update

task/tractate_analysis_in_7000.py

@@ -190,8 +190,9 @@ temp_null_count = 0
 #     500: [0, 0, 0],
 #     1000: [0, 0, 0],
 # }
-
+print(device_id_dict)
 for redis_count, device_id in enumerate(device_id_dict):
+
     # if redis_count >= 50:break
     second_demands = []
     projects = []
@@ -262,7 +263,7 @@ exposure_sql = """
 
 print(exposure_sql)
 exposure_df = spark.sql(exposure_sql)
-exposure_df.createOrReplaceTempView("exposure_df")
+# exposure_df.createOrReplaceTempView("exposure_df")
 exposure_df.show(1)
 sql_res = exposure_df.collect()
 session_pv_all = 0
@@ -276,6 +277,7 @@ baoguang_dict = {}
 # 找出card_id 对应帖子的标签 并分类汇总 得到 标签-计数字段
 for res in sql_res:
     # partition_date = res.partition_date
+    print(res)
     cl_id = res.cl_id
     if cl_id in device_id_dict:
         card_id = res.card_id
@@ -313,3 +315,14 @@ for tag_id in second_demands_id_count:
         # "session_pv_count": len(set(second_demands_id_count[tag_id]))
     }
     print(temp_dict)
+
+
+for tag_id in projects_demands_id_count:
+    temp_dict = {
+        "tag_name": tag_id,
+        "device_count": projects_demands_tag_count[tag_id],
+        "tractate_count": tags_v3_count_dict.get(tag_id),
+        "exporsure_count": len(set(projects_demands_id_count[tag_id])),
+        # "session_pv_count": len(set(second_demands_id_count[tag_id]))
+    }
+    print(temp_dict)
\ No newline at end of file