Builder API: Loading DataFrame containing more than just IDs about neo4j-spark-connector

[jexp]

Sure there is a loadDataFrame method that takes a schema:

scala> import org.neo4j.spark._
import org.neo4j.spark._

scala> val neo = Neo4j(sc)
neo: org.neo4j.spark.Neo4j = org.neo4j.spark.Neo4j@750cd36d

scala> val df = neo.cypher("match (n) return id(n) as id1, id(n) as id2").partitions(4).batch(100).loadDataFrame(("id1"->"int"),("id2"->"int"))
df: org.apache.spark.sql.DataFrame = [id1: bigint, id2: bigint]

scala> df.take(3)
res2: Array[org.apache.spark.sql.Row] = Array([0,0], [1,1], [2,2])

[jexp]

But it also work with just the query (it inspects the result metadata then)

val df = neo.cypher("match (n) return id(n) as id1, id(n) as id2").partitions(4).batch(100).loadDataFrame
df: org.apache.spark.sql.DataFrame = [id1: bigint, id2: bigint]

[sjishan]

Thanks. Solved.

[anks2024]

@jexp

scala> val df = neo.cypher("match (n:Emp) return id(n) as id, n.name as name, n.dept as dept").loadDataFrame(("id"->"long"),("name"->"string"),("dept"->"long"))
df: org.apache.spark.sql.DataFrame = [id: bigint, name: string ... 1 more field]

scala> df.count()
res15: Long = 28 // When I have only 7 nodes in my Graph. Any specific reason for this repetitions?

[jexp]

@asarraf What happens if you run the same query as Cypher in Neo4j?