Thank you Micheal I will check typos, in the example I posted ones just for example.
However, I do have indexes in my nodes:type (topic), node.properties: name and ID (id is integer) and set up node_auto_index But it takes AGES to locate a node, and ages to compute a simple query! I suspect the problem is in indexes, because quering by internal id is faster. Please, have a look here below, and could you tell me some benchmark to have an idea of how long should take a simple query like MATCH (n:topic)-[r:]-(m) where.name = 'TITLE' RETURN m ORDER BY r.weight DESC LIMIT 6 on a graph of 4M nodes and 100M rels, to understand if i am doing things right ? http://localhost:7474/webadmin/#/index/ topic lucene {"to_lower_case":"true", "type":"fulltext"} Deletename lucene {"to_lower_case":"true", "type":"fulltext"} Deleteid lucene {"type":"exact"} Deletenode_auto_index lucene {"type":"exact"} Delete Il giorno sabato 30 agosto 2014 01:01:27 UTC+2, Michael Hunger ha scritto: > > Did you create a schema index? > > Labels and property names are case sensitive, you use "USER" (without s) > in your batch import but "Users" (with s and different caps in cypher) > > create index on :Users(id); > > otherwise see for the difference between legacy and schema indexes > http://nigelsmall.com/neo4j/index-confusion > the batch-importer currently only supports legacy indexes > > Michael > > Am 14.08.2014 um 18:33 schrieb gg4u <[email protected] <javascript:>>: > > Hello folks! > > I've done a batch import of 4M nodes and 100M rels. > > I set up autoindexes before import, but something got wrong apparently: > it takes ages for querying a single node like: > match (n:Users)-[r:REL]-() > where n.id = 25 > > > Please help me to understand if I am doing the following steps correct, in > order to create proper indexing for my nodes and properties. > > *Goal* > > - I wanna query my nodes by their *name* property, so I am gonna > index :User(*name*) > - I want to *traverse* the graph: does also the node-id for each node > to be indexed :User(*id*)? Please note that node-id is a key for > indexing third part sources, so I cannot change it. > > ** > My data are as following: > #node.csv > *id:int:student MyLabel:label name:string:user* > 3212 USER Mark > 6367 USER Paula > ... > > (I set the headers by looking at the tutorial, but I am confused: > is *name:string:user * correct syntax to apply the property name to > each node with Label 'USER' ? > Why the keyword 'student' does not apply as a label in the > localhost:7474/browser? > Is it a label of nodes, or the name of the index?) > > > > #rels.csv > *id:int:student **id:int:student* *type property* > 3212 6367 LOVERS april > > *** > In *batch.properties* I specified: > > *batch_import.node_index.name > <http://batch_import.node_index.name>=fulltext* > *batch_import.node_index.student=exact* > *batch_import.node_index.node_auto_index=exact* > > > Once I uploaded, I cannot search neither for USER.id, neither for USER.name > > *** > In *neo4j.properties*, > I had *node_auto_indexing* and *node_keys_indexable* commented out, so i > batch imported without them. > > Is the indexing failed because of this? > > Do I have to redo the import with this settings? > > # Enable auto-indexing for nodes, default is false > node_auto_indexing=true > > # The node property keys to be auto-indexed, if enabled > node_keys_indexable=name,id > > > Any help for shedding light? > > -- > 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] <javascript:>. > For more options, visit 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.
