smart_rank 相关的统计指标

eda/feededa/src/main/scala/com/gmei/temp_analysis.scala

@@ -602,7 +602,6 @@ object smart_rank_count {
          """.stripMargin
       )
       blacklist_id.createOrReplaceTempView("blacklist_id")
-      blacklist_id.show()
 
       val final_id = sc.sql(
         s"""
@@ -614,20 +613,20 @@ object smart_rank_count {
          """.stripMargin
       )
       final_id.createOrReplaceTempView("final_id")
-      final_id.show()
 
 
       val user_city_meigou_view = sc.sql(
         s"""
-           |select cl_id as device_id,city_id as device_city,params['business_id'] as meigou_id
-           |from online.tl_hdfs_maidian_view
-           |where action = "page_view"
-           |and params['page_name']="welfare_detail"
-           |and partition_date >='20181201'
-           |and city_id is not null
+           |select ov.cl_id as device_id,ov.city_id as device_city,ov.params['business_id'] as meigou_id
+           |from online.tl_hdfs_maidian_view ov left join final_id
+           |on ov.cl_id = final_id.device_id
+           |where ov.action = "page_view"
+           |and ov.params['page_name']="welfare_detail"
+           |and ov.partition_date >='20181201'
+           |and ov.city_id is not null
+           |and final_id.device_id  is null
        """.stripMargin
       )
-      user_city_meigou_view.show()
       user_city_meigou_view.createOrReplaceTempView("user_city_meigou_view")
 
       val meigou_city = sc.sql(
@@ -702,17 +701,17 @@ object smart_rank_count {
 
       //开始计算每个地区每月新增设备
 
-//      val device_new_count = sc.sql(
-//        s"""
-//           |select first_city,count(distinct(device_id))
-//           |from online.ml_device_day_active_status
-//           |where active_type != '4'
-//           |and partition_date >='20181201'
-//           |group by first_city
-//           |order by first_city
-//       """.stripMargin
-//      )
-//      device_new_count.show()
+      val device_new_count = sc.sql(
+        s"""
+           |select first_city,count(distinct(device_id))
+           |from online.ml_device_day_active_status
+           |where active_type != '4'
+           |and partition_date >='20181201'
+           |group by first_city
+           |order by first_city
+       """.stripMargin
+      )
+      device_new_count.show()