Jon Stevens wrote:
> <http://freshmeat.net/news/2000/08/01/965139735.html>
>
> not sure if you guys can use this at all...but it sounds cool.
>
> -jon
Kevin wrote:
> Castor currently does this.. both ways. :) XML -> Java (using XML
> Schema) or Java -> XML (Using Reflection)
We probably can contribute our code to turbine (making it available
under BSD/ASF license - I need to check with my project leader).
See the language outline specification and examples at:
ftp://ftp.dfd.dlr.de/put/reck/XML_Configuration.html
It is more powerful than IBMs BML (Bean Markup Language)
(at http://www.alphaworks.ibm.com/tech/bml ) which is restricted
to beans. Similar applies to Castor (http://castor.exolab.org/)
and KBML (http://www-sop.inria.fr/koala/kbml/), but are much richer in
tools around them. Suns JDK 1.3 serialization is also neat but limited.
The DLR XmlConfig reader allows defining any type of objects, many into
one file and instantiating them when requested. It allows reuse of
object definitions using references (e.g. defining an URL radix once
and appending filenames). The reference to the instantiated object
is cached and returned in further accesses. A writer allows serializing
objects that conform to the bean specification (as well as allowing
direct access to public fields).
The sources XmlConfigReader.java has 1350 lines of code (20K class file)
and the XmlConfigWriter.java has 1013 LOC, 13K class file. these classes
are running either with SUN XML DOM (for over a year now) or Xerces
(test).
Please let me know if this is of interest for the Turbine (or Avalon)
project.
- Christoph
------------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
Search: <http://www.mail-archive.com/turbine%40list.working-dogs.com/>
Problems?: [EMAIL PROTECTED]
