New schema spec

[Rick Jelliffe]

From: roddey at us.ibm.com <roddey at us.ibm.com>
 >I am interested in some comments on the new Schema spec. In
particular, what are
>people's thoughts about the fact that the AND connector has been
reintroduced
>and that SEQ,ALT,AND blocks now support an optional min/max repetition
count?

Excellent things. XML was right to simplify SGML and get rid of & and
exceptions. I hope XSchema will reintroduce both of them, and more.

>Just the m to n repetition system means that DFAs wouldn't work anymore
right

n{2 to 5} can be replaced by  (n, n, (n, (n, (n)?)?)?)

(n, m){2,5} can be replaced by ((n,m), (n,m), (n,m, (n,m (n,m)?)?)?)

(n|m){2,5} can be replaced by
    ((n|m), (n|m), (n, ((n, (n|m))
                                    | (m, (n|m))))
                            | (m, (n, (n|m))
                                   | (m, (n|m))))

(n&m){2 to 5} can be replaced by
     ( ((n, m)| (m,n)),
       ((n, m)| (m,n)),
       ( ((n, m)| (m,n)), ( ((n, m)| (m,n)), ((n, m)| (m,n))?)?)? )

What would upset things is if  (n&m){2 to 5} could be satisfied by the
input
     n n n n n m m m m m

Also, content models such as  (n?, m?){2 to 5} are ambiguous, but so
would
(n?, m?)* or (n?, m?)+

Rick Jelliffe

P.S. Am I the only one freaking out that the current Schema draft is not
compatible with XML 1.0?  It introduces a new class "nearly well-formed"
which is not WF: thus all current XML processors will treat these
documents in error. This is a terrible excercise in backwards
incompatability--all existing XML processors will be incompatible: the
only reason they have this class is because they try to cram into the
Schema spec some way to declare entities (I thought that XLink was our
way to improve entities: now we have 3 ways to do them...Yikes).  It
reduces XSchema's credibility a lot--it makes it look like there is a
petty dislike of XML's markup declaration syntax which was so strong
that it overweighed the requirement to conform to XML 1.0 WF rules.

P.P.S. The archetype and datatype parts of the Schema proposal are
excellent, IMHO.


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)