On Mon, 2002-05-13 at 15:00, Jonathan Robie wrote:
> >Or you can use tools like Regular Fragmentations to break the lexical
> >parts down into markup-identifiable and unambiguous day/month/year
> >elements.
>
> Or you could use a regular expression to do the same thing, and associate
> the type with a name that identifies the regular expression to be used. And
> allow other constraints, such as ranges, to be associated with the name.
That's pretty much precisely what Regular Fragmentations does, actually,
though it leaves constraints for a later stage in the pipeline.
> Ooops, I just reinvented XML Schema simple types.
Except that XML Schema doesn't let you do anything with the results of
the regular expression except binary "yes it is/no it ain't". With
RegFrag, it's easy to say "this part is a XYZ and that part is a CBA."
I'm just pointing out that typing isn't the only option for markup
processing by far, and it may in fact be a very poor approach.
--
Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
http://simonstl.com
