Hehehe. Quite right...I just have to manage my time a lot closer these days and don't see a direct benefit (to myself) of trying to debate or discuss complicated internal tapestry design aspects unless it's with people that are already very familiar with the overall picture...Not trying to be an ass, just the way it is...
I suppose you could call what tacos is doing "full page rendering", but that's a sort of loaded term. Plenty of other things do full page rendering as well, like forms. This is a design point that originally made the tacos integration a little hard, but is also why tapestry "just works" for almost all normal use cases. I think this is a perfectly acceptable tradeoff for a framework. Complicated (at least initially ) internally, and easy to use externally. Direct component renders are currently hard because of what you already know, component id's , and not knowing what they will be in forms and loops. It's an issue that will be resolved in tapestry 4.1. I don't know what else to say about it. Give us a little time and it will be there? Unless there is some other use case floating around that I don't know about? You can't very well just instantiate a component and render it on the fly, just as much as you can't take a button in awt and render it on the fly without a Graphics object with which to render it. Granted, they make it a little easier in the client world, but the browser (and all the html constraints) doesn't allow this model to be ~quite~ as flexible and easy to use. (yet) Am I missing something? j On 1/30/06, Patrick Casey <[EMAIL PROTECTED]> wrote: > > > That was my understanding as well, although the last time I got > into > a discussion about the topic with the Tacos crew there was a degree of > "wait > and see ... we may be more clever than you think". > > --- Pat > > > -----Original Message----- > > From: Leonardo Quijano Vincenzi [mailto:[EMAIL PROTECTED] > > Sent: Monday, January 30, 2006 10:43 AM > > To: Tapestry users > > Subject: Re: tapestry not really component based? > > > > Jesse, correct me if I'm wrong, but an Ajax request involves a whole > > page render, right? From that page, the updated components are retrieved > > and sent through HTTP. > > > > That's whole page rendering, but partial update. Right? > > > > (by the way, I'm in the Tacos crew :P - just haven't been for such a > > long time as Jesse). > > > > -- > > Ing. Leonardo Quijano Vincenzi > > DTQ Software > > > > > > Howard Lewis Ship wrote: > > > If partial rendering was not possible, Tacos would not exist. But it > > > does, which indicates that rendering partial pages is possible. > > > > > > It's all just objects that implement methods, so partials are > > > completely doable. The Tacos crew, including Jesse, have that working > > > for 4.0. Jesse is working on improvements to the API for 4.1 to > > > support partial rendering more efficiently. > > > > > > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > >
