From 250da2805fbd9ec25833d954ecaa224286563a83 Mon Sep 17 00:00:00 2001
From: Gabor Szarnyas <szarnyasg@gmail.com>
Date: Fri, 16 Sep 2022 14:27:50 +0000
Subject: [PATCH] Factorgen: Fix sampling of messageIds table

---
 .../ldbc/snb/datagen/factors/FactorGenerationStage.scala  | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/src/main/scala/ldbc/snb/datagen/factors/FactorGenerationStage.scala b/src/main/scala/ldbc/snb/datagen/factors/FactorGenerationStage.scala
index 281d5cd5d..ac3fcd597 100644
--- a/src/main/scala/ldbc/snb/datagen/factors/FactorGenerationStage.scala
+++ b/src/main/scala/ldbc/snb/datagen/factors/FactorGenerationStage.scala
@@ -173,10 +173,10 @@ object FactorGenerationStage extends DatagenStage with Logging {
           date_trunc("day", $"creationDate").as("creationDay"),
           date_trunc("day", $"deletionDate").as("deletionDay"),
           $"MessageId")
-      val sampleSize = 20000
-        val count = messages.count()
-        val sampleFraction = Math.min(sampleSize / count, 1.0)
-        messages.sample(sampleFraction, 42)
+      val sampleSize = 20000.0
+      val count = messages.count()
+      val sampleFraction = Math.min(sampleSize / count, 1.0)
+      messages.sample(sampleFraction, 42)
     },
     "countryNumPersons" -> Factor(PlaceType, PersonType) { case Seq(places, persons) =>
       val cities    = places.where($"type" === "City").cache()