Re: HTML 4.0/SGML, (#PCDATA)* problem in TEXTAREA

Neil St.Laurent (neil@bigpic.com)
Wed, 30 Jul 1997 09:04:46 -0600


Message-Id: <199707301500.JAA00415@underworld.bigpic.com>
From: "Neil St.Laurent" <neil@bigpic.com>
To: Paul Prescod <papresco@technologist.com>
Date: Wed, 30 Jul 1997 09:04:46 -0600
Subject: Re: HTML 4.0/SGML, (#PCDATA)* problem in TEXTAREA
CC: www-html@w3.org

In the time of 29 Jul 97:22:03, www-html@w3.org pronounced:

> wrong as documents that did not. I can't find anything that would
> recommend an error recovery mechanism in HTML 4.0, but I haven't read
> the whole spec. Like you, many people misread the earlier comment to be
> some form of rule..

Thanks for that clarification.  What we've decided to do for the 
purpose of using invalid documents is define a generic Pseudo-SGML 
rule:

<!ELEMENT name - O (#PCDATA)>
<!ATTLIST name
    (name  CDATA  #IMPLIED)* >

Where "name" is the corresponding SGML rule.

In our final representation of the document we'll simply ignore 
unrecognized elements [& their attributes] caught under this rule.  
It also allows validation to catchmore than one error at a time.
__
| Mortar: Advanced Web Development <http://bigpic.com/mortar/>
| Neil St.Laurent  neil@bigpic.com
| Big Picture Multimedia