Thank you, but we decided not to going on with the jcr2spi/spi2davex solution since the network latencies were too costly. We are writing our own client/server application with a local jackrabbit on the server.
François Le 2010-03-04 à 8:33 AM, Michael Dürig a écrit : >> >> Depends on the tree. If you fetch the root node and spi fetches more >> than one level, you might potentially include a lot of nodes, more >> compared to a more deeper subtree. >> >> If you are using spi2davex for the remote connection, you can change >> the BatchReadConfig to not fetch deeper levels. See also >> http://markmail.org/message/nzcbrf27b2edkbzr for some links. > > Some more details on how caching and batch reading interact here: > http://markmail.org/message/qwh3rwstpjr56ghs > > Michael
