W3C home > Mailing lists > Public > www-style@w3.org > February 2005

[CSS3] Staged scanning

From: Christian Roth <roth@visualclick.de>
Date: Wed, 2 Feb 2005 13:44:16 +0100
To: "www-style Mailing List" <www-style@w3.org>
Message-Id: <20050202124416.1332@mail.visualclick.de>

Am I correct that scanning a CSS3 stylesheet requires staged scanning,
i.e. at least two passes of tokenization with different tokenization
rules each?

The way I currently figure it needs to be done:

1. Tokenize stylesheet according to CSS Core syntax:
<http://www.w3.org/TR/2003/WD-css3-syntax-20030813/#tokenization>

2. Determine where the selector part is:
<http://www.w3.org/TR/2003/WD-css3-syntax-20030813/#style>

3. Recombine all tokens making up that selector part again into a
character stream.

4. Re-scan the selector part, but now with the tokenization rules of the
Selector scanner:
<http://www.w3.org/TR/2001/CR-css3-selectors-20011113/#lex>

Note that this tokenizer definition is quite different than the one from
the Core Syntax (SIGNED_INTEGER, INTEGER).


If this rather complex process is not the way how to parse CSS3, how is
the intended way to do it?

Regards, Christian.
Received on Wednesday, 2 February 2005 12:44:51 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:35 GMT