This is what you would use indexing for, for storing starting points that
you can then use for traversing the graph.
<updated-example>
GraphDatabaseService graphDb = new EmbeddedGraphDatabase("poc_test");
// ADDED: add an index service to your system
IndexService index = new LuceneIndexService( graphDb );
Transaction tx = graphDb.beginTx();
try {
for(CSVObject object : csv){
Node trigramme = graphDb.createNode();
Node cp = graphDb.createNode();
Node description = graphDb.createNode();
Node category = graphDb.createNode();
Relationship ispartOf =
trigramme.createRelationshipTo(cp, RelTypes.IS_PART_OF);
Relationship isDescribedBy
= trigramme.createRelationshipTo(description, RelTypes.IS_DESCRIBED_BY);
Relationship isCategorizedIn
= description.createRelationshipTo(category, RelTypes.IS_CATEGORIZED_IN);
trigramme.setProperty("value", object.getTrigram());
cp.setProperty("value", object.getCp());
description.setProperty("value", object.getDescription());
category.setProperty("value", object.getCategory());
ispartOf.setProperty("relationship", "ispartOf ");
isDescribedBy.setProperty("relationship", "ispartOf");
isCategorizedIn.setProperty("relationship", "ispartOf");
// ADDED: add the code for indexing your start nodes
index.index( description, "description",
description.getProperty("value") );
}
tx.success();
}catch (Exception e){
e.printStackTrace();
}
finally {
tx.finish();
}
public void retreiveStuff( Object descriptionValue ) {
// ADDED: retrieve your start node through the index before
traversal
Node description = index.getSingleNode( "description",
descriptionValue );
if ( description == null ) { throw
theDescriptionValueTheUserProvidedDoesNotExist(); }
Traverser friendsTraverser = description.traverse(
Traverser.Order.BREADTH_FIRST,
StopEvaluator.DEPTH_ONE,
ReturnableEvaluator.ALL_BUT_START_NODE,
RelTypes.IS_DESCRIBED_BY,
Direction.INCOMING );
// Traverse the node space and print out the result
//System.out.println( "Testing values : " );
for ( Node friend : friendsTraverser )
{
System.out.println( "At depth : "+
friendsTraverser.currentPosition().depth()+"
with value : "+
friend.getProperty( "value" ) );
}
}
</updated-example>
Cheers,
Tobias
On Tue, Jul 6, 2010 at 8:39 AM, Boban Erakovic <[email protected]>wrote:
> Hi,
> the reason of using neo4j is the main problem of my question. Just as Toni
> Menzel said "Directly falling back to the index service does not really
> embrace why you are a graph db at all." The question is, because my
> employee
> want to perform some project, within what he (me) will demonstrate the use
> of graph db. That is the problem, because there is no any particular
> problem, I am aiming to resolve, and for that and that I need graph db.
> However, one thing which bothering me is : Can I perform selection like in
> relational database (so I have saved some data some time ago, and I want to
> retrieve it again)?
> In all neo4j examples, data is retrieved from the same part of the code
> which perform saving earlier.
> Is it at all possible to select data, like 10 days after saving?
>
> Every example is like this :
>
> GraphDatabaseService graphDb = new
> EmbeddedGraphDatabase("poc_test");
>
> Transaction tx = graphDb.beginTx();
> try {
> for(CSVObject object : csv){
> Node trigramme = graphDb.createNode();
> Node cp = graphDb.createNode();
> Node description = graphDb.createNode();
> Node category = graphDb.createNode();
>
> Relationship ispartOf = trigramme.createRelationshipTo(cp,
> RelTypes.IS_PART_OF);
> Relationship isDescribedBy =
> trigramme.createRelationshipTo(description, RelTypes.IS_DESCRIBED_BY);
> Relationship isCategorizedIn =
> description.createRelationshipTo(category, RelTypes.IS_CATEGORIZED_IN);
>
> trigramme.setProperty("value", object.getTrigram());
> cp.setProperty("value", object.getCp());
> description.setProperty("value", object.getDescription());
> category.setProperty("value", object.getCategory());
>
> ispartOf.setProperty("relationship", "ispartOf ");
> isDescribedBy.setProperty("relationship", "ispartOf");
> isCategorizedIn.setProperty("relationship", "ispartOf");
>
> Traverser friendsTraverser = description.traverse(
> Traverser.Order.BREADTH_FIRST,
> StopEvaluator.DEPTH_ONE,
> ReturnableEvaluator.ALL_BUT_START_NODE,
> RelTypes.IS_DESCRIBED_BY,
> Direction.INCOMING );
> // Traverse the node space and print out the result
> //System.out.println( "Testing values : " );
> for ( Node friend : friendsTraverser )
> {
>
> System.out.println( "At depth : "+
> friendsTraverser.currentPosition().depth()+" with
> value : "+
> friend.getProperty( "value" ) );
> }
> }
>
> tx.success();
> }catch (Exception e){
> e.printStackTrace();
> }
> finally {
> tx.finish();
> graphDb.shutdown();
> }
> }
>
> So Traverser is using the node previously created, in order to iterate over
> database.
> But, how can I perform (simple example), to save user's data when he
> register, and after 10 days when he log in again, to retrieve that data?
> And
> is this kind of concept, which I use in relational db, possible to apply to
> graph database? I mean, if I cannot get saved data after some time, what is
> the purpose of the graph db?
> Hope you understood my problem and question, and thanks in advance to your
> answers !
> _______________________________________________
> Neo4j mailing list
> [email protected]
> https://lists.neo4j.org/mailman/listinfo/user
>
--
Tobias Ivarsson <[email protected]>
Hacker, Neo Technology
www.neotechnology.com
Cellphone: +46 706 534857
_______________________________________________
Neo4j mailing list
[email protected]
https://lists.neo4j.org/mailman/listinfo/user
