- From: Charles F. Goldfarb <Charles@SGMLsource.com>
- Date: Fri, 13 Sep 1996 00:25:20 GMT
- To: Tim Bray <tbray@textuality.com>
- Cc: w3c-sgml-wg@w3.org
On Wed, 11 Sep 1996 13:13:35 -0700, Tim Bray <tbray@textuality.com> wrote:
From the postings on this subject so far it would seem that:
1) MS INCLUDE/IGNORE is useful in a DTD.
It helps tailor a DTD for different development tool quirks (and version
control, and other things). However, it requires a parser to keep a stack
because the MSs can be nested.
2)MS CDATA/RCDATA is useful in an instance.
It lets you imbed TeX directly, for example (]]> hardly ever occurs anywhere).
As Erik Naggum has pointed out, it is safer to use an MS for a special parsing
context because there is a visible label. For XML, there is the added advantage
that you don't need to parse the DTD, as you would for a CDATA or RCDATA element
type.
So, if the above are deemed valid requirements for XML, we can simplify the
implementation of MS as follows:
INCLUDE/IGNORE is allowed only in a DTD and may nest.
CDATA/RCDATA is allowed only in an instance and may not nest.
TEMP is allowed anywhere; it is just a comment (or disallow it and save an entry
in the keyword table).
--
Charles F. Goldfarb * Information Management Consulting * +1(408)867-5553
13075 Paramount Drive * Saratoga CA 95070 * USA
International Standards Editor * ISO 8879 SGML * ISO/IEC 10744 HyTime
Prentice-Hall Series Editor * CFG Series on Open Information Management
--
Received on Thursday, 12 September 1996 20:23:28 UTC