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

Re: A multi-step approach on defining object-orientednatureofD

• To: Rich Salz <rsalz@d...>
• Subject: Re: A multi-step approach on defining object-orientednatureofDOM
• From: Paul Prescod <paul@p...>
• Date: Wed, 28 Aug 2002 10:22:16 -0700
• Cc: xml-dev@l...
• References: <Pine.LNX.4.33.0208251850120.29139-100000@e...>

Rich Salz wrote:
> 
> > Namespaces should never have been scoped. That's a big part of what
> > drove the implementation complexity up to such a large degree.
> 
> But it's conceptually much cleaner.  Who likes pascal style variable
> declarations at the top, isntead of local scope rules?

The current namespace design would be fine if applications just read XML
(as Pascal apps do) and never had to worry about editing or transforming
them. Admittedly namespaces are mostly a hassle for writers of
XML-generic tools that others build upon but we are still numerous.

> It also makes component software easier, since I can just locally use a
> "ds" prefix to indicate XML DSIG, without having to globally coordinate
> with every part of a program working on a document.

Each part of the program should not worry about its prefixes. The thing
that manages the XML representation (DOM or SAX writer) can figure out
what the right prefix is based on what was declared at the top level.
Even better, we could keep scoped namespace declarations but disallow
the reuse of prefixes and URIs so that no matter how they are expressed
*textually*, the prefix/namespace binding can be managed globally.

streamer.startElement(uri="http://.....dsig", localname="XXXX",
suggested_prefix="dc")

 * http://lists.xml.org/archives/xml-dev/200204/msg00170.html

Something needs to manage these mappings *anyhow* if you are going to
generate XML wherein the same namespace is declared a dozen times just
because that's how the code happens to be organized. I consider that an
important quality of implementation issue. After all, XML is supposed to
be human readable and that means minimizing the number of redundant URI
strings.
-- 
 Paul Prescod

• References:
  • Re: A multi-step approach on defining object-orientednatureof DOM
    • From: Rich Salz <rsalz@d...>

• Prev by Date: Re: problem adding new encoding to perl XML::Parser
• Next by Date: RE: Getting dropped off the mail list
• Previous by thread: Re: A multi-step approach on defining object-orientednatureofDOM
• Next by thread: Re: A multi-step approach on defining object-oriented natureof DOM
• Index(es):
  • Date
  • Thread