cocoon-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Bruno Dumon <>
Subject Re: JXTemplates - what's in a name?
Date Wed, 05 May 2004 08:03:37 GMT
On Wed, 2004-05-05 at 08:56, Derek Hohls wrote:
> Bruno
> Yes, I suggest you read the paper first before we continue
> the discussion on templates ;-)
> In the meantime, perhaps you can expand on why there should
> not be logic in generators such as XSP and JXT and, if not there, 
> where it should, in fact, be (short of writing custom Java code 
> each time) ?
> I agree that flow control should not be in XSP; but no data
> manipulation?  Even writing conditional SQL to bring out "part" 
> of the data, based on input parameters, is kind of manipulation, 
> and soon as you do something like add two values from a row 
> result on the basis of the third value, you have added logic.
> Where does all this go instead?? 

I thought you meant manipulation of external data. Of course,
manipulation of the data that goes through the pipeline is no problem,
after all, that's what each transformer does.

When seeing data manipulation, I first thought you meant things like
inserting database records, modifying object models, etc. In other
words, everything that has external side-effects.

Not that this is disallowed per-se, but I see a pipeline
(generator-transformers-serializer) as the the view part of an
application. People have also been successfully using the pipeline
itself for implementing all their business logic, and when that's your
vision of how stuff should be done, no problem.

>   And why are all the docs then 
> mis-leading us about doing these type of things?
> Thanks in advance for any advice or "better practice" suggestions...
> Derek

Bruno Dumon                   
Outerthought - Open Source, Java & XML Competence Support Center                

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

View raw message