add chain rate

.idea/misc.xml

 <?xml version="1.0" encoding="UTF-8"?>
 <project version="4">
-  <component name="MavenProjectsManager">
-    <option name="originalFiles">
-      <list>
-        <option value="$PROJECT_DIR$/pom.xml" />
-      </list>
-    </option>
-  </component>
-  <component name="ProjectRootManager" version="2" languageLevel="JDK_1_8" project-jdk-name="1.8.0_212" project-jdk-type="JavaSDK">
+  <component name="ProjectRootManager" version="2" languageLevel="JDK_1_8" project-jdk-name="1.8" project-jdk-type="JavaSDK">
     <output url="file://$PROJECT_DIR$/classes" />
   </component>
 </project>
\ No newline at end of file

src/main/scala/com/gmei/data/dq/Main.scala

@@ -30,7 +30,8 @@ object Main {
     }
 
     // check pv
-    pvReferCheker.check(spark, partition_date)
+    // pvReferCheker.check(spark, partition_date)
+    actionCheck.check(spark, partition_date)
 
     spark.stop()
   }

src/main/scala/com/gmei/data/dq/actionCheck.scala

+package com.gmei.data.dq
+
+import java.text.SimpleDateFormat
+import java.util.Date
+
+import org.apache.spark.sql.SparkSession
+
+object actionCheck {
+
+  import java.util.Calendar
+
+  private def getYesterday(date: Date) = {
+    val cal = Calendar.getInstance
+    cal.setTime(date)
+    cal.add(Calendar.DATE, -1)
+    cal.getTime
+  }
+
+  def check(sc: SparkSession, partition_date: String) = {
+    import sc.implicits._
+
+    val dateFormat = new SimpleDateFormat("yyyyMMdd")
+    val time_date = dateFormat.parse(partition_date)
+    val yesterday = getYesterday(time_date)
+
+    val df = sc.sql(
+      s"""
+        |select
+        | a.partition_date as today,
+        | b.partition_date as yesterday,
+        | a.cl_type,
+        | a.app_version,
+        | a.c as todayCount,
+        | b.c as yesterdayCount,
+        | case when b.c = 0 then 1.0 else (a.c - b.c) * 1.0 / b.c end as chainRate
+        |(
+        | select partition_date, cl_type, app_version, count(1) as c
+        | from online.bl_hdfs_maidian_updates
+        | where partition_date = '${partition_date}'
+        | group by partition_date, cl_type, app_version
+        |) a left join (
+        | select
+        | select partition_date, cl_type, app_version, count(1) as c
+        | from online.bl_hdfs_maidian_updates
+        | where partition_date = '${yesterday}'
+        | group by partition_date, cl_type, app_version
+        |) b on a.cl_type = b.cl_type and a.app_version = b.app_version
+      """.stripMargin)
+
+    df.show
+  }
+}