with that.

~fantasai

Received on Thursday, 23 October 2008 12:17:25 UTC