Merge pull request #315 from ldbc/fix-nondeterminism

szarnyasg · web-flow · commit 58a363806c1d · 2021-07-19T20:12:47.000+02:00
Sort collection of Persons before assigning them to blocks. Fixes #306
diff --git a/src/main/scala/ldbc/snb/datagen/generation/serializer/SparkActivitySerializer.scala b/src/main/scala/ldbc/snb/datagen/generation/serializer/SparkActivitySerializer.scala
@@ -15,6 +15,7 @@ import org.apache.spark.sql.SparkSession
 
 import java.nio.charset.StandardCharsets
 import java.util
+import java.util.Collections
 import java.util.function.Consumer
 import scala.collection.JavaConverters._
 
@@ -68,6 +69,7 @@ object SparkActivitySerializer {
             }
             friends.write(strbuf.toString().getBytes(StandardCharsets.UTF_8))
           }
+          Collections.sort(clonedPersons)
 
           val activities = generator.generateActivityForBlock(blockId.toInt, clonedPersons)
 

Original file line number	Diff line number	Diff line change
`@@ -15,6 +15,7 @@ import org.apache.spark.sql.SparkSession`
`15`	`15`
`16`	`16`	`import java.nio.charset.StandardCharsets`
`17`	`17`	`import java.util`
	`18`	`+import java.util.Collections`
`18`	`19`	`import java.util.function.Consumer`
`19`	`20`	`import scala.collection.JavaConverters._`
`20`	`21`
`@@ -68,6 +69,7 @@ object SparkActivitySerializer {`
`68`	`69`	`}`
`69`	`70`	`friends.write(strbuf.toString().getBytes(StandardCharsets.UTF_8))`
`70`	`71`	`}`
	`72`	`+ Collections.sort(clonedPersons)`
`71`	`73`
`72`	`74`	`val activities = generator.generateActivityForBlock(blockId.toInt, clonedPersons)`
`73`	`75`