Re: CDATA sections in W3C XML Infoset

[Joe English <jenglish@flightlab.com>]

Richard Lanyon wrote:
>
> Could someone explain to me why CDATA section start/end markers were
> taken out of the W3C Infoset?

I don't know the official rationale, but it makes sense:
CDATA section boundaries are a lexical artifact, much
like the distinction between '"' and ''' in attribute
value specifications, whitespace in tags, and entity
references, all of which are also omitted from the
Infoset.

> I was hoping to use the Infoset in writing the spec for XML Script,
> but the removal of these markers makes it very difficult to write an
> implementation of XML Script which is based around a DOM.  [...]
>
> Is there any chance that CDATA section start/end markers could be put
> back in?

Probably not, but there's no reason why you can't include CDATA
section boundaries in *your* Infoset.  The W3C XML Infoset
is a lowest common denominator; for XML Script to be "Infoset
compliant" it shouldn't throw anything in the W3C infoset out,
but it's free to add any new items that are appropriate.

That's my understanding anyway.

--Joe English

  jenglish@flightlab.com