RE: [Fwd: XML Schema Part 2 should provide BNF for all primitive types.]

Paul and I have long wanted to add this but its not gotten done.
Its a significant amount of work.  I'm not sure we are allowed
to make this amount of change before rec.  Do the chairs have any
advice?  Paul, what do you think?

Would folks prefer BNF or regex's?

Ashok

-----Original Message-----
From: James Clark [mailto:jjc@jclark.com]
Sent: Monday, March 26, 2001 6:19 PM
To: www-xml-schema-comments@w3.org
Subject: [Fwd: XML Schema Part 2 should provide BNF for all primitive
types.]


I would strongly endorse Kawaguchi-san's comments below.  I have tried
to implement this and the spec is not clear.  There is no excuse for the
spec not to provide regexs for both the lexical space and the canonical
forms of all primitive types.  This needs to be fixed before Rec.

-------- Original Message --------
Subject: XML Schema Part 2 should provide BNF for all primitive types.
Date: Mon, 26 Mar 2001 12:18:30 -0800
From: Kohsuke KAWAGUCHI <k-kawa@bigfoot.com>
To: Steve.Rosenberry@ElectronicSolutionsCo.com
CC: "Thomas B. Passin" <tpassin@home.com>, xml-dev@lists.xml.org
References:
<001801c0b5fa$27f08720$7cac1218@reston1.va.home.com><3ABF53E0.EB4674A9@v
erizon.net>


> but none of these documents actually has the RE that defines a string
as
> a valid float value that I was hoping to copy (at least not that I
could
> find).

You couldn't find them because there is none. And I think the spec
should provide more precise definition of what are valid lexical
representation of primitive types. BNF is certainly a good way to do
this.

Good news is, XML Schema is still a PR. That means there is still a
chance to have them add BNF. So please post your comment to
www-xml-schema-comments@w3.org about adding
BNF to the spec. I've already posted one, but the same comment from a
different person should counts.


Currently, the spec describes lexical representation without using any
formal language. For example, you can find the following statement in
the description of "number" type.

> An optional leading sign is allowed. If the sign is omitted, "+" is
> assumed. Leading and trailing zeroes are optional. 

OK. I think "0.0" is a valid "number". Since leading zeroes are
optional,
it seems to me that the current spec allows ".0" as a valid "number".

There is more, by removing optional trailing zeroes, "." becomes a valid
"number". Really?  But why not?

These are happening everywhere in the spec, and much worse in date/time
related types.


So please let WG knows your thought.

regards,
----------------------
K.Kawaguchi
E-Mail: k-kawa@bigfoot.com


------------------------------------------------------------------
The xml-dev list is sponsored by XML.org, an initiative of OASIS
<http://www.oasis-open.org>

The list archives are at http://lists.xml.org/archives/xml-dev/

To unsubscribe from this elist send a message with the single word
"unsubscribe" in the body to: xml-dev-request@lists.xml.org

Received on Tuesday, 27 March 2001 13:10:24 UTC