W3C home > Mailing lists > Public > www-style@w3.org > November 2003

Re: [CSS2.1] Text 16.6.2 Bidirectionality with Whitespace Collapsing

From: fantasai <fantasai@escape.com>
Date: Wed, 05 Nov 2003 17:58:49 -0500
Message-ID: <3FA980A9.1030000@escape.com>
Cc: www-style@w3.org, W3c I18n Group <w3c-i18n-ig@w3.org>

Tex Texin wrote:

> fantasai wrote:
> 
>>Tex Texin wrote:
>> >
>> > http://www.w3.org/TR/2003/WD-CSS21-20030915/text.html#q9
>>...
>>That sentence is there so that spaces separated by bidi formatting
>>characters will collapse; whether a character follows another character
>>is independent of reordering. (Think: which side comes first -- left or
>>right? The order we're talking about here has to be the byte order.)
> 
> Yes, the sentence says spaces separated by bidi formatting characters will
> collapse.
> However, supposing instead of the Unicode bidi formatting characters e.g. RLE,
> supposing we use
> <span dir="rtl">. The effect should be the same right?

Yes. Spaces collapse across element boundaries, as explained in the collapsing
rule.

> Yes, but the spaces are going to have to be somewhere, they are not always in
> the middle of elements.
> The GEO FAQ concluded it was better to have them in the beginning.

With the exception of some restrictive content models, you can almost always
get away with putting them *outside* the elements -- which works better when
you format them anyway. Imagine an underline or a border on <em>:

    some <em> more </em> text

It's better to have

    some <em>more</em> text

~fantasai
Received on Wednesday, 5 November 2003 17:58:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:25 GMT