throw exception when check equality between external and internal row

Author: cloud-fan

sql/catalyst/src/main/scala/org/apache/spark/sql/Row.scala

@@ -364,16 +364,29 @@ trait Row extends Serializable {
     false
   }
 
+  /**
+   * Returns true if we can check equality for these 2 rows.
+   * Equality check between external row and internal row is not allowed.
+   * Here we do this check to prevent call `equals` on external row with internal row.
+   */
   protected def canEqual(other: Any) = {
-    // Note that InternalRow overrides canEqual. These two canEqual's together makes sure that
-    // comparing the external Row and InternalRow will always yield false.
+    // Note that `Row` is not only the interface of external row but also the parent
+    // of `InternalRow`, so we have to ensure `other` is not a internal row here to prevent
+    // call `equals` on external row with internal row.
+    // `InternalRow` overrides canEqual, and these two canEquals together makes sure that
+    // equality check between external Row and InternalRow will always fail.
     // In the future, InternalRow should not extend Row. In that case, we can remove these
     // canEqual methods.
     other.isInstanceOf[Row] && !other.isInstanceOf[InternalRow]
   }
 
   override def equals(o: Any): Boolean = {
-    if (o == null || !canEqual(o)) return false
+    if (!canEqual(o)) {
+      throw new UnsupportedOperationException(
+        "cannot check equality between external and internal rows")
+    }
+
+    if (o == null) return false
 
     val other = o.asInstanceOf[Row]
     if (length != other.length) {

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/InternalRow.scala

@@ -54,6 +54,11 @@ abstract class InternalRow extends Row {
   // A default implementation to change the return type
   override def copy(): InternalRow = this
 
+  /**
+   * Returns true if we can check equality for these 2 rows.
+   * Equality check between external row and internal row is not allowed.
+   * Here we do this check to prevent call `equals` on internal row with external row.
+   */
   protected override def canEqual(other: Any) = other.isInstanceOf[InternalRow]
 
   // Custom hashCode function that matches the efficient code generated version.

sql/catalyst/src/test/scala/org/apache/spark/sql/RowTest.scala

@@ -17,6 +17,7 @@
 
 package org.apache.spark.sql
 
+import org.apache.spark.sql.catalyst.InternalRow
 import org.apache.spark.sql.catalyst.expressions.{GenericRow, GenericRowWithSchema}
 import org.apache.spark.sql.types._
 import org.scalatest.{Matchers, FunSpec}
@@ -68,4 +69,29 @@ class RowTest extends FunSpec with Matchers {
       sampleRow.getValuesMap(List("col1", "col2")) shouldBe expected
     }
   }
+
+  describe("row equals") {
+    val externalRow = Row(1, 2)
+    val externalRow2 = Row(1, 2)
+    val internalRow = InternalRow(1, 2)
+    val internalRow2 = InternalRow(1, 2)
+
+    it("equality check for external rows") {
+      externalRow shouldEqual externalRow2
+    }
+
+    it("equality check for internal rows") {
+      internalRow shouldEqual internalRow2
+    }
+
+    it("throws an exception when check equality between external and internal rows") {
+      def assertError(f: => Unit): Unit = {
+        val e = intercept[UnsupportedOperationException](f)
+        e.getMessage.contains("cannot check equality between external and internal rows")
+      }
+
+      assertError(internalRow.equals(externalRow))
+      assertError(externalRow.equals(internalRow))
+    }
+  }
 }