I'd be very interested to see the XMLForm feedback wizard demo re-written to use the sitemap (flowmap) for navigation. The reason why I insist on another version of the demo instead of a new application is that it was specificly designed to cover many subtle problems with HTML form handling and workflow. Another implementation will also allow direct comparison between two flow techniques.
Andrew C. Oliver wrote: > Ivelin, > > I'm following up on the dev group as not to scare users off. > >> >> The HowtoWizardAction has an identical role as a Struts action. >> It makes the decision, which is the following *logical* page to go to. >> Notice that the *sitemap* converts the logical name to a *URL*! > > > > *so* ? > >> >> Actions are appropriate for evaluating the current form content, >> applying business rules and making decision what is the next logical >> page to go to. Actions *do not* deal with absolute *URL*s and they >> *only* aware and interested in pages which immediately follow as a >> result of the Action processing. > > > I disagree. Thats one solution. I'm going to look at doing other > things for the bringmethis auction example. I don't like the use of > actions here. Most of my actions are trivial classes, and the complex > ones I'm convinced can be handled via the sitemap. > >> >> In real world application decisions for the outcome of form submit can >> be quite complex, requiring the Action to consult with the business >> layer and possibly the database and other resources. > > > I disagree. In most real world applications, the decisions for handling > the outcome of a form submit is simple. What you do with that output is > complex. But using the Action to route to pages is not only "page" > centric (which I find inappropriate) but creates a larger compile test > debug loop and is very akward. Finding a sitemap approach is much > cleaner. > -Andy. > >> >> >>>>> 2. When I repeated the steps and unchecked the boxes. Then when I >>>>> arrive at >>>>> confirm input everthing is still true. Do I have to invalidate the >>>>> session >>>>> or something? Shouldn't that happen the moment you push finish? >>>> >>>> >> >> This should not happen. Possibly a bug. >> Can you reproduce the problem from the time you open a new browser >> window and send me the steps. >> >> >> Ivelin >> >> >> >>>>> >>>>> Thanks, >>>>> >>>>> Edgar >>>>> >>>>> --------------------------------------------------------------------- >>>>> Please check that your question has not already been answered in the >>>>> FAQ before posting. <http://xml.apache.org/cocoon/faq/index.html> >>>>> >>>>> To unsubscribe, e-mail: <[EMAIL PROTECTED]> >>>>> For additional commands, e-mail: <[EMAIL PROTECTED]> >>>>> >>>> >>>> --------------------------------------------------------------------- >>>> Please check that your question has not already been answered in the >>>> FAQ before posting. <http://xml.apache.org/cocoon/faq/index.html> >>>> >>>> To unsubscribe, e-mail: <[EMAIL PROTECTED]> >>>> For additional commands, e-mail: <[EMAIL PROTECTED]> >>>> >> >> >> > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, email: [EMAIL PROTECTED] > > -- -= Ivelin =- --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, email: [EMAIL PROTECTED]
