Re: Relax NG specs

[Kohsuke KAWAGUCHI <kohsuke.kawaguchi@eng.sun.com>]

> | How accessible is the current TREX formal spec?  Is it too hard?  Do we need
> | something providing the same information as the formal spec but in an easier
> | to understand, less formal way (perhaps with the formalism in an appendix)?
> 
> The current formal spec for TREX is tough going. OTOH, I like its
> formality and I think that's an important aspect of this work.

I think the current formal spec of TREX is very legible, especially
compared to that of the other.

But I would suggest the use of the following syntax,

> function boolean allowsChildren(pattern p) {
>   switch (p) {
>   case <oneOrMore> p1 </oneOrMore>:
>      return allowsChildren(p1);
>   case <interleave> p1 p2 </interleave>:
>   case <group> p1 p2 </group>:
>   case <choice> p1 p2 </choice>:

rather than the current M [[p]](a, c, e) syntax.



--
Kohsuke KAWAGUCHI                          +1 650 786 0721
Sun Microsystems                   kohsuke.kawaguchi@eng.sun.com