On Sat, May 16, 2015 at 5:21 AM, Peter Norwich <[email protected]> wrote: > Using replicator means client has to implement replication protocol.
Replication protocol is not a magic, but a usage of CouchDB API in special order to send only missing docs/revisions from source to target in the end. It's pretty good described in the docs: http://docs.couchdb.org/en/latest/replication/protocol.html In fact, it would be the same as want you already doing, but you'll add few more requests (see API reference in the end) which will help you avoid unnecessary conflicts creation and help to make your sync incremental. -- ,,,^..^,,,
