RE: [css3-text] Does text-transform:fullwidth transform U+0020 SPACE characters as well?

Apart from U+0020, but I found fantasai's point important.

Alphabet letters like 'a' U+0061 has different characteristics than full-width 'a' U+FF41 in terms of line-wrapping, justification, word-spacing, text-autospace, etc.

We should probably state that:
* U+0020 is not transformed even though Unicode says it should
* Transforming to full-width must happen before line-wrapping
* Line-wrapping, justification, word-spacing, text-autospace, etc. follow the characteristics of transformed code points (this may cover the 2nd item)

I hope this won't impact existing implementations. Although ''uppercase'' and other values do not affect characteristics, the widths of glyphs are affected, so the wild guess is that existing implementations are transforming before line-wrapping.


Regards,
Koji

-----Original Message-----
From: public-i18n-core-request@w3.org [mailto:public-i18n-core-request@w3.org] On Behalf Of Koji Ishii
Sent: Thursday, February 17, 2011 11:34 AM
To: fantasai; Kang-Hao (Kenny) Lu
Cc: ML public-i18n-core; CJK discussion; WWW Style
Subject: RE: [css3-text] Does text-transform:fullwidth transform U+0020 SPACE characters as well?

Thank you for pointing out the potential risks, Kenny and fantasai.

The use case exists in the original mail[1] but it is not strong, and it looks like the discussion in Japan seem to be "either", so I agree with you, let's NOT take the potential risk.

I've posted this to Japanese ML. As long as we don't see strong objections in a few days, I agree to modify the spec to exclude U+0020 from the transform.


Regards,
Koji

[1] http://lists.w3.org/Archives/Public/www-style/2011Feb/0467.html


-----Original Message-----
From: www-style-request@w3.org [mailto:www-style-request@w3.org] On Behalf Of fantasai
Sent: Thursday, February 17, 2011 6:35 AM
To: Kang-Hao (Kenny) Lu
Cc: ML public-i18n-core; CJK discussion; WWW Style
Subject: Re: [css3-text] Does text-transform:fullwidth transform U+0020 SPACE characters as well?

On 02/16/2011 12:03 PM, Kang-Hao (Kenny) Lu wrote:
> (11/02/17 4:33), fantasai wrote:
>> [snip]
>> I am not sure if text-transform: fullwidth should affect U+0020. It 
>> makes sense to me either way.
>
> I tried some examples and found it weird to have U+0020 in between two 
> fullwidth Latin characters. I'll be interested in seeing a real world 
> example and its use case. Perhaps I am missing something.

It is potentially a very significant change to implementations to have U+0020 change to U+3000 because this will require that text-transform happens after white space collapsing. Right now implementations can do either: they can transform before or after white space collapsing. So I think we should make sure that this is needed in real-world cases before we make this change. Note that U+3000 has very different behavior from U+0020 -- it is not just a visual change. It has different line-wrapping behavior, it has different justification behavior, and it is not affected by the 'word-spacing' property.

~fantasai

Received on Thursday, 17 February 2011 04:54:59 UTC