- From: Sam Ruby <rubys@us.ibm.com>
- Date: Sat, 04 Aug 2007 12:20:11 -0400
- To: Simon Pieters <simonp@opera.com>
- CC: public-html@w3.org
Simon Pieters wrote:
> On Sat, 04 Aug 2007 17:11:15 +0200, Sam Ruby <rubys@us.ibm.com> wrote:
>
>>> Sorry about being thick, but I'm still not sure what changes you are
>>> suggesting to CDATA tokenization.
>>
>> Given this input:
>>
>> <p><![CDATA[x]]></p>
>>
>> HTML5 current interprets that exactly the same as:
>>
>> <p><!--[CDATA[x]]--></p>
>>
>> I'm proposing that given this input:
>>
>> <o:p><![CDATA[x]]></o:p>
>>
>> HTML5 would treat that as:
>>
>> <o:p>x/o:p>
>
> The use-case for this is to hide text inside embedded SVG from IE. [1]
That's one use case, yes.
> Using the above construct will also hide the text from Safari and
> Firefox (which I guess is the intension), but not Opera.
>
> Assuming that "<svg:svg>" ends up in the DOM with the localName "svg" in
> HTML5-aware UAs (with your proposal), it is possible to hide such text
> from legacy UAs (and IE) with:
>
> svg\:svg { display:none; }
>
> ...which avoids the need for special-casing "<![CDATA[". Furthermore, it
> will achieve the intended result in Opera, and you can use characters
> such as [, ] and > in the text. [2]
I did some more testing, and this DOES appear to hide the text on IE,
but only if xmlns:svg is declared (weird!), which might explain why I
did not notice this before.
Thanks!
> [1] http://www.w3.org/mid/46B35F07.6050603@us.ibm.com
> [2] http://www.w3.org/mid/46B22F81.8020708@zaynar.demon.co.uk
> --Simon Pieters
> Opera Software
- Sam Ruby
Received on Saturday, 4 August 2007 16:20:35 UTC