Re: pragmas proposal - please read and comment

Are you suggesting that the string

   {my:blue}

be ambiguous because parseable both as a comment and as a pragma?

Or that the definition of comment be changed so that the grammar is
unambiguous?

Michael

> On 11,Jan2022, at 7:40 AM, Steven Pemberton <steven.pemberton@cwi.nl> wrote:
> 
> My very strongly held position is that pragmas should be a form of comment, thus using the { } notation, and not use the character set notation [  ].
> That would already help fulfil the pragma-oblivious requirement.
> 
> Steven
> 
> 
> On Tuesday 04 January 2022 18:45:31 (+01:00), C. M. Sperberg-McQueen wrote:
> 
> > A somewhat shorter version of the pragmas proposal Tom Hillman and I
> > have developed is now in the ixml repo on the proposal-pragmas branch:
> >
> > https://github.com/invisibleXML/ixml/blob/proposal-pragmas/misc/pragmas-proposal.md
>> 
> > Basic summary: syntactically a pragma is a pair of square brackets
> > enclosing an optional mark, a QName or URI-qualified name, and optional
> > pragma data (which may contain paired square brackets). Examples:
> >
> > [my:blue]
> > [Q{http://example.com/ns/props}color blue]
> > [their:rewrite
> > comment: -"{", (cchars; comment)*, -"}".
> > [ls:token] -cchars: cchar+.
> > ]
> >
> > The use of square brackets allows all processors to recognize them and
> > process or ignore them. The use of QNames allow processors to understand
> > whether the pragma is one they support or not using the same logic
> > familiar from the use of namespaces in other XML-related contexts.
> >
> > Pragmas also have an XML form, which follows naturally from the
> > grammatical definition of pragmas.
> >
> > To allow pragmas to use prefixed names in the usual way, we propose to
> > add a pragma for namespace declarations. A pragma whose QName uses the
> > prefix 'ixmlns' is a namespace declaration. Examples:
> >
> > [ixmlns:xsd http://www.w3.org/2001/XMLSchema]
> > [ixmlns:rng http://relaxng.org/ns/structure/1.0]
> >
> > As things stand, colons are not allowed in ixml nonterminal names, so
> > namespace declarations are used only in expanding QNames in pragmas. A
> > pragma-oblivious processor can thus ignore all pragmas. If we decide to
> > allow colons in nonterminal names and specify that namespace bindings
> > should apply to the XML output, then all processors willl need to
> > understand namespace declarations.
> >
> > MIchael
> >
> >
> 

********************************************
C. M. Sperberg-McQueen
Black Mesa Technologies LLC
cmsmcq@blackmesatech.com
http://www.blackmesatech.com
********************************************

Received on Tuesday, 11 January 2022 15:14:07 UTC