Michael Wechner wrote:
Andreas Hartmann wrote:
[...]
That's true, but the DocumentIdToPathMapper is just an indirection
between the content object model and the storage layer. It determines
how documents are mapped to nodes. IMO it is sufficient if the node
factory maps the nodes to files (or JCR nodes or whatever).
I don't fully understand how this relates to the above. Can you give an
example?
Sure:
- URL: /foo/bar.html
- Document ID: /foo/bar
The DocumentIdToPathMapper might map this to the path
/foo/bar/index_en
which leads to the repository source URI
lenya://[...]/foo/bar/index_en
Then, the Node object can map this arbitrarily for instance to Excalibur
sources.
IMO the customized mapping could be handled entirely by the Node class.
For documents, the URI would be for instance
lenya://[...]/foo/bar_en
and the Node could decide to map this to
context://[...]/content/foo/bar/index_en.xml
I'm not yet sure if we need this level of abstraction (handling documents,
site structures, user data etc. in the same way), but IMO it simplifies
things a lot and makes the repository layer more maintainable.
-- Andreas
--
Andreas Hartmann
Wyona Inc. - Open Source Content Management - Apache Lenya
http://www.wyona.com http://lenya.apache.org
[EMAIL PROTECTED] [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
