SAX2 RFD: LexicalHandler draft v.1.1

[Kay Michael]

> >Do we really need to know about CDATA sections 
> 
> 
> Debatable perhaps, but supported by the DOM. (Anyone know why?)
> But I'd really like to see better SAX/DOM integration, so Yes!
> 

It seems we are trying to provide two views of a document, the reader's view
and the writer's view. The reader's view needs to present roughly what's in
SAX1. The writer's view arguably should preserve all the arbitrary choices
made by the document author, including whether to use CDATA or entity
references or character references, where to put the line breaks, whether to
use empty element syntax, where to put optional spaces, what kind of quotes
to use round attributes, etc, etc. If we are retaining any of this for the
benefit of people who want to edit the document, then logically we should
retain all of it.

Mike Kay
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ic.ac.uk/pipermail/xml-dev/attachments/19990322/c3c24046/attachment.htm