cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Christian Haul <>
Subject Re: Variable Expanders
Date Fri, 11 Oct 2002 11:56:09 GMT
On 11.Oct.2002 -- 09:38 PM, Jeff Turner wrote:
> On Fri, Oct 11, 2002 at 11:49:50AM +0200, Christian Haul wrote:
> > On 10.Oct.2002 -- 11:50 PM, Stefano Mazzocchi wrote:

> [snip naming debate]

Just to make things clear: I don't _really_ mind changing the name.

> [snip (unneeded) expander-sets]
> > > isn't it more precisely describing what you want to achieve?
> > 
> > No. It changes a name and not for the better and places the
> > declaration to a place were SoC is breached. What really is needed is
> > multiple xconfs. Most (all?) sitemap components should be in a local
> > xconf.
> How would multiple configs help? I'd imagine:
> cocoon.xconf -> declare and configure generic components
> sitemap.xconf -> declare and configure sitemap components (map:components stuff)
> sitemap.xmap -> use components
> But even then, input modules would belong in cocoon.xconf (being
> generic), so multiple configs isn't the answer to the problem.

One would only declare stuff relevant to the application. This is
mostlikely a different person than the one caring about cocoon.xconf.

> > Until then, it should stay in the one that we have.
> How about if they could be declared in either cocoon.xconf (when used
> generically) or sitemap.xmap (when used as 'expanders')? Two different
> 'contexts', two different configuration syntaxes.

And another clarification: I don't _really_ mind declaring them in

> > BTW did someone notice, that these thingies that we spend so much time
> > about now are in CVS since Q1 02??
> :) Probably because no-one had any clue they were _useful_. Their general
> usefulness was so well camouflaged by docs like the one above.

And a third one: I didn't intent to blame anyone for not noticing
earlier. I just wondered why it draws so much attention now.

And yes, I hope to find some spare time during the week end and come
up with a new "glue" module like the one we discussed and agreed
on. In addition, I will change the interface to return an Iterator
instead of an Enumeration. Provided I find the time.

Oh, I almost don't dare to suggest it: but what if a module could gain
access to the sitemap variable stack? a) it could provide easy XPath
access to all vars and b) could provide any naming scheme desired.

C h r i s t i a n       H a u l
    fingerprint: 99B0 1D9D 7919 644A 4837  7D73 FEF9 6856 335A 9E08

To unsubscribe, e-mail:
For additional commands, email:

View raw message