Re: [relax-ng] ooRelaxNG (was: Limitation ...)

[John Cowan <jcowan@reutershealth.com>]

"MURATA Makoto (FAMILY Given)" scripsit:

> In most o-o programming languages, when we declare a variable of some 
> class, we allow not only instances of this class but also instances 
> of its subclasses.  Thus, we cannot determine the type of an 
> object from the class of the parent object.  In fact, each object 
> typically has a pointer to its class.

It seems to me, though, that it's fully determined from the element
name what the class must be, as far as RNG validation is concerned.
The instance-to-class link is essentially the element-to-element type link.

> I just mimicked this approach in my XML representation of instances.  
> Yes, instances do look very redundant, but this approach is very simple.
> I am not sure if the overhead of specifying oi:class is really problematic.
> >From the beginning, XML is verbose.

It's not the verboseness that bothers me, it's the appearance of flexibility
when there actually is none.  You can't introduce a new element name with
an existing class name, nor can you use a single element name with two
different classes in different places.

-- 
John Cowan <jcowan@reutershealth.com>     http://www.reutershealth.com
I amar prestar aen, han mathon ne nen,    http://www.ccil.org/~cowan
han mathon ne chae, a han noston ne 'wilith.  --Galadriel, _LOTR:FOTR_