Re: [Neo] how to decouple transaction management from access to underlying nodes

2009-06-29 Thread Johan Svensson
On Thu, Jun 25, 2009 at 2:51 PM, Mattias Persson wrote: > 2009/6/25 Symeon (Akis) Papadopoulos : >> [My apologies for insisting on this issue.] >>> There's (at least) two solutions to this problem. One is that, as you >>> suggest, to have the iterator wrap its methods in transactions in >>> additio

Re: [Neo] how to decouple transaction management from access to underlying nodes

2009-06-25 Thread Mattias Persson
2009/6/25 Symeon (Akis) Papadopoulos : > [My apologies for insisting on this issue.] >> There's (at least) two solutions to this problem. One is that, as you >> suggest, to have the iterator wrap its methods in transactions in >> addition to the code that returns the iterator (look at >> org.neo4j.

Re: [Neo] how to decouple transaction management from access to underlying nodes

2009-06-25 Thread Symeon (Akis) Papadopoulos
[My apologies for insisting on this issue.] > There's (at least) two solutions to this problem. One is that, as you > suggest, to have the iterator wrap its methods in transactions in > addition to the code that returns the iterator (look at > org.neo4j.util.TxIterator in neo-utils component). This

Re: [Neo] how to decouple transaction management from access to underlying nodes

2009-06-25 Thread Tobias Ivarsson
On Thu, Jun 25, 2009 at 10:31 AM, Symeon (Akis) Papadopoulos wrote: > > Currently, according to the Design Guide of Neo4j, the Model is defined > in a Neo4j-agnostic way and only the implementations of the defined > interfaces (e.g. CustomerImpl, OrderImpl) are Neo4j-dependent. In case I > transfe

Re: [Neo] how to decouple transaction management from access to underlying nodes

2009-06-25 Thread Symeon (Akis) Papadopoulos
Thank you for your prompt answer. See one further question regarding your second suggestion. > Another (and probably better and more logical) is to manage your > transactions at a slightly higher level. One example is an MVC model > where it'd be a good idea to put your transaction handling in the

Re: [Neo] how to decouple transaction management from access to underlying nodes

2009-06-24 Thread Mattias Persson
Hi! There's (at least) two solutions to this problem. One is that, as you suggest, to have the iterator wrap its methods in transactions in addition to the code that returns the iterator (look at org.neo4j.util.TxIterator in neo-utils component). This solution can be tedious and perhaps not the mo

[Neo] how to decouple transaction management from access to underlying nodes

2009-06-24 Thread Symeon (Akis) Papadopoulos
Hi all I am a newbie to Neo4j, but as it seems very pertinent to my research interests (web 2.0 mining), therefore I am planning to test it as an infrastructure layer for massive graph analysis. One very basic (even naive) question I have regards the separation between transaction management a