update dairyQueueUpdate file

local/diary2.0.py

@@ -172,7 +172,7 @@ def update_dairy_queue(score_df,predict_score_df):
         return score_df.index.tolist()
 
 
-def update_sql_dairy_queue(queue_name, diary_id,device_city):
+def update_sql_dairy_queue(queue_name, diary_id,device_id, city_id):
     db = pymysql.connect(host='rdsmaqevmuzj6jy.mysql.rds.aliyuncs.com', port=3306, user='work',
                          passwd='workwork', db='doris_test')
     cursor = db.cursor()
@@ -182,7 +182,7 @@ def update_sql_dairy_queue(queue_name, diary_id,device_city):
     print("写入前")
     print(id_str[:80])
     sql = "update device_diary_queue set {}='{}' where device_id = '{}' and city_id = '{}'".format\
-        (queue_name,id_str,device_city[0],device_city[1])
+        (queue_name,id_str,device_id, city_id)
     cursor.execute(sql)
     db.commit()
     db.close()
@@ -211,10 +211,10 @@ def get_native_queue(device_id,city_id):
 def multi_update(queue_name,queue_arg,device_id,city_id):
     if queue_arg[0] != []:
         diary_id = predict(queue_name,queue_arg,device_id,city_id)
-        update_sql_dairy_queue(queue_name, diary_id,device_id,city_id)
-        print("更新结束")
+        return diary_id
     else:
         print("预测集是空，不需要预测")
+        return False
 
 
 def get_queue(device_id, city_id,queue_name):
@@ -227,35 +227,33 @@ def get_queue(device_id, city_id,queue_name):
     cursor.execute(sql)
     result = cursor.fetchall()
     df = pd.DataFrame(list(result))
-
     if not df.empty:
-
         queue_list = df.loc[0,0].split(",")
         queue_list = list(map(lambda x: "diary|" + str(x), queue_list))
-
         db.close()
-
-        return True, queue_list
+        return queue_list
     else:
         print("该用户对应的日记队列为空")
-        return False, []
+        return False
 
 
 def user_update(device_id,city_id):
+    global native_queue_list
     queue_name_list = ["native_queue","nearby_queue","nation_queue","megacity_queue"]
     for queue_name in queue_name_list:
-        exist,queue_list = get_queue(device_id, city_id,queue_name)
-        # 下面的代码是用来对比native_queue是否发生变化，如果发生了变化，就不更新日记队列了
-        # if queue_name == "native_queue":
-        #     native_queue_list =
-
-
-        if exist:
-            queue_predcit = list(set(queue_list) & set(data_set_cid))
-            queue_not_predcit = list(set(queue_list) - set(data_set_cid))
-            queue_arg = [queue_predcit,queue_not_predcit,queue_list]
-            multi_update(queue_name,queue_arg,device_id,city_id)
-
+        queue_list = get_queue(device_id, city_id,queue_name)
+        if queue_name == "native_queue":
+            native_queue_list = queue_list
+        if queue_list:
+            queue_predict = list(set(queue_list) & set(data_set_cid))
+            queue_not_predict = list(set(queue_list) - set(data_set_cid))
+            queue_arg = [queue_predict,queue_not_predict,queue_list]
+            diary_id = multi_update(queue_name, queue_arg, device_id, city_id)
+            if diary_id and (native_queue_list == get_native_queue(device_id,city_id)):
+                update_sql_dairy_queue(queue_name, diary_id, device_id, city_id)
+                print("更新结束")
+            else:
+                print("不需要更新日记队列")
         else:
             print("日记队列为空")
 

userProfile.py

 from utils import con_sql
 from datetime import datetime
+from config import *
+import pandas as pd
+import os
+import time
 
 
 # 获取当下一分钟内活跃用户
@@ -7,19 +11,40 @@ def get_active_users():
     now = datetime.now()
     now_start = str(now)[:16] + ":00"
     now_end = str(now)[:16] + ":59"
-    没有city_id的是“” 这个值可能是空
-    sql = "select device_id from user_active_time order by active_time desc limit 1;"
-    # sql = "select device_id from user_active_time " \
+    sql = "select device_id,city_id from user_active_time limit 1;"
+    # TODO 正式上线后用下面的sql语句
+    # sql = "select device_id,city_id from user_active_time " \
     #       "where active_time <= '{}' and active_time >= '{}'".format(now_end,now_start)
-    device_id_df = con_sql(sql)
-    if device_id_df.empty:
+    df = con_sql(sql)
+    if df.empty:
         print("当下这一分钟没有活跃用户，不需要预测")
-        return True,None
+        for eachFile in os.listdir("/tmp"):
+            if "xlearn" in eachFile:
+                os.remove("/tmp" + "/" + eachFile)
+        time.sleep(58)
+
+        return False
     else:
-        device_id_list = device_id_df[0].values.tolist()
-        # 对device_id 进行去重
-        device_id_list = list(set(device_id_list))
-        return False,device_id_list
+        df = df.rename(columns={0: "device_id", 1: "city_id"})
+        old_device_id_list = pd.read_csv(DIRECTORY_PATH + "data_set_device_id.csv")["device_id"].values.tolist()
+        # 求活跃用户和老用户的交集，也就是只预测老用户
+        df = df.loc[df["device_id"].isin(old_device_id_list)]
+        if df.empty:
+            print("该列表是新用户，不需要预测")
+        else:
+            # TODO 正式上线后注释下面的只预测尾号是6的代码
+            # 只预测尾号是6的ID，这块是测试要求的，这块也可以在数据库取数据时过滤一下
+            device_temp_list = df["device_id"].values.tolist()
+            predict_list = list(filter(lambda x: str(x)[-1] == "6", device_temp_list))
+            df = df.loc[df["device_id"].isin(predict_list)]
+            if df.empty:
+                print("没有尾号是6的用户，不需要预测")
+            else:
+                device_list = df["device_id"].values.tolist()
+                city_list = df["city_id"].values.tolist()
+                device_city_list = list(zip(device_list, city_list))
+
+                return device_city_list
 
 
 def fetch_user_profile(device_id):