[relax-ng-comment] augmented rng --xsl--> reference documentation

[Andreas Leitner <nozone@sbox.tugraz.at>]

Hi,

I am new to Relax NG so please excuse any stupid things I say and ask in
the following mail (;

For the GOBO Eiffel project (www.sf.net/projects/gobo) we are going to
use Relax NG schemas to specify our home-brewn XML grammars (which are
currenlty two). I just read the Relax NG tutorial and played a bit with
jing and I have to say Relax NG looks really nice: it's simple and yet
powerfull. I am glad we choose Relax NG over XML Schemas (;

What I am currently exploring is whether the following is feasable:

*) Produce reference documentation about a XML Grammar from an augmented
RNG Schema.

Now the augmentation, I imagine using simple additional tags like
<description> etc. The tutorial says this is explicitly allowed as long
as the extra tags have a seperate namespace.

Has anybody done something similar before? Could the whole process be
done using XSL only? I would prefere very much using XSL only.

What I imaging as output is similar to the zvon.org Relax NG
Specification (plus more descriptive text):

For each Schema I would like to output:

*) List of all elements (with link to detailed description)
*) List of all attributes (with link to detailed description)
*) notes, description and examples 

For each element I would like to output:

*) Possible attributes 
*) Possible child elements
*) Possible parent elements
*) notes, description and examples

For each attribute I would like to output

*) Possible values
*) Possible parent elements
*) notes description and examples


And the killer thing would be if this whole thing would somehow
integrate with DocBook, because that's (or maybe better a super/subset
what we use for the rest of the documentation)

I am interested in all kinds of feedback, be it "this does not make
sense, because", "this is far to complicated, much simpler would be",
"hey, that other guy is doing exactly the same thig/something similar" ,
...

many thanks in advance,
Andreas
PS: I am a bit unsure if the structure of my output can be archived at
all since a gloabal list of elements does not make sense if elements
with completly different semantics are allowed (each as child of
different parrents)