cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Unico Hommes" <Un...@hippo.nl>
Subject Progress on 2.2
Date Tue, 30 Dec 2003 12:09:15 GMT


Carsten Ziegeler wrote:
> 
> Unico Hommes wrote:
> >
> > Hi Carsten,
> >
> > I am trying to get sitemap mounts to work in 2.2 but am 
> running into a 
> > problem with context changing. It seems
> > EnvironmentHelper.changeContext() is not working for me. I find it 
> > difficult to understand how exactly to use the 
> EnvironmentHelper. In a 
> > child processors should I construct an EnvironmentHelper with the 
> > child processor's parent's EnvironmentHelper? How do the 
> > EnvironmentHelper prefix, lastPrefix and context member fields get 
> > initialized? Should I call changeContext(String,String) for that?
> >
> > Any light you can shed on this would be greatly appreciated :-)
> >
> Hmm, it's long ago...ok, I completly rewrote the environment 
> handling and tried to make it cleaner and faster. 
> Unfortunately, I could never test/run it, so there might be 
> many bugs in it! I think, the basic idea is to have on 
> EnvironmentHelper per sitemap. A helper of a sub sitemap 
> should be initialized with the parent helper, yes. And the 
> component has to be initialized using the usual Avalon 
> lifecycle interfaces, so you have to use the lifecycle helper 
> or the ContainerUtil.
> 
> As soon as Cocoon 2.2 is startable (with only a main sitemap) 
> I can have a look at it and find possible bugs. What's the status?
> 

Was just about to write a report. Basically, 2.2 is compiling and
running. If you do a build webapp and launch cocoon you should be able
to see the welcome page :-)

That's actually about all you will be able to though for the moment. The
componentized processor is really very sketchy still and mostly not
tested. I am just going to work through testing and fixing all the
individual sitemap nodes over the next few days.

In order to make things compile I've had to do some work on
I18nTransformer. It should work but I haven't tested.

XSP is completely broken ATM. There was some ECM specific stuff that I
had to throw out (GeneratorSelector) for which we need to come up with
something new.

I've also taken a look at what would be involved for putting back the
treeprocessor because componentized processor won't be stable for a
while. From what I could see I think it would be quite some work.
Especially due to some customized ECM stuff treeprocessor relies on
(ExtendedComponentSelector) that was removed. I'm doubtful it would be
worth the effort. Additionally, I've noticed some dependencies on the
processor implementation throughout the code (for instance in the flow
interpreter) that would need to be resolved.

Unico

Mime
View raw message