Re: atoms, molecules

["Simon St.Laurent" <simonstl@simonstl.com>]

At 10:57 AM 4/19/01 +0200, Anders W. Tell wrote:
There is another mechanism which may be a part of a solution.
>I have ealier proposed to the XML schema group a mechanism where it is 
>possible
>to handle "parsable encodings" such as XPath expressions and SVG paths in 
>strings (attributes, etc.),
>called "microparsing".
>In these cases are regular expressions not really sufficient because of 
>complex parsing rules.

Is 'microparsing' something readily defined with a declarative unpacking 
mechanism?

Or does it bind more tightly to a particular processing environment?

Regular expressions don't solve every problem, but I find them attractive 
because of their relative independence from particular platforms and their 
inclusion in XML Schema.

>What I proposed was a mechanism to annotate "compressed" attributes with 
>an equivalent
>XML representatation, ie a binding between string and a corresponding XML 
>fragment.
>This means that XPath strings can be unpacked or tokenized into XML.
>
>However this leaves some problems:  how should these specialized parser be 
>identified and bound at runtime.

I'd think some kind of annotation mechanism might be able to help here - 
sounds like the microparsers would just need identification and some kind 
of infrastructure.  It's not that different from what already happens in a 
lot of applications, just abstracted a layer higher to become metadata 
about a type.


Simon St.Laurent - Associate Editor, O'Reilly and Associates
XML Elements of Style / XML: A Primer, 2nd Ed.
XHTML: Migrating Toward XML
http://www.simonstl.com - XML essays and books