Angela Schreiber wrote:
...
Isn't that incorrect for open-scoped locks?
you mean in terms of jcr? no.... not that i would
know of. the semantics of javax.jcr.lock.Lock.getToken()
will change slightly with jsr283 (regarding session-scoped
locks) but for with jsr 170 its defined to be:
"If this <code>Session</code> holds the lock token for this lock,
then this method will return that lock token. If this
<code>Session</code> does not hold the applicable lock
token then this method will return null."
and the dav-layer only uses open-scoped locks (if i'm not
mistaken).
...
Hm, wait. It took my another day to realize what that means in practice :-)
So when I use LOCK with Jackrabbit WebDAV, I get the locktoken back in
the LOCK response, but a subsequent PROPFIND/DAV:lockdiscovery will not
reveal the lock token?
...this definitively will cause problems for some clients -- not all of
them keep the lock token, and rely on DAV:lockdiscovery to re-discover
them. But with the JCR API, there seems to be no way to avoid that --
maybe we should document this restriction somewhere...
BR, Julian
