Hi,
I just committed a prototype implementation of a JCR repository based on the Microkernel into the sandbox [1]. In contrast to the SPI based prototype [2], this is direct implementation not relying on an existing stack. This is very much work in progress still and many (if not most) things are not implemented. What currently (should) work is standard read and session write operations. However, there is no support for name spaces nor for node types. Furthermore not all property types are supported yet.
The intent of this implementation is to be able to experiment with different approaches. It should help us to decide whether we want to go with a SPI based implementation and refactor this stack to fit our needs or whether we want to do a rewrite and reuse existing code where possible.
Michael [1] http://svn.apache.org/viewvc/jackrabbit/sandbox/jackrabbit-microkernel/ [2] http://svn.apache.org/viewvc/jackrabbit/sandbox/jackrabbit-mk/ On 8.12.11 18:18, [email protected] wrote:
Author: mduerig Date: Thu Dec 8 18:18:42 2011 New Revision: 1212018 URL:http://svn.apache.org/viewvc?rev=1212018&view=rev Log: add Microkernel based prototype of JCR implementation (WIP)
