Skip to content

[WIP][159] Fix non-deterministic ID assignment #192

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
phi-dbq wants to merge 3 commits into from
Closed

[WIP][159] Fix non-deterministic ID assignment #192

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
2c89c6d
try a possible fix
mengxr May 4, 2017
398d8c6
update the code that leads to a more optimized plan
mengxr May 8, 2017
f659d12
remove `zipWithUniqueId` helper functions
phi-dbq May 10, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 10 additions & 4 deletions src/main/scala/org/graphframes/GraphFrame.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,9 +22,8 @@ import java.util.Random
import scala.reflect.runtime.universe.TypeTag

import org.apache.spark.graphx.{Edge, Graph}
import org.apache.spark.sql.SQLHelpers._
import org.apache.spark.sql._
import org.apache.spark.sql.functions.{array, broadcast, col, count, explode, struct, udf}
import org.apache.spark.sql.functions.{array, broadcast, col, count, explode, struct, udf, monotonically_increasing_id}
import org.apache.spark.sql.types._
import org.apache.spark.storage.StorageLevel

Expand Down Expand Up @@ -466,8 +465,15 @@ class GraphFrame private(
indexedVertices.select(
col(ATTR + "." + ID).cast("long").as(LONG_ID), col(ATTR + "." + ID).as(ID), col(ATTR))
} else {
val indexedVertices = zipWithUniqueId(vertices)
indexedVertices.select(col("uniq_id").as(LONG_ID), col("row." + ID).as(ID), col("row").as(ATTR))
val withLongIds = vertices.select(ID)
.repartition(col(ID))
.distinct()
.sortWithinPartitions(ID)
.withColumn(LONG_ID, monotonically_increasing_id())
.persist(StorageLevel.MEMORY_AND_DISK)
vertices.select(col(ID), nestAsCol(vertices, ATTR))
.join(withLongIds, ID)
.select(LONG_ID, ID, ATTR)
}
}

Expand Down
15 changes: 0 additions & 15 deletions src/main/spark-1.x/org/apache/spark/sql/SQLHelpers.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,21 +26,6 @@ object SQLHelpers {

def expr(e: String): Column = functions.expr(e)

/**
* Appends each record with a unique ID (uniq_id) and groups existing fields under column "row".
* This is a workaround for SPARK-9020 and SPARK-13473.
*/
def zipWithUniqueId(df: DataFrame): DataFrame = {
val sqlContext = df.sqlContext
val schema = df.schema
val rdd = df.rdd.zipWithUniqueId().map { case (row, id) =>
Row(row, id)
}
val outputSchema = StructType(Seq(
StructField("row", schema, false), StructField("uniq_id", LongType, false)))
sqlContext.createDataFrame(rdd, outputSchema)
}

def callUDF(f: Function1[_, _], returnType: DataType, arg1: Column): Column = {
org.apache.spark.sql.functions.callUDF(f, returnType, arg1)
}
Expand Down
15 changes: 0 additions & 15 deletions src/main/spark-2.x/org/apache/spark/sql/SQLHelpers.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,21 +26,6 @@ object SQLHelpers {

def expr(e: String): Column = functions.expr(e)

/**
* Appends each record with a unique ID (uniq_id) and groups existing fields under column "row".
* This is a workaround for SPARK-9020 and SPARK-13473.
*/
def zipWithUniqueId(df: DataFrame): DataFrame = {
val sqlContext = df.sqlContext
val schema = df.schema
val rdd = df.rdd.zipWithUniqueId().map { case (row, id) =>
Row(row, id)
}
val outputSchema = StructType(Seq(
StructField("row", schema, false), StructField("uniq_id", LongType, false)))
sqlContext.createDataFrame(rdd, outputSchema)
}

def callUDF(f: Function1[_, _], returnType: DataType, arg1: Column): Column = {
val u = udf(f, returnType)
u(arg1)
Expand Down