cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefano Mazzocchi <>
Subject Re: XML-Based Selection (Redirect Serializer?)
Date Thu, 28 Feb 2002 16:54:11 GMT
Michael Homeijer wrote:
> Hi Stephano,

Stefano, please.

> What do you think of the suggestion in the mailing list from Bruno Dumon?

As I stated previously, the question is: should a selector have
knowledge on what flows into the pipe? do the advantages outweight the

Ok, let's analyize this further.

> Most transformers that write data or writable sources/drains return a small
> amount of data (ie. success/failure, message, generated keyvalue). Would it
> be a performance problem to stream the result into a dom tree, use a
> selector (xpath based) to select a new pipeline based on the result and
> continue with the rest of the pipeline (with the option to stream the result
> into the new pipeline again). 

Yes, I think this can be done by extending the pipe object model:
passing DOMs around seems like a very bad idea to me. (at least, there
are easier ways to accomplish this, IMO).

> If this is considered not clean or resulting
> in a performance problem, maybe somebody (hint) could post an RT on how to
> support this with generic cocoon components, as from my opinion this is a
> "big" problem in the current cocoon implementation.

Please, explain what you consider a big problem in cocoon implementation
with a scenario of use. I think this is the best way to attack a
> In the current situation, the pipelines parts for incoming data and the
> parts for outgoing data are executed without having a possibility to change
> the outgoing pipeline based on the data returned from the incoming part.

Again, please, give us a clear example on this.
> I think we need a general solution to this kind of problem in which you will
> still be able to use standard cocoon components and not everyone having to
> write his own selectors (FS? ;-)).

My irrational perception says that having selectors react on what flows
into the pipes mixes concerns: in fact, selectors should 'route' the
content of the pipeline, while the pipelines content should 'passively'
be routed around by the external machinery.

What do others think about this?

Stefano Mazzocchi      One must still have chaos in oneself to be
                          able to give birth to a dancing star.
<>                             Friedrich Nietzsche

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

View raw message