Sylvain Wallez wrote: > > Carsten Ziegeler wrote: > > >Sylvain Wallez wrote: > > > >><snipped> > >> > >> > >>But I have also have a question : should enter/leaveEnvironment be > >>called only once per request (i.e. on the root sitemap), or also on > >>every nested subsitemap ? > >> > >>In sitemap.handler, it seems to be called for each subsitemap, which > >>means for example that a RequestLifecycle component looked up in the > >>main sitemap is hidden when we enter in a subsitemap. Is this the > >>intended behaviour ? > >> > >Argh, no of course not - the intended behaviour is that a > RequestLifecycle > >component is valid for a request regardless in what sitemap it > is looked up. > >But, if an internal request is process (using the cocoon: protocol), > >the RequestLifecycle component of the calling process are not > valid inside > >the internal request. That is the intention of the lines in the > >sitemap.handler. > > > >Well, lets see if I can fix it somehow. Thanks for reporting it! > > > >Carsten > > > One more detail : should a "cocoon:" URL considered as a new request, or > should it inherit RequestLifecycle components of the calling request ? > > I would say it should be considered as a new request, in order to behave > independently of the calling environment, but what do you think ? > Yupp, it's a new request. That was actually what I meant by my previous mail and this should be fixed with my patch from yesterday (which is a little bit hacky, but it should work).
Thanks, Carsten --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, email: [EMAIL PROTECTED]
