[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: XML 1.1 and other specs

• To: John Cowan <jcowan@r...>
• Subject: Re: XML 1.1 and other specs
• From: Elliotte Rusty Harold <elharo@m...>
• Date: Tue, 22 Oct 2002 09:49:00 -0400
• Cc: xml-dev@l...
• In-reply-to: <200210221344.JAA08947@m...>
• References: <200210221344.JAA08947@m...>

At 9:30 AM -0400 10/22/02, John Cowan wrote:

>In fact the Infoset has no rules about what is a valid name, either
>directly or by reference.  It's true that you can't construct an
>infoset with names that contain non-XML-1.0-name characters by parsing
>a 1.0 document, but the section "Synthetic Infosets" makes it clear
>that there are other ways to construct an infoset than by parsing.

In other words it's OK to have an element information item in which 
the name of an element is, for example, seven ASCII space characters 
in a row? My god, I had no idea.

What else can we throw out? The Infoset doesn't appear to require 
that there be only one ID type attribute per element. Hell, it 
doesn't even require that there be one element with the name foo. 
Give the element 23 foo attributes. The infoset's OK with it. Looking 
at 2.6. Character Information Items without assuming that XML rules 
apply, I now realize that infoset character information items can 
contain C0 controls including null, unmatched surrogate halves, and 
other nasty stuff. About the only well-formedness constraint the 
infoset seems to satisfy is the tree structure. Everything else is 
out the window.

Thanks. I think you've finally succeeded in convincing me of what 
Simon and Walter have been claiming for the last several years, that 
the Infoset is not only useless but actively harmful. It should be 
deprecated, and future specs such as XQuery/XPath 2.0 should base 
themselves on XML 1.0 plus namespaces and actively ignore the infoset.
-- 

+-----------------------+------------------------+-------------------+
| Elliotte Rusty Harold | elharo@m... | Writer/Programmer |
+-----------------------+------------------------+-------------------+
|          XML in a  Nutshell, 2nd Edition (O'Reilly, 2002)          |
|              http://www.cafeconleche.org/books/xian2/              |
|  http://www.amazon.com/exec/obidos/ISBN%3D0596002920/cafeaulaitA/  |
+----------------------------------+---------------------------------+
|  Read Cafe au Lait for Java News:  http://www.cafeaulait.org/      |
|  Read Cafe con Leche for XML News: http://www.cafeconleche.org/    |
+----------------------------------+---------------------------------+

• Follow-Ups:
  • Re: XML 1.1 and other specs
    • From: John Cowan <jcowan@r...>

• References:
  • Re: XML 1.1 and other specs
    • From: John Cowan <jcowan@r...>

• Prev by Date: Re: XML 1.1 and other specs
• Next by Date: Re: XML 1.1 and other specs
• Previous by thread: Re: XML 1.1 and other specs
• Next by thread: Re: XML 1.1 and other specs
• Index(es):
  • Date
  • Thread