repair bug

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

-//#该文件统计ffm\node2vec\搜索推荐三个策略merge后的全量用户的ctr
-//#device_id尾号为1的用户为对照组
-//#且去除每天新用户
-
 package com.gmei
 
 import java.io.Serializable
@@ -57,11 +53,11 @@ object Recommendation_strategy_all {
 
       import sc.implicits._
       //      val stat_date = GmeiConfig.getMinusNDate(1)
-     // println(param.date)
+      println(param.date)
       val partition_date = param.date.replace("-","")
       val decive_id_oldUser = sc.sql(
         s"""
-           |select distinct(device_id) as device_id
+           |select distinct(device_id) as decive_id
            |from online.ml_device_day_active_status
            |where active_type = '4'
            |and first_channel_source_type not in ('yqxiu1','yqxiu2','yqxiu3','yqxiu4','yqxiu5','mxyc1','mxyc2','mxyc3'
@@ -77,61 +73,47 @@ object Recommendation_strategy_all {
       decive_id_oldUser.show()
       decive_id_oldUser.createOrReplaceTempView("device_id_old")
 
-      val clk_count_oldUser_experiment = sc.sql(
-        s"""
-           |select '${param.date}' as stat_date, count(cid_id) as clk_count_oldUser_experiment
-           |from data_feed_click jd inner join device_id_old
-           |on jd.device_id = device_id_old.device_id
-           |where  (jd.cid_type = 'diary' or jd.cid_type = 'diary_video')
-           |and jd.device_id not in (select device_id from bl_device_list)
-           |and jd.device_id not in (select device_id from blacklist)
-           |and jd.stat_date ='${param.date}'
-         """.stripMargin
-          )
-      clk_count_oldUser_experiment.show()
 
-      val imp_count_oldUser_experiment = sc.sql(
-        s"""
-           |select '${param.date}' as stat_date, count(cid_id) as imp_count_oldUser_experiment
-           |from data_feed_exposure je inner join device_id_old
-           |on je.device_id = device_id_old.device_id
-           |where je.cid_type = 'diary'
-           |and je.device_id not in (select device_id from bl_device_list)
-           |and je.device_id not in (select device_id from blacklist)
-           |and je.stat_date ='${param.date}'
+      val strategies = Seq("1$","[^1]$")
+      for (strategy <- strategies){
+        val clk_count_oldUser = sc.sql(
+          s"""
+             |select '${param.date}' as stat_date, count(cid_id) as get_clk_count_old
+             |from data_feed_click jd inner join device_id_old
+             |on jd.device_id = device_id_old.decive_id
+             |where  (jd.cid_type = 'diary' or jd.cid_type = 'diary_video')
+             |and jd.device_id regexp'${strategy}'
+             |and jd.device_id not in (select device_id from bl_device_list)
+             |and jd.device_id not in (select device_id from blacklist)
+             |and jd.stat_date ='${param.date}'
          """.stripMargin
-          )
-      imp_count_oldUser_experiment.show()
-      val clk_count_oldUser_all = sc.sql(
-        s"""
-           |select '${param.date}' as stat_date, count(cid_id) as clk_count_oldUser_all
-           |from data_feed_click jd inner join device_id_old
-           |on jd.device_id = device_id_old.device_id
-           |where  (jd.cid_type = 'diary' or jd.cid_type = 'diary_video')
-           |and jd.device_id not in (select device_id from bl_device_list)
-           |and jd.device_id not in (select device_id from blacklist)
-           |and jd.stat_date ='${param.date}'
+        )
+
+        val imp_count_oldUser = sc.sql(
+          s"""
+             |select '${param.date}' as stat_date, count(cid_id) as get_imp_count_old
+             |from data_feed_exposure je inner join device_id_old
+             |on je.device_id = device_id_old.decive_id
+             |where je.cid_type = 'diary'
+             |and je.device_id regexp'${strategy}'
+             |and je.device_id not in (select device_id from bl_device_list)
+             |and je.device_id not in (select device_id from blacklist)
+             |and je.stat_date ='${param.date}'
          """.stripMargin
-          )
+        )
+
+
+
+
+        val result = clk_count_oldUser.join(imp_count_oldUser,"stat_date")
+        result.show()
+
+
+
+      }
+
 
-      clk_count_oldUser_all.show()
 
-      val imp_count_oldUser_all = sc.sql(
-        s"""
-           |select '${param.date}' as stat_date, count(cid_id) as imp_count_oldUser_all
-           |from data_feed_exposure je inner join device_id_old
-           |on je.device_id = device_id_old.device_id
-           |where je.cid_type = 'diary'
-           |and je.device_id not in (select device_id from bl_device_list)
-           |and je.device_id not in (select device_id from blacklist)
-           |and je.stat_date ='${param.date}'
-         """.stripMargin
-          )
-      imp_count_oldUser_all.show()
-      val result2 = clk_count_oldUser_experiment.join(imp_count_oldUser_experiment,"stat_date")
-              .join(clk_count_oldUser_all,"stat_date")
-              .join(imp_count_oldUser_all,"stat_date")
-          result2.show()
 
     }