From 06f5dc841517e7156f5f445655d97ba541ebbd7e Mon Sep 17 00:00:00 2001
From: Nattavut Sutyanyong <nsy.can@gmail.com>
Date: Mon, 8 Aug 2016 12:14:11 +0200
Subject: [SPARK-16804][SQL] Correlated subqueries containing non-deterministic
 operations return incorrect results

## What changes were proposed in this pull request?

This patch fixes the incorrect results in the rule ResolveSubquery in Catalyst's Analysis phase by returning an error message when the LIMIT is found in the path from the parent table to the correlated predicate in the subquery.

## How was this patch tested?

./dev/run-tests
a new unit test on the problematic pattern.

Author: Nattavut Sutyanyong <nsy.can@gmail.com>

Closes #14411 from nsyca/master.
---
 .../spark/sql/catalyst/analysis/Analyzer.scala     | 13 ++++++++++
 .../sql/catalyst/analysis/AnalysisErrorSuite.scala | 17 +++++++++++++
 .../scala/org/apache/spark/sql/SubquerySuite.scala | 29 ++++++++++++++++++++++
 3 files changed, 59 insertions(+)

diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
index 660f523698..25202b521a 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
@@ -1021,6 +1021,19 @@ class Analyzer(
         case e: Expand =>
           failOnOuterReferenceInSubTree(e, "an EXPAND")
           e
+        case l : LocalLimit =>
+          failOnOuterReferenceInSubTree(l, "a LIMIT")
+          l
+        // Since LIMIT <n> is represented as GlobalLimit(<n>, (LocalLimit (<n>, child))
+        // and we are walking bottom up, we will fail on LocalLimit before
+        // reaching GlobalLimit.
+        // The code below is just a safety net.
+        case g : GlobalLimit =>
+          failOnOuterReferenceInSubTree(g, "a LIMIT")
+          g
+        case s : Sample =>
+          failOnOuterReferenceInSubTree(s, "a TABLESAMPLE")
+          s
         case p =>
           failOnOuterReference(p)
           p
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/analysis/AnalysisErrorSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/analysis/AnalysisErrorSuite.scala
index 8363a1b1cd..13bf034f83 100644
--- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/analysis/AnalysisErrorSuite.scala
+++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/analysis/AnalysisErrorSuite.scala
@@ -548,5 +548,22 @@ class AnalysisErrorSuite extends AnalysisTest {
       Exists(Union(LocalRelation(b), Filter(EqualTo(OuterReference(a), c), LocalRelation(c)))),
       LocalRelation(a))
     assertAnalysisError(plan3, "Accessing outer query column is not allowed in" :: Nil)
+
+    val plan4 = Filter(
+      Exists(
+        Limit(1,
+          Filter(EqualTo(OuterReference(a), b), LocalRelation(b)))
+      ),
+      LocalRelation(a))
+    assertAnalysisError(plan4, "Accessing outer query column is not allowed in a LIMIT" :: Nil)
+
+    val plan5 = Filter(
+      Exists(
+        Sample(0.0, 0.5, false, 1L,
+          Filter(EqualTo(OuterReference(a), b), LocalRelation(b)))().select('b)
+      ),
+      LocalRelation(a))
+    assertAnalysisError(plan5,
+                        "Accessing outer query column is not allowed in a TABLESAMPLE" :: Nil)
   }
 }
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/SubquerySuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/SubquerySuite.scala
index afed342ff8..52387b4b72 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/SubquerySuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/SubquerySuite.scala
@@ -571,4 +571,33 @@ class SubquerySuite extends QueryTest with SharedSQLContext {
       Row(1.0, false) :: Row(1.0, false) :: Row(2.0, true) :: Row(2.0, true) ::
         Row(3.0, false) :: Row(5.0, true) :: Row(null, false) :: Row(null, true) :: Nil)
   }
+
+  test("SPARK-16804: Correlated subqueries containing LIMIT - 1") {
+    withTempView("onerow") {
+      Seq(1).toDF("c1").createOrReplaceTempView("onerow")
+
+      checkAnswer(
+        sql(
+          """
+            | select c1 from onerow t1
+            | where exists (select 1 from onerow t2 where t1.c1=t2.c1)
+            | and   exists (select 1 from onerow LIMIT 1)""".stripMargin),
+        Row(1) :: Nil)
+     }
+   }
+
+  test("SPARK-16804: Correlated subqueries containing LIMIT - 2") {
+    withTempView("onerow") {
+      Seq(1).toDF("c1").createOrReplaceTempView("onerow")
+
+      checkAnswer(
+        sql(
+          """
+            | select c1 from onerow t1
+            | where exists (select 1
+            |               from   (select 1 from onerow t2 LIMIT 1)
+            |               where  t1.c1=t2.c1)""".stripMargin),
+        Row(1) :: Nil)
+     }
+   }
 }
-- 
cgit v1.2.3