[SPARK-26352][SQL] join reorder should not change the order of output…

… attributes

## What changes were proposed in this pull request?

The optimizer rule `org.apache.spark.sql.catalyst.optimizer.ReorderJoin` performs join reordering on inner joins. This was introduced from SPARK-12032 (apache#10073) in 2015-12.

After it had reordered the joins, though, it didn't check whether or not the output attribute order is still the same as before. Thus, it's possible to have a mismatch between the reordered output attributes order vs the schema that a DataFrame thinks it has.
The same problem exists in the CBO version of join reordering (`CostBasedJoinReorder`) too.

This can be demonstrated with the example:
```scala
spark.sql("create table table_a (x int, y int) using parquet")
spark.sql("create table table_b (i int, j int) using parquet")
spark.sql("create table table_c (a int, b int) using parquet")
val df = spark.sql("""
  with df1 as (select * from table_a cross join table_b)
  select * from df1 join table_c on a = x and b = i
""")
```
here's what the DataFrame thinks:
```
scala> df.printSchema
root
 |-- x: integer (nullable = true)
 |-- y: integer (nullable = true)
 |-- i: integer (nullable = true)
 |-- j: integer (nullable = true)
 |-- a: integer (nullable = true)
 |-- b: integer (nullable = true)
```
here's what the optimized plan thinks, after join reordering:
```
scala> df.queryExecution.optimizedPlan.output.foreach(a => println(s"|-- ${a.name}: ${a.dataType.typeName}"))
|-- x: integer
|-- y: integer
|-- a: integer
|-- b: integer
|-- i: integer
|-- j: integer
```

If we exclude the `ReorderJoin` rule (using Spark 2.4's optimizer rule exclusion feature), it's back to normal:
```
scala> spark.conf.set("spark.sql.optimizer.excludedRules", "org.apache.spark.sql.catalyst.optimizer.ReorderJoin")

scala> val df = spark.sql("with df1 as (select * from table_a cross join table_b) select * from df1 join table_c on a = x and b = i")
df: org.apache.spark.sql.DataFrame = [x: int, y: int ... 4 more fields]

scala> df.queryExecution.optimizedPlan.output.foreach(a => println(s"|-- ${a.name}: ${a.dataType.typeName}"))
|-- x: integer
|-- y: integer
|-- i: integer
|-- j: integer
|-- a: integer
|-- b: integer
```

Note that this output attribute ordering problem leads to data corruption, and can manifest itself in various symptoms:
* Silently corrupting data, if the reordered columns happen to either have matching types or have sufficiently-compatible types (e.g. all fixed length primitive types are considered as "sufficiently compatible" in an `UnsafeRow`), then only the resulting data is going to be wrong but it might not trigger any alarms immediately. Or
* Weird Java-level exceptions like `java.lang.NegativeArraySizeException`, or even SIGSEGVs.

## How was this patch tested?

Added new unit test in `JoinReorderSuite` and new end-to-end test in `JoinSuite`.
Also made `JoinReorderSuite` and `StarJoinReorderSuite` assert more strongly on maintaining output attribute order.

Closes apache#23303 from rednaxelafx/fix-join-reorder.

Authored-by: Kris Mok <[email protected]>
Signed-off-by: Wenchen Fan <[email protected]>

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/CostBasedJoinReorder.scala

@@ -48,6 +48,7 @@ object CostBasedJoinReorder extends Rule[LogicalPlan] with PredicateHelper {
           if projectList.forall(_.isInstanceOf[Attribute]) =>
           reorder(p, p.output)
       }
+
       // After reordering is finished, convert OrderedJoin back to Join
       result transformDown {
         case OrderedJoin(left, right, jt, cond) => Join(left, right, jt, cond)
@@ -175,11 +176,20 @@ object JoinReorderDP extends PredicateHelper with Logging {
         assert(topOutputSet == p.outputSet)
         // Keep the same order of final output attributes.
         p.copy(projectList = output)
+      case finalPlan if !sameOutput(finalPlan, output) =>
+        Project(output, finalPlan)
       case finalPlan =>
         finalPlan
     }
   }
 
