Hello,
I have questions about how /jcr:root/jcr:system/jcr:typnodes is supposed to work.
If I'm not mistaking, the specification mentions that the exposure of jcr:typnodes is left up to the implementation. Furthermore, when a customized typenode is created, it should be created under /jcr:root/jcr:system/jcr:typnodes. And actually, this is done perfectly by CRX.
The point is the jcr:typnodes node doesn't exist when the repository is created. It has to be created manually. Is it desired ? Are there plans to create a internal node (like jcr:versionStorage) for jcr:typenodes ? And expose all the created type nodes under that node ? In this case, what to do with the system type nodes like nt:base, nt:folder ? Expose them under jcr:typenodes or not ? - On on hand, it might be expected to have them under jcr:typenodes (as protected of course). It is quite elegant (after all AFAIK metadata is treated as plain data by the jcr). - On the other hand, do we have the risk that it make an export of jcr:typnodes not useable, because we will have to strip the definiton of system type nodes before importing the file again. It doesn't make sense to import nt:base for example, does it ?
Finally, when a node type is added, it has to be registered by jackrabbit specific methods. Once again, I'm wondering if this is a feature, or simply because jackrabbit is under development.
If some of these points are not features and requires work to get them right, I would be very interested in helping.
What do you think ?
Regards, Bertrand.
This message contains information that may be privileged or confidential and is the property of the Capgemini Group. It is intended only for the person to whom it is addressed. If you are not the intended recipient, you are not authorized to read, print, retain, copy, disseminate, distribute, or use this message or any part thereof. If you receive this message in error, please notify the sender immediately and delete all copies of this message.
