W3C home > Mailing lists > Public > public-svg-wg@w3.org > October to December 2010

Re: Fw: Re: text-intro-05-t

From: Alex Danilo <alex@abbra.com>
Date: Thu, 09 Dec 2010 09:13:10 +1100
Message-Id: <Y1R4DL.0PB5P6HEZ91@abbra.com>
To: Cameron McCormack <cam@mcc.id.au>
Cc: public-svg-wg@w3.org
Hi All,

	One comment inline that should be discussed. (BTW, we pass that
test so I don't believe it's in error).

--Original Message--:
>[Edited and forwarded with permission.]
>
>----- Forwarded message from Piers Wombwell <piers@ekioh.com> -----
>
>From: Piers Wombwell <piers@ekioh.com>
>Date: Wed, 8 Dec 2010 13:43:11 +0000
>To: cam@mcc.id.au
>Subject: Re: text-intro-05-t
>
>Hi Cam,
>
>I think some of the SVG Tiny (and possibly Full) tests are incorrect, including this one.
>
>The SVG Tiny spec and SVG 1.1 Full specs disagree on the handling, though SVG 1.1Full2 now agrees with SVG Tiny.
>
>The key differences between the 1.1F1 spec is highlighted below.
>
>SVG1.1F1:
>For the ‘direction’ property to have any effect, the ‘unicode-bidi’ property's value must be embed or bidi-override.
>
>SVG1.2T (and SVG1.1F2):
>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 in SVG 1.2 Tiny), the 'unicode-bidi' property's value must be embed or bidi-override.
>
>unicode-bidi's default value is 'normal'.
>
>The 'direction' property's default value is 'ltr', so the text laid out with the assumption that the primary reading direction will be left-to-right. However, the text happens to be Arabic, so appears reversed. The use-case for this is when the majority of your text is LtR, but it includes some inline RtL.

"The assumption that the primary reading direction will be left-to-right" is a fair interpretation
of the text in the spec. "However, the text happens to be Arabic" is also notable.

As the text is Arabic, it consists of purely right to left code points. It would be a mistake
to ever assume that there is an implied left-to-right reading direction. The Arabic is
not embedded in anything, and so the user agent should treat this as RTL text and
the test in Tiny 1.2 reflects this. If the locale was somewhere in an Arabian speaking
country, then it's pretty obvious that RTL would be the assumed direction if any.

Same for Hebrew speaking nations.

If anything the text in the spec is misleading/wrong and should be clarified.

One should never assume a writing direction in content that contains no Unicode
characters corresponding to that direction. Explicit overrides are another matter entirely.

Alex

>In summary, I believe text-intro-05-t would be valid under SVG1.1Full1, but invalid under SVG1.1Full2 and SVGTiny.
>
>Another way of looking at it is all the examples in the SVGT spec, plus the W3C SVG Tiny RtL tutorial:
>	http://www.w3.org/International/tutorials/svg-tiny-bidi/
>
>They seemingly contradict the SVG T test suite's understanding of the text-anchor property.
>
>Piers.
>
>
>
>----- End forwarded message -----
>
>-- 
>Cameron McCormack ≝ http://mcc.id.au/
>
>
Received on Wednesday, 8 December 2010 22:14:01 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 8 December 2010 22:14:04 GMT