Return-Path: Delivered-To: apmail-cocoon-users-archive@www.apache.org Received: (qmail 31777 invoked from network); 7 Apr 2004 22:08:36 -0000 Received: from daedalus.apache.org (HELO mail.apache.org) (208.185.179.12) by minotaur-2.apache.org with SMTP; 7 Apr 2004 22:08:36 -0000 Received: (qmail 30528 invoked by uid 500); 7 Apr 2004 22:07:40 -0000 Delivered-To: apmail-cocoon-users-archive@cocoon.apache.org Received: (qmail 30391 invoked by uid 500); 7 Apr 2004 22:07:39 -0000 Mailing-List: contact users-help@cocoon.apache.org; run by ezmlm Precedence: bulk list-help: list-unsubscribe: list-post: Reply-To: users@cocoon.apache.org Delivered-To: mailing list users@cocoon.apache.org Received: (qmail 30117 invoked from network); 7 Apr 2004 22:07:37 -0000 Received: from unknown (HELO mail.gmx.net) (213.165.64.20) by daedalus.apache.org with SMTP; 7 Apr 2004 22:07:37 -0000 Received: (qmail 2141 invoked by uid 65534); 7 Apr 2004 22:07:43 -0000 Received: from a183069.studnetz.uni-leipzig.de (EHLO gmx.de) (139.18.183.69) by mail.gmx.net (mp006) with SMTP; 08 Apr 2004 00:07:43 +0200 X-Authenticated: #3483660 Message-ID: <4073D124.9080106@gmx.de> Date: Wed, 07 Apr 2004 12:00:04 +0200 From: Joerg Heinicke User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.6) Gecko/20040113 X-Accept-Language: de-de, de, en-us, en-gb, en MIME-Version: 1.0 To: users@cocoon.apache.org Subject: Re: Toward better cohesion in the sitemap References: <3B376616-7DB8-11D8-9989-000A95908E0E@wrinkledog.com> In-Reply-To: <3B376616-7DB8-11D8-9989-000A95908E0E@wrinkledog.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 8bit X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N X-Spam-Rating: minotaur-2.apache.org 1.6.2 0/1000/N On 24.03.2004 18:25, Mark Lundquist wrote: > Partitioning the 'container' elements for pipelines the way the sitemap > does today yields a kind of "utility cohesion" which is not so great. > The aspects I have in mind specifically are (a) as the > container for , and (b) @internal-only applying at the > level. > > (BTW, the colloquialism seems to be to refer to matchers or selectors as > 'pipelines', which totally makes sense IMHO. I'll follow that > convention here, and I'll use "" to refer to that element > itself. To be honest, it's not really clear to me what a > is, other than a series of 'pipelines' that share one or the other value > for the "internal-only" property. The mismatch between the formalism > and the colloquialism seems unfortunate... If I'm totally off-base > here, I'm sure someone will enlighten me...) No, you are not, at least I have the same feelings here, then we were already two :) map:pipeline does not only hold one pipeline and the name is therefore irritating. But using pipeline for a matcher is also a bad idea: How many "pipelines" would you have in this sample? Even without nested matchers you don't need to have a complete pipeline inside on matcher, e.g. when using resources. > Anyway... when using flow with CForms or JXTemplate, the same pattern > occurs over and over again in my sitemap: > > > > > . > . > . > > Now, the matcher for "something.display" really belongs in an internal > . But to do that, I have to violate the "functional > cohesion" of having these two matchers occur together. I like having > them together, it makes it easier to tell what's going on (and I can > make new instances of the pattern with a single "yank/put" in the > editor...). It's easy to fix as above. More to type, but therefore a map:pipeline really contains one pipeline ;-) I only have one of those pairs in my cforms app sitemap at the moment, but that might change. > I wish that I could just write > > > > Similarly, if I have a resource that's meant to be called by just a > couple of pipelines, it would be nice to be able to locate it > contiguously w/ the pipelines that call it � once again, it just would > make the sitemap easier to follow IMHO. Especially if you're using > uplevel naming of matcher parameters (e.g. "{../1}") in the resource. > If were allowed in the same context as , then > I could put the related elements together in the same "block" or > division in the sitemap, comment it all together etc. > > To do this right, it s would probably need to be nestable > like matchers, actions etc. � consider a set of pipelines protected by > ... you'd want to be able to locate the resource > inside the action along with the pipelines that call it. For such discussions you probably have to go to the developers list to get a response. Joerg --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org For additional commands, e-mail: users-help@cocoon.apache.org