At 10:44 PM 30/08/2002, Carsten Ziegeler wrote:
>I don't understand exactly what you mean - you could (easily) write
>a transformer interacting with business objects managed by a
>a prevalent system. For example for reading
><read businessObject="identifer for the object">
> <name>
> <street>
></read>
>
>This is interpreted by the transformer into something like:
> object o = getBusinessObject("identifier for the object">
> o.getName();
> o.getStreet();
>And the results are inserted into the SAX stream.
>
>And the way back could look like this:
><write businessObject="identifer for the object">
> <name>some name</name>
> <street>and a street</street>
></write>
>etc.
>
>Is this the way you mean?
No, probably because I hadn't seen a tie-in at the transformer level to
business objects before - I've only ever seen XSP used to access bo's and
at this point I am staying away from XSP (but as I said my knowledge of
Cocoon internals is minimal at this point in time).
The main reason I hadn't thought along these lines is that I am taking a
more document-centric rather than data-centric view and as such I want to
persist things as xml documents rather than 1 or more bo's.
Further to using the SourceWritingTransformer as a starting point, it
appears to be a write-only solution, requiring a separate generator to do
the reading. Maybe I am missing something important here but I think what I
am looking for is a non-stream based read/write component that can act as
both a generator and write-able transformer as does FileSource. Is there a
sample of such a beast or am I on the wrong track ?
Robert
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]
