on 5/28/03 7:22 AM Carsten Ziegeler wrote: > Stefano Mazzocchi wrote: > >> <map:serializer name="xhtml"> >> <map:transformer type="link-translation"/> >> <map:serializer type="xhtml"/> >> <map:serializer> >> >> <map:match pattern="..."> >> <map:generate src="..."/> >> <map:transform src="..."/> >> <map:serialize type="xhtml"/> >> </map:match> >> > I think this is a very useful concept. Do you remember our discussion from > two years ago about adding url rewriting to the html serializer?
Yes I do. I've been thinking about an elegant way to solve that since then ;-) (well not continously, of course) > I wanted > only to write <map:serialize type="xhtml"/> doing url rewriting and > the usual xhtml serialization stuff and you said, adding a transformer > before is much cleaner. Yep. I think that anything that happens transparently is harmful in the long-term (for example, the fact that actions don't expose the name of the parameters the pass, unlike matchers who always use a fixed numbering scheme) > (Yes, sometimes I'm a little elephant remembering > technical discussions :) ). With the concept from above we satisfy > both of us, get a clean SoC and easy use as well. Nice! I'm happy you like it. >>Resources were supposed to be reusable pipelines, but *complete* ones! >>Later, they were implemented to be usable as pipeline fragments but, >>IMO, they impose some readability problems in the sitemap. The above >>aims to correct that. >> >>In fact, if the above is introduced, I would deprecate the use of >>resources that do not have a complete pipeline. >> > > +1, the most use cases for resources I saw is exactly the thing from above. Cool. Sounds like many people are liking this ideas. But let's defer this for post-2.1 when I will have more sitemap-based RT to throw in. -- Stefano.
