Scott Tavares wrote:
>
> ----- Original Message -----
> From: george stewart <[EMAIL PROTECTED]>
> To: Turbine <[EMAIL PROTECTED]>
> Sent: Sunday, January 09, 2000 3:53 PM
> Subject: Re: OPaL the Attr class
>
> <snip>
> > To get a better understanding of the current
> > implementation, it would be helpful to build a sample.
> > All you need is a derived class, one class map, a db
> > broker, and some test data. Try save, delete, and
> > retrieve and step into these with a debugger. This
> > will answer many of your questions.
> >
> > jon kindly provided sample data.
> >
> > The derived class just needs a bunch of set and get
> > methods for its data.
> >
> > Everything else is already there.
> >
>
> Whoa.. you have brought this thing along even further than i though you
> did... this is freaking cool! its giving me goose bumps to see this thing
> getting so close to actually being useful.
>
> as far as the "...one class map..." is how i got into that whole
> contestation with my new best buddy Brett ; )
I know I'm wasting bandwidth but I just had to insert a nice <grin/>
here... ;-)
> I had started to code
> something but i've been compelled to re-think it all. You had mentioned that
> you have coded an example also, is this something you are planning to post
> up? I know that we've talked about using Jyve to base the example on, is
> this what you have been working on?... i'm just curious to know where you
> are.
>
> I plan on over the next day or two schooling myself on this XML Schema stuff
> and go from there as per the class maps. I'm not sure at this point where
> this is going to take me but when i get there i will let everyone know by
> posting some proposals and requests for feedback.
Some suggestions (nicely phrased, even!):
1) Start out using XML Schema just as a straight "replacement" for the
DTD you outlined. I lost the earlier one(s), as my mail is screwy
today, but if you want to repost I can give you an example, or you can
work it out and post and I can help using that as a starting block. In
other words, don't jump straight into using XML Schema as a pure data
representation, as there are lots of complexities there that are easier
to get incrementally.
2) Move away from the sample data [temporarily] and using just the XML
Schema, generate some conforming XML documents (and we on the list can
too), ignoring if they would be useful. This is often a good test to
make sure your schema is right; amazing the odd ways you can "break" an
XML format that was intended without violating any validity constraints.
3) Once that's done, let's create *lots* of example class maps. That
will be the second "litmus" test of how robust the XML Schema will be.
[ 3a) If problems, iterate... ]
4) Once we have a working set of data and a Schema, let's delve into
more advanced Schema applications; like using XML Schema to represent a
database map, and then XML instance documents to represent subsets of
that data that logically are bound together; or XML Schema to represent
a global schema, and XML instance documents to represent different
vendor implementations... etc, etc.
Just my thoughts ....
--Brett
>
> -scott-
>
> ------------------------------------------------------------
> To subscribe: [EMAIL PROTECTED]
> To unsubscribe: [EMAIL PROTECTED]
> Problems?: [EMAIL PROTECTED]
------------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
Problems?: [EMAIL PROTECTED]
