You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@carbondata.apache.org by GitBox <gi...@apache.org> on 2019/02/18 07:10:48 UTC

[GitHub] akashrn5 commented on a change in pull request #3128: [CARBONDATA-3294] Fix that MV datamap throw error when using count(1) and case when expression

akashrn5 commented on a change in pull request #3128: [CARBONDATA-3294] Fix that MV datamap throw error when using count(1) and case when expression
URL: https://github.com/apache/carbondata/pull/3128#discussion_r257576751
 
 

 ##########
 File path: datamap/mv/core/src/test/scala/org/apache/carbondata/mv/rewrite/MVCountAndCaseTestCase.scala
 ##########
 @@ -0,0 +1,97 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.carbondata.mv.rewrite
+
+import org.apache.spark.sql.catalyst.plans.logical.LogicalPlan
+import org.apache.spark.sql.execution.datasources.LogicalRelation
+import org.apache.spark.sql.test.util.QueryTest
+import org.scalatest.BeforeAndAfterAll
+
+class MVCountAndCaseTestCase  extends QueryTest with BeforeAndAfterAll{
+
+
+  override def beforeAll(): Unit = {
+    drop
+    sql("create table region(l4id string,l4name string) using carbondata")
+    sql(
+      s"""create table data_table(
+         |starttime int, seq long,succ long,LAYER4ID string,tmp int)
+         |using carbondata""".stripMargin)
+  }
+
+  def drop(): Unit ={
+    sql("drop table if exists region")
+    sql("drop table if exists data_table")
+  }
+
+  test("test mv count and case when expression") {
+    sql("drop datamap if exists data_table_mv")
+    sql(s"""create datamap data_table_mv using 'mv' as
+           | SELECT STARTTIME,LAYER4ID,
+           | SUM(seq) AS seq_c,
+           | SUM(succ) AS succ_c
+           | FROM data_table
+           | GROUP BY STARTTIME,LAYER4ID""".stripMargin)
+
+    sql("rebuild datamap data_table_mv")
+
+    var frame = sql(s"""SELECT  MT.`3600` AS `3600`,
+                       | MT.`2250410101` AS `2250410101`,
+                       | count(1) over() as countNum,
+                       | (CASE WHEN (SUM(COALESCE(seq_c, 0))) = 0 THEN NULL
+                       |   ELSE
+                       |   (CASE WHEN (CAST((SUM(COALESCE(seq_c, 0))) AS int)) = 0 THEN 0
+                       |     ELSE ((CAST((SUM(COALESCE(succ_c, 0))) AS double))
+                       |     / (CAST((SUM(COALESCE(seq_c, 0))) AS double)))
+                       |     END) * 100
+                       |   END) AS rate
+                       | FROM (
+                       |   SELECT sum_result.*, H_REGION.`2250410101` FROM
+                       |   (SELECT cast(floor((starttime + 28800) / 3600) * 3600 - 28800 as int) AS `3600`,
+                       |     LAYER4ID,
+                       |     COALESCE(SUM(seq), 0) AS seq_c,
+                       |     COALESCE(SUM(succ), 0) AS succ_c
+                       |       FROM data_table
+                       |       WHERE STARTTIME >= 1549866600 AND STARTTIME < 1549899900
+                       |       GROUP BY cast(floor((STARTTIME + 28800) / 3600) * 3600 - 28800 as int),LAYER4ID
+                       |   )sum_result
+                       |   LEFT JOIN
+                       |   (SELECT l4id AS `225040101`,
+                       |     l4name AS `2250410101`,
+                       |     l4name AS NAME_2250410101
+                       |       FROM region
+                       |       GROUP BY l4id, l4name) H_REGION
+                       |   ON sum_result.LAYER4ID = H_REGION.`225040101`
+                       | WHERE H_REGION.NAME_2250410101 IS NOT NULL
+                       | ) MT
+                       | GROUP BY MT.`3600`, MT.`2250410101`
+                       | ORDER BY `3600` ASC LIMIT 5000""".stripMargin)
+
+    assert(verifyMVDataMap(frame.queryExecution.analyzed, "data_table_mv"))
+  }
+
+  def verifyMVDataMap(logicalPlan: LogicalPlan, dataMapName: String): Boolean = {
 
 Review comment:
   `verifyMVDataMap` this method is used in many places, all test cases of MV, so better move ti to some utility class, or some Object and use the same everywhere. Refactor in all the places whoever is using this method.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services