Re: [relax-ng] Combining RELAX NG Grammars

[Eric van der Vlist <vdv@dyomedea.com>]

On Thu, 2003-01-02 at 17:02, John Cowan wrote:
> Eric van der Vlist scripsit:
> 
> > It's often the case with embedded structures (in programming languages)
> > that each level inherits the definitions done in its "ancestors" (with
> > the possibility to overide them) and I am wondering why this is not the
> > case with Relax NG. 
> 
> I can't say specifically, but I've always considered this automatic
> import to be a bug in Algol 60 that propagated to its intellectual descendants.
> References to outer scopes should IMHO be explicit ones.

That's a strong POV :-) ...

> > Also, while it is possible to refer to named patterns of the parent
> > grammar, it doesn't seem to be possible to refer to named patterns of
> > other ancestors and I am wondering what's the reason for this.
> 
> I proposed this at the time, but it was difficult to find use cases
> outside the simple parent-child relationship.

Yes, I agree, but the whole point of nested grammars is to be a
convenience to facilitate the combination of grammars and it's difficult
to see the full scope in which it will be used...

I guess that if this was really needed, "relais" could be defined in
each of the intermediate grammars to give access to named patterns in
their parents like:

grammar {

  my-pattern=...

  grammar {

   my-parent-pattern = parent my-pattern

   grammar {
   ...
     parent my-parent-pattern
    
that's verbose and requires an adaptation of the intermediate grammars
but should do the trick in some cases.

Thanks

Eric
-- 
Curious about Relax NG? My book in progress is waiting for your review!
                                   http://books.xmlschemata.org/relaxng/
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------