Re: [Neo] Question about "Neo mistakes" wiki page: What's wrong with the synchronization example?

2010-02-14 Thread Tobias Ivarsson
Any modification to a Node will block it for other transactions. Assume nodes A, B and C, A and B connected with a relationship (AB), let's look at a few scenarios (all starting from this initial state): tx1 changes property on node A, tx2 changes property on node B: no conflict tx1 changes prop

Re: [Neo] Question about "Neo mistakes" wiki page: What's wrong with the synchronization example?

2010-02-12 Thread Thomas Andersson
Hi, Thanks for the quick and detailed reply! This mailing list is a great help. I have a follow-up question: You say that Neo4j "will synchronize access between concurrent transactions (on a node and relationship level)". I am a bit curious about the granularity of the synchronization on the node

Re: [Neo] Question about "Neo mistakes" wiki page: What's wrong with the synchronization example?

2010-02-10 Thread Johan Svensson
Hi, On Wed, Feb 10, 2010 at 10:48 AM, Thomas Andersson wrote: > Hi, > ... > // don't do this > synchronized void methodA() > { >    nodeA.setProperty( "prop1", 1 ); >    methodB(); > } > > synchronized void methodB() > { >    nodeB.setProperty( "prop2", 2 ); > } > > According to the test, "The co

[Neo] Question about "Neo mistakes" wiki page: What's wrong with the synchronization example?

2010-02-10 Thread Thomas Andersson
Hi, I have just started using neo4j and have just read through the documentation found on the wiki and have a question about synchronization. On the neo mistakes page ( http://wiki.neo4j.org/content/Neo_Mistakes ), there is an example of what not to do called "Unnecessary synchronization" that rea