+  private def sameOutput(plan: LogicalPlan, expectedOutput: Seq[Attribute]): Boolean = {
+    val thisOutput = plan.output
+    thisOutput.length == expectedOutput.length && thisOutput.zip(expectedOutput).forall {
+      case (a1, a2) => a1.semanticEquals(a2)
+    }
+  }
+
   /** Find all possible plans at the next level, based on existing levels. */
   private def searchLevel(
       existingLevels: Seq[JoinPlanMap],

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/joins.scala

@@ -86,9 +86,9 @@ object ReorderJoin extends Rule[LogicalPlan] with PredicateHelper {
   }
 
   def apply(plan: LogicalPlan): LogicalPlan = plan transform {
-    case ExtractFiltersAndInnerJoins(input, conditions)
+    case p @ ExtractFiltersAndInnerJoins(input, conditions)
         if input.size > 2 && conditions.nonEmpty =>
-      if (SQLConf.get.starSchemaDetection && !SQLConf.get.cboEnabled) {
+      val reordered = if (SQLConf.get.starSchemaDetection && !SQLConf.get.cboEnabled) {
         val starJoinPlan = StarSchemaDetection.reorderStarJoins(input, conditions)
         if (starJoinPlan.nonEmpty) {
           val rest = input.filterNot(starJoinPlan.contains(_))
@@ -99,6 +99,14 @@ object ReorderJoin extends Rule[LogicalPlan] with PredicateHelper {
       } else {
         createOrderedJoin(input, conditions)
       }
+
+      if (p.sameOutput(reordered)) {
+        reordered
+      } else {
+        // Reordering the joins have changed the order of the columns.
+        // Inject a projection to make sure we restore to the expected ordering.
+        Project(p.output, reordered)
+      }
   }
 }
 

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/JoinOptimizationSuite.scala

@@ -102,16 +102,19 @@ class JoinOptimizationSuite extends PlanTest {
         x.join(y).join(z).where(("x.b".attr === "z.b".attr) && ("y.d".attr === "z.a".attr)),
         x.join(z, condition = Some("x.b".attr === "z.b".attr))
           .join(y, condition = Some("y.d".attr === "z.a".attr))
+          .select(Seq("x.a", "x.b", "x.c", "y.d", "z.a", "z.b", "z.c").map(_.attr): _*)
       ),
       (
         x.join(y, Cross).join(z, Cross)
           .where(("x.b".attr === "z.b".attr) && ("y.d".attr === "z.a".attr)),
         x.join(z, Cross, Some("x.b".attr === "z.b".attr))
           .join(y, Cross, Some("y.d".attr === "z.a".attr))
+          .select(Seq("x.a", "x.b", "x.c", "y.d", "z.a", "z.b", "z.c").map(_.attr): _*)
       ),
       (
         x.join(y, Inner).join(z, Cross).where("x.b".attr === "z.a".attr),
         x.join(z, Cross, Some("x.b".attr === "z.a".attr)).join(y, Inner)
+          .select(Seq("x.a", "x.b", "x.c", "y.d", "z.a", "z.b", "z.c").map(_.attr): _*)
       )
     )
 

