apoc.periodic.iterate with apoc.export.csv.data

[pankaj_malviya]

Hi there,

             I am using this query to export particular Node data to csv using apoc.periodic.iterate. But getting error.

Please look at the query and do suggest.

MATCH (nod:${node}) WITH collect(distinct nod) as a

CALL apoc.periodic.iterate(

'CALL apoc.export.csv.data(a,[],"person.csv", {})

RETURN distinct testFile, nodes, relationships, properties, data'

,{batchSize:10000, parallel:true,params:{nod:a}}) YIELD batches RETURN batches`)

[glilienfield]

What is the error?

In the meantime, a couple of observations:

• I am not familiar with this notation MATCH (nod:${node}). I highlited the confusing part. What are you trying to achieve here?
• apoc.periodic.iterate requires two cypher statements, one to get the data, and one to operate one the data. I only see one cypher statement in your query. 
• you can use ‘CALL {} IN TRANSACTIONS’ instead. See https://neo4j.com/docs/cypher-manual/current/clauses/call-subquery/#subquery-call-in-transactions

[pankaj_malviya]

CALL apoc.periodic.iterate(

'MATCH (nod:${node}) WITH collect(distinct nod) as a RETURN a',

'CALL apoc.export.csv.data(a,[],"person.csv", {})

YIELD file, nodes, relationships, properties, data

RETURN distinct file, nodes, relationships, properties, data'

,{batchSize:100}) YIELD batch,operations RETURN batch,operations`);

Now Query Get Executed but there is no batch work.

[ThomasVarias]

Just thinking out loud here... how is a multi-threaded export going to work? Halving multiple threads writing to the same file sounds like a bad idea to me.  Please keep in mind that I haven't tried this at all, ymmv.

Thomas

[glilienfield]

I agree with @ThomasVarias, this doesn’t seem to make sense. You did remove the parallel execution, but now it will probably just overwrite the existing export and create a new one. You probably would need to append a unique identifier to the end of the file name so a new file is created each time. Are you intending to generate the export across multiple files? Have you tried exporting the data directly without the periodic iterate? 

[pankaj_malviya]

Yes i have tried it works well. I need to migrate data from neo4j to mongoDB. And it is one time process. Is it the right way i am doing saving data in CSV file and then do the further process or something else i can try to migrate data.

[pankaj_malviya]

Neo4j Data is near about 10 GB, is it the right way by using apoc procedure