It's a bit tricky.

I think the fastest way to achieve that is to use the neo4j batch-inserter 
APIs, (which you might want to use for this import of your store anyway as it 
is not that large), I think you can even use my github.com/jexp/batch-import 
<http://github.com/jexp/batch-import> tool for populating the auto-index.

If you want to keep your store and just add the auto-index entries:

- open the store, 
- create a lucene-batch-inserter-impl configured for fulltext and named 
node_auto_index
- walk over all node-id's retrieve the property and add it to the index.

Michael

> Am 02.04.2015 um 19:20 schrieb Jean Reptile <[email protected]>:
> 
> Hello,
> 
> For a project, I need to import 5 million nodes and 15 millions relations.
> I tried to import by batch but it was very slow, so I used the new tool 
> 'Neo4j-import' from Neo4j 2.2. We generate some specifics .csv and use the 
> 'neo4j-import'. It is very fast, the whole database is created in 1mn30.
> But the problem is that I need to do a regex query on one property (find a 
> movie with only the beginning of his name). And the average response time is 
> between 2.5 and  4 seconds, which is huge.
> 
> I read that with Lucene query it would be much more efficient. But with 
> Neo4-import, nodes are created without the node_auto_indexing.
> 
> Is there a way to use Neo4j-import and have node_auto_indexing in order to 
> use the Lucene query ?
> 
> Thanks a lot
> 
> -- 
> 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 [email protected] 
> <mailto:[email protected]>.
> For more options, visit https://groups.google.com/d/optout 
> <https://groups.google.com/d/optout>.

-- 
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 [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to