Beautiful!
Yep. Nice addition to the TinkerPop toolbox.
Josh
> What's the motivation for allowing an Edge to reside on a partition other than the Partition of both its "in" and "out" vertices?
An edge across partitions can be seen as a merge point for the two partitions.
A user can write/read to partition "a".
Another user can write/read to partition "b".
A final user can read from "a", "b", and "c" and write to "c". As such, this final user is able to make a new partition that is the merge of the first two users' work.
Cool?,
Marko.
Yea. However, if you are doing it via the PartitionGraph wrapper, then you can do:
.sideEffect{it.setPartition('x')}
Marko.
An edge across partitions can be seen as a merge point for the two partitions.
A user can write/read to partition "a".
Another user can write/read to partition "b".
A final user can read from "a", "b", and "c" and write to "c". As such, this final user is able to make a new partition that is the merge of the first two users' work.
Good news. Considering to load RDF quadruple using SailGraph#loadRDF and naming the named graph with the "baseGraph" parameter, what is the best way to use this new implementation of graph ? in order to traverse a particular named graph or a set of named graphs.
Alex