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/JoinReorderSuite.scala

@@ -20,8 +20,8 @@ package org.apache.spark.sql.catalyst.optimizer
 import org.apache.spark.sql.catalyst.dsl.expressions._
 import org.apache.spark.sql.catalyst.dsl.plans._
 import org.apache.spark.sql.catalyst.expressions.{Attribute, AttributeMap}
-import org.apache.spark.sql.catalyst.plans.{Inner, PlanTest}
-import org.apache.spark.sql.catalyst.plans.logical.{ColumnStat, LogicalPlan}
+import org.apache.spark.sql.catalyst.plans.{Cross, Inner, PlanTest}
+import org.apache.spark.sql.catalyst.plans.logical.{ColumnStat, LocalRelation, LogicalPlan}
 import org.apache.spark.sql.catalyst.rules.RuleExecutor
 import org.apache.spark.sql.catalyst.statsEstimation.{StatsEstimationTestBase, StatsTestPlan}
 import org.apache.spark.sql.internal.SQLConf.{CBO_ENABLED, JOIN_REORDER_ENABLED}
@@ -124,7 +124,8 @@ class JoinReorderSuite extends PlanTest with StatsEstimationTestBase {
     // the original order (t1 J t2) J t3.
     val bestPlan =
       t1.join(t3, Inner, Some(nameToAttr("t1.v-1-10") === nameToAttr("t3.v-1-100")))
-      .join(t2, Inner, Some(nameToAttr("t1.k-1-2") === nameToAttr("t2.k-1-5")))
+        .join(t2, Inner, Some(nameToAttr("t1.k-1-2") === nameToAttr("t2.k-1-5")))
+        .select(outputsOf(t1, t2, t3): _*)
 
     assertEqualPlans(originalPlan, bestPlan)
   }
@@ -139,7 +140,9 @@ class JoinReorderSuite extends PlanTest with StatsEstimationTestBase {
     val bestPlan =
       t1.join(t3, Inner, Some(nameToAttr("t1.v-1-10") === nameToAttr("t3.v-1-100")))
         .join(t2, Inner, Some(nameToAttr("t1.k-1-2") === nameToAttr("t2.k-1-5")))
+        .select(outputsOf(t1, t2, t3): _*) // this is redundant but we'll take it for now
         .join(t4)
+        .select(outputsOf(t1, t2, t4, t3): _*)
 
     assertEqualPlans(originalPlan, bestPlan)
   }
@@ -202,6 +205,7 @@ class JoinReorderSuite extends PlanTest with StatsEstimationTestBase {
       t1.join(t2, Inner, Some(nameToAttr("t1.k-1-2") === nameToAttr("t2.k-1-5")))
         .join(t4.join(t3, Inner, Some(nameToAttr("t4.v-1-10") === nameToAttr("t3.v-1-100"))),
           Inner, Some(nameToAttr("t1.k-1-2") === nameToAttr("t4.k-1-2")))
+        .select(outputsOf(t1, t4, t2, t3): _*)
 
     assertEqualPlans(originalPlan, bestPlan)
   }
@@ -219,6 +223,23 @@ class JoinReorderSuite extends PlanTest with StatsEstimationTestBase {
     }
   }
 
+  test("SPARK-26352: join reordering should not change the order of attributes") {
+    // This test case does not rely on CBO.
+    // It's similar to the test case above, but catches a reordering bug that the one above doesn't
+    val tab1 = LocalRelation('x.int, 'y.int)
+    val tab2 = LocalRelation('i.int, 'j.int)
+    val tab3 = LocalRelation('a.int, 'b.int)
+    val original =
+      tab1.join(tab2, Cross)
+          .join(tab3, Inner, Some('a === 'x && 'b === 'i))
+    val expected =
+      tab1.join(tab3, Inner, Some('a === 'x))
+          .join(tab2, Cross, Some('b === 'i))
+          .select(outputsOf(tab1, tab2, tab3): _*)
+
+    assertEqualPlans(original, expected)
+  }
+
   test("reorder recursively") {
     // Original order:
     //          Join
@@ -266,8 +287,17 @@ class JoinReorderSuite extends PlanTest with StatsEstimationTestBase {
   private def assertEqualPlans(
       originalPlan: LogicalPlan,
       groundTruthBestPlan: LogicalPlan): Unit = {
-    val optimized = Optimize.execute(originalPlan.analyze)
+    val analyzed = originalPlan.analyze
+    val optimized = Optimize.execute(analyzed)
     val expected = groundTruthBestPlan.analyze
+
+    assert(analyzed.sameOutput(expected)) // if this fails, the expected plan itself is incorrect
+    assert(analyzed.sameOutput(optimized))
+
     compareJoinOrder(optimized, expected)
   }
+
+  private def outputsOf(plans: LogicalPlan*): Seq[Attribute] = {
+    plans.map(_.output).reduce(_ ++ _)
+  }
 }

