Hi everyone, My use case (probably not common but not the only one either) is to ensure the database is accessed sequentially.
As you can find in my initial Stack Overflow post <http://stackoverflow.com/questions/30303071/neo4j-pessimistic-locking-with-jdbc-driver>, Liquigraph <https://github.com/fbiville/liquigraph> checks for a Lock node and creates it or wait for its deletion before proceeding. However, Liquigraph is *agnostic* of the underlying Neo4j setup, thus relying on the JDBC driver to do the plumbing. While my external synchronization <https://github.com/fbiville/liquigraph/blob/02b276897b5d3c5595714c2054f03c284b0e7062/liquigraph-core/src/main/java/org/liquigraph/core/api/LockManager.java> certainly reduces the time window during which another Liquigraph instances executes concurrently, the latter is still possible and could compromise data. I was therefore wondering why there was no syntax equivalent to SELECT FOR UPDATE <http://www.techonthenet.com/oracle/cursors/for_update.php> for Cypher. I am sure it brings nontrivial issues, but this seems a much better bet than the currently documented, do-it-yourself and *server-specific* pessimistic locking <http://neo4j.com/docs/stable/tutorials-java-embedded-unique-nodes.html#tutorials-java-embedded-unique-pessimistic> . I know it's not Christmas yet, but could such a language addition be implemented and even better, from 2.0.x branch onwards? :) Thanks in advance, Florent P.S.: I'd be thrilled to help anyhow on this subject -- You received this message because you are subscribed to the Google Groups "Neo4j" group. To unsubscribe from this group and stop receiving emails from it, send an email to neo4j+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
