Sylvain Wallez wrote: > Stefano, the above "outside-in" sentence shed a completely new light on > my understanding of views, and I'm now convinced that views should start > from the last label !
That's great! > The point that made me wait before answering was about the execution of > indirect components and their possible side-effects (actions). I now > think that side effects of indirect components are part of a pipeline > semantics. And views should not change this semantic, but only provide a > different mmmh... view ;) on the pipeline result. Precisely :) the name 'view' in fact, comes exactly from this thought! > So branching on a view should not change the actions taken in the > pipeline. If we need different side-effects, then we should consider > having different pipelines. What do you think ? Absolutely! in fact, the design pattern I use for 'is this a view or another resource' is something along the lines of: if this 'data I need' can be applied to an entire collection of resources, I need a view, otherwise, I need another resource. > Ah, and since you gave us so brilliant explanations, would you please > (finally ;) consider answering my other question about views and > aggregation ? I'm doing right after this. > In short, the question is about the behaviour associated > to <map:part> labels ? Should they cause view branching, or only filter > those parts that belong to the view, branching being decided by the > labels on <map:aggregate>. My (current) opinion is filtering, but you > may again have an explanation that will make me change my mind. > > Again, see > http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=101683844805545 for > more details on this. > > Thanks for being such a great mind and sharing it with us. I'm no great mind, Sylvain, I'm just overly concerned by esthetic elegance :) And believe me, here it is considered a value, in some other realms surely it is not :/ -- Stefano Mazzocchi One must still have chaos in oneself to be able to give birth to a dancing star. <[EMAIL PROTECTED]> Friedrich Nietzsche -------------------------------------------------------------------- --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, email: [EMAIL PROTECTED]
