fix comments

linhongliu-db · linhongliu-db · commit 4df4f7c8ebc8 · 2020-09-09T10:37:08.000+08:00
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala
@@ -195,6 +195,8 @@ abstract class Optimizer(catalogManager: CatalogManager)
       EliminateSorts) :+
     Batch("Decimal Optimizations", fixedPoint,
       DecimalAggregates) :+
+    // This batch must run after "Decimal Optimizations", as that one may change the
+    // aggregate distinct column
     Batch("Distinct Aggregate Rewrite", Once,
       RewriteDistinctAggregates) :+
     Batch("Object Expressions Optimization", fixedPoint,
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DataFrameSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DataFrameSuite.scala
@@ -2557,16 +2557,12 @@ class DataFrameSuite extends QueryTest
   }
 
   test("SPARK-32816: aggregating multiple distinct DECIMAL columns") {
-    withTempPath { path =>
-      spark.range(0, 100, 1, 1)
-        .selectExpr("id", "cast(id as decimal(9, 0)) as decimal_col")
-        .write.mode("overwrite")
-        .parquet(path.getAbsolutePath)
-      spark.read.parquet(path.getAbsolutePath).createOrReplaceTempView("test_table")
-      checkAnswer(
-        sql("select avg(distinct decimal_col), sum(distinct decimal_col) from test_table"),
-        Row(49.5, 4950))
-    }
+    spark.range(0, 100, 1, 1)
+      .selectExpr("id", "cast(id as decimal(9, 0)) as decimal_col")
+      .createOrReplaceTempView("test_table")
+    checkAnswer(
+      sql("select avg(distinct decimal_col), sum(distinct decimal_col) from test_table"),
+      Row(49.5, 4950))
   }
 }
 