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/StarJoinCostBasedReorderSuite.scala

@@ -218,6 +218,7 @@ class StarJoinCostBasedReorderSuite extends PlanTest with StatsEstimationTestBas
         .join(d1, Inner, Some(nameToAttr("f1_fk1") === nameToAttr("d1_pk")))
         .join(t2, Inner, Some(nameToAttr("f1_c2") === nameToAttr("t2_c1")))
         .join(t1, Inner, Some(nameToAttr("f1_c1") === nameToAttr("t1_c1")))
+        .select(outputsOf(f1, t1, t2, d1, d2): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -256,6 +257,7 @@ class StarJoinCostBasedReorderSuite extends PlanTest with StatsEstimationTestBas
         .join(t3.join(t2, Inner, Some(nameToAttr("t2_c2") === nameToAttr("t3_c1"))), Inner,
           Some(nameToAttr("d1_c2") === nameToAttr("t2_c1")))
         .join(t1, Inner, Some(nameToAttr("t1_c1") === nameToAttr("f1_c1")))
+        .select(outputsOf(d1, t1, t2, f1, d2, t3): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -297,6 +299,7 @@ class StarJoinCostBasedReorderSuite extends PlanTest with StatsEstimationTestBas
           Some(nameToAttr("t3_c1") === nameToAttr("t4_c1")))
         .join(t1.join(t2, Inner, Some(nameToAttr("t1_c1") === nameToAttr("t2_c1"))), Inner,
           Some(nameToAttr("t1_c2") === nameToAttr("t4_c2")))
+        .select(outputsOf(d1, t1, t2, t3, t4, f1, d2): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -347,6 +350,7 @@ class StarJoinCostBasedReorderSuite extends PlanTest with StatsEstimationTestBas
           Some(nameToAttr("d3_c2") === nameToAttr("t1_c1")))
         .join(t5.join(t6, Inner, Some(nameToAttr("t5_c2") === nameToAttr("t6_c2"))), Inner,
           Some(nameToAttr("d2_c2") === nameToAttr("t5_c1")))
+        .select(outputsOf(d1, t3, t4, f1, d2, t5, t6, d3, t1, t2): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -375,6 +379,7 @@ class StarJoinCostBasedReorderSuite extends PlanTest with StatsEstimationTestBas
       f1.join(d3, Inner, Some(nameToAttr("f1_fk3") === nameToAttr("d3_pk")))
         .join(d2, Inner, Some(nameToAttr("f1_fk2") === nameToAttr("d2_pk")))
         .join(d1, Inner, Some(nameToAttr("f1_fk1") === nameToAttr("d1_pk")))
+        .select(outputsOf(d1, d2, f1, d3): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -400,13 +405,27 @@ class StarJoinCostBasedReorderSuite extends PlanTest with StatsEstimationTestBas
       f1.join(t3, Inner, Some(nameToAttr("f1_fk3") === nameToAttr("t3_c1")))
         .join(t2, Inner, Some(nameToAttr("f1_fk2") === nameToAttr("t2_c1")))
         .join(t1, Inner, Some(nameToAttr("f1_fk1") === nameToAttr("t1_c1")))
+        .select(outputsOf(t1, f1, t2, t3): _*)
 
