Alexander Klimetschek wrote:
On Fri, Sep 4, 2009 at 6:50 PM, Kenneth Yue<[email protected]> wrote:
Any ideas? I'm beginning to think this may be a bug.
org.apache.jackrabbit.jcr2spi.RepositoryImpl.java expects a
javax.naming.spi.ObjectFactory implementation, but I looked over the entire
Jackrabbit workspace and found only the following classes implement
javax.naming.spi.ObjectFactory:
org.apache.jackrabbit.core.jndi.BindableRepositoryFactory
org.apache.jackrabbit.jcr2spi.RepositoryImpl$Factory
org.apache.jackrabbit.rmi.client.ClientRepositoryFactory
BindableRepositoryFactory and ClientRepositoryFactory are unrelated to
WebDAV. RepositoryImpl$Factory is, well, the code that's doing the
checking, So is the code looking for itself??? How does the code expect
users to use it? Does the code expect users to write their own
ObjectFactory implementation and plug it in? Then how exactly is
org.apache.jackrabbit.client.RepositoryFactoryImpl useful?
What do you all think? Thanks.
Maybe this thread helps: http://markmail.org/message/4oy65gmvjttg42ht
Unfortunately it doesn't help. I was the one who started that thread
and the one who ended that thread by posting the last message:
http://markmail.org/message/r5jbob2f7pob77k7
I added the request of an ObjectFactory implementation to
https://issues.apache.org/jira/browse/JCR-1877 but this issue has been
fixed still without an ObjectFactory implementation. :-(
There is a related issue https://issues.apache.org/jira/browse/JCR-2041
which is also marked as fixed. But again, still no ObjectFactory
implementation.
I would love to have someone explaining to me what the logic in
RepositoryImpl$Factory is. Why is it checking for an ObjectFactory
implementation while none exists (besides BindableRepositoryFactory and
ClientRepositoryFactory which are totally unrelated)? Thanks.
Regards,
Ken
