On Thu, 11 Apr 2002 20:13:28 -0500, "Ivelin Ivanov" <[EMAIL PROTECTED]> wrote:
> > Ovidiu, > > did you miss this question, or I'm getting to the point where you stop > responding ;)) No, I saw the message, but forgot to respond to it. Thanks for the reminder! > > A random thought: > > > > > > <map:pipeline> > > <map:match test="mywizard"> > > <map:fetch src="page1"/> > > > > <map:select> > > <map:when test="hasCar"> > > <map:fetch src="carInfo"/> > > </map:when> > > </map:select> > > > > <map:fetch src="confirmData"/> > > > > <map:select> > > <map:when test="confirmed"> > > <map:call src="cocoon://commitAction"/> > > </map:when> > > <map:otherwise> > > <map:call src="cocoon://home"/> > > </map:otherwise> > > </map:select> > > > > </map:match> The problem with this approach is that the flow is described in the sitemap, which has no notion of continuation. You'd have to translate this in an equivalent JavaScript program or something similar. I went down this path with my previous design of Schecoon, whose sitemap engine was written in Scheme. There's just too much work to do it, and the benefits of such an approach are debatable. In general, I don't see why would anyone try to come up with a programming language whose syntax is based on XML. It may be OK to express declarative type languages in XML, like the finite state machine Konstantin implemented. But to implement full blown languages, this is a bit too much: languages are for humans, not for machines. Regards, -- Ovidiu Predescu <[EMAIL PROTECTED]> http://www.geocities.com/SiliconValley/Monitor/7464/ (GNU, Emacs, other stuff) --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, email: [EMAIL PROTECTED]