     assertEqualPlans(query, expected)
   }
 
   private def assertEqualPlans( plan1: LogicalPlan, plan2: LogicalPlan): Unit = {
-    val optimized = Optimize.execute(plan1.analyze)
+    val analyzed = plan1.analyze
+    val optimized = Optimize.execute(analyzed)
     val expected = plan2.analyze
+
+    assert(equivalentOutput(analyzed, expected)) // if this fails, the expected itself is incorrect
+    assert(equivalentOutput(analyzed, optimized))
+
     compareJoinOrder(optimized, expected)
   }
+
+  private def outputsOf(plans: LogicalPlan*): Seq[Attribute] = {
+    plans.map(_.output).reduce(_ ++ _)
+  }
+
+  private def equivalentOutput(plan1: LogicalPlan, plan2: LogicalPlan): Boolean = {
+    normalizeExprIds(plan1).output == normalizeExprIds(plan2).output
+  }
 }

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/StarJoinReorderSuite.scala

@@ -182,6 +182,7 @@ class StarJoinReorderSuite extends PlanTest with StatsEstimationTestBase {
         .join(d1, Inner, Some(nameToAttr("f1_fk1") === nameToAttr("d1_pk1")))
         .join(d3, Inner, Some(nameToAttr("f1_fk3") === nameToAttr("d3_pk1")))
         .join(s3, Inner, Some(nameToAttr("d3_fk1") === nameToAttr("s3_pk1")))
+        .select(outputsOf(d1, d2, f1, d3, s3): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -220,6 +221,7 @@ class StarJoinReorderSuite extends PlanTest with StatsEstimationTestBase {
         .join(d3, Inner, Some(nameToAttr("f1_fk3") === nameToAttr("d3_pk1")))
         .join(d2, Inner, Some(nameToAttr("f1_fk2") < nameToAttr("d2_pk1")))
         .join(s3, Inner, Some(nameToAttr("d3_fk1") === nameToAttr("s3_pk1")))
+        .select(outputsOf(d1, f1, d2, s3, d3): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -255,7 +257,7 @@ class StarJoinReorderSuite extends PlanTest with StatsEstimationTestBase {
         .join(d3, Inner, Some(nameToAttr("d3_fk1") === nameToAttr("s3_pk1")))
         .join(d2, Inner, Some(nameToAttr("f1_fk2") === nameToAttr("d2_pk1")))
         .join(s3, Inner, Some(nameToAttr("f1_fk3") === nameToAttr("s3_c2")))
-
+        .select(outputsOf(d1, f1, d2, s3, d3): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -292,6 +294,7 @@ class StarJoinReorderSuite extends PlanTest with StatsEstimationTestBase {
         .join(d3, Inner, Some(nameToAttr("f1_fk3") === nameToAttr("d3_pk1")))
         .join(d2, Inner, Some(nameToAttr("f1_fk2") === nameToAttr("d2_c2")))
         .join(s3, Inner, Some(nameToAttr("d3_fk1") < nameToAttr("s3_pk1")))
+        .select(outputsOf(d1, f1, d2, s3, d3): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -395,6 +398,7 @@ class StarJoinReorderSuite extends PlanTest with StatsEstimationTestBase {
         .join(d2.where(nameToAttr("d2_c2") === 2), Inner,
           Some(nameToAttr("f1_fk2") === nameToAttr("d2_pk1")))
         .join(s3, Inner, Some(nameToAttr("f11_fk1") === nameToAttr("s3_pk1")))
+        .select(outputsOf(d1, f11, f1, d2, s3): _*)
 
