W3C home > Mailing lists > Public > www-style@w3.org > October 2001

Re: Escapes in CSS identifiers are equivalent?

From: Bjoern Hoehrmann <derhoermi@gmx.net>
Date: Mon, 08 Oct 2001 15:28:38 +0200
To: Etan Wexler <ewexler@stickdog.com>
Cc: Web style list <www-style@w3.org>, "David Baron" <dbaron@fas.harvard.edu>
Message-ID: <0b93stcm1vp769epio9oo02jifv32gj8dt@4ax.com>
* Etan Wexler wrote:
>Are escaped characters in CSS reserved identifiers equivalent to the 
>literal forms?  Or, rather, are the identifiers then equivalent?  
>Observe:
>
>@impo\rt \url("style") \screen;

Interpreting CSS Level 2, this one is invalid, the import keyword is a
literal string in Appendix D.2, same for the the url function as in
section 4.1.1 and as per section 4.1.3 identifiers may *contain*
escapes, but the prose doesn't state that nmstart tokens may contain
escapes, section 4.1.1 conflicts with this, one of those sections is in
error.

>If the escaped forms are equivalent, the grammars of CSS1, CSS2, and 
>CSS3 need an overhaul.  If the escaped forms are not equivalent, the 
>CSS object model needs a way to differentiate between escapes and 
>literal characters.

Hm, it has, if they are not equivalent, an \url(...) would be a
CSS_UNKNOWN primitive value while url(...) would be a CSS_URI value...
However, it's impossible to make this distinction, if

  background\-color: green

isn't canonically equivalent to

  background-color: green

what is it then? The question rather is, do escapes apply to predefined
identifiers, i.e. is the escaped property valid at all? This one of
those cases where I said the CSS Level 2 grammar is way too lose.

regards,
-- 
Björn Höhrmann { mailto:bjoern@hoehrmann.de } http://www.bjoernsworld.de
am Badedeich 7 } Telefon: +49(0)4667/981028 { http://bjoern.hoehrmann.de
25899 Dagebüll { PGP Pub. KeyID: 0xA4357E78 } http://www.learn.to/quote/
Received on Monday, 8 October 2001 09:29:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:11 GMT