     assertEqualPlans(query, equivQuery)
   }
@@ -430,6 +434,7 @@ class StarJoinReorderSuite extends PlanTest with StatsEstimationTestBase {
         .join(d2.where(nameToAttr("d2_c2") === 2), Inner,
           Some(nameToAttr("f1_fk2") === nameToAttr("d2_c4")))
         .join(s3, Inner, Some(nameToAttr("d3_fk1") === nameToAttr("s3_pk1")))
+        .select(outputsOf(d1, d3, f1, d2, s3): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -465,6 +470,7 @@ class StarJoinReorderSuite extends PlanTest with StatsEstimationTestBase {
         .join(d2.where(nameToAttr("d2_c2") === 2), Inner,
           Some(nameToAttr("f1_fk2") === nameToAttr("d2_pk1")))
         .join(s3, Inner, Some(nameToAttr("d3_fk1") === nameToAttr("s3_pk1")))
+        .select(outputsOf(d1, d3, f1, d2, s3): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -499,6 +505,7 @@ class StarJoinReorderSuite extends PlanTest with StatsEstimationTestBase {
         .join(d2.where(nameToAttr("d2_c2") === 2),
           Inner, Some(nameToAttr("f1_fk2") === nameToAttr("d2_pk1")))
         .join(s3, Inner, Some(nameToAttr("d3_fk1") === nameToAttr("s3_pk1")))
+        .select(outputsOf(d1, d3, f1, d2, s3): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -532,6 +539,7 @@ class StarJoinReorderSuite extends PlanTest with StatsEstimationTestBase {
         .join(d3, Inner, Some(nameToAttr("f1_fk3") < nameToAttr("d3_pk1")))
         .join(d2, Inner, Some(nameToAttr("f1_fk2") < nameToAttr("d2_pk1")))
         .join(s3, Inner, Some(nameToAttr("d3_fk1") < nameToAttr("s3_pk1")))
+        .select(outputsOf(d1, d3, f1, d2, s3): _*)
 
     assertEqualPlans(query, expected)
   }
@@ -565,13 +573,27 @@ class StarJoinReorderSuite extends PlanTest with StatsEstimationTestBase {
         .join(d3, Inner, Some(nameToAttr("f1_fk3") === nameToAttr("d3_pk1")))
         .join(d2, Inner, Some(nameToAttr("f1_fk2") === nameToAttr("d2_pk1")))
         .join(s3, Inner, Some(nameToAttr("d3_fk1") === nameToAttr("s3_pk1")))
+        .select(outputsOf(d1, d3, f1, d2, s3): _*)
 
     assertEqualPlans(query, expected)
   }
 
-  private def assertEqualPlans( plan1: LogicalPlan, plan2: LogicalPlan): Unit = {
-    val optimized = Optimize.execute(plan1.analyze)
+  private def assertEqualPlans(plan1: LogicalPlan, plan2: LogicalPlan): Unit = {
+    val analyzed = plan1.analyze
+    val optimized = Optimize.execute(analyzed)
     val expected = plan2.analyze
+
+    assert(equivalentOutput(analyzed, expected)) // if this fails, the expected itself is incorrect
+    assert(equivalentOutput(analyzed, optimized))
+
     compareJoinOrder(optimized, expected)
   }
+
+  private def outputsOf(plans: LogicalPlan*): Seq[Attribute] = {
+    plans.map(_.output).reduce(_ ++ _)
+  }
+
+  private def equivalentOutput(plan1: LogicalPlan, plan2: LogicalPlan): Boolean = {
+    normalizeExprIds(plan1).output == normalizeExprIds(plan2).output
+  }
 }

sql/core/src/test/scala/org/apache/spark/sql/JoinSuite.scala

@@ -895,4 +895,18 @@ class JoinSuite extends QueryTest with SharedSQLContext {
       checkAnswer(res, Row(0, 0, 0))
     }
   }
+
+  test("SPARK-26352: join reordering should not change the order of columns") {
+    withTable("tab1", "tab2", "tab3") {
+      spark.sql("select 1 as x, 100 as y").write.saveAsTable("tab1")
+      spark.sql("select 42 as i, 200 as j").write.saveAsTable("tab2")
+      spark.sql("select 1 as a, 42 as b").write.saveAsTable("tab3")
+
+      val df = spark.sql("""
+        with tmp as (select * from tab1 cross join tab2)
+        select * from tmp join tab3 on a = x and b = i
+      """)
+      checkAnswer(df, Row(1, 100, 42, 200, 1, 42))
+    }
+  }
 }