需求如下：
第一个：每30张卡片最多包含2张过去14天没有登录的用户，卡片尽量不要出现在前15张的位置
# 需求：每30张卡片最多包含2张过去14天没有登录的用户，卡片尽量不要出现在前15张的位置 from knight_util.get_db_token import tk_ import requests from knight_util.util import KnightUtil from knight_util.write_txt import write_log from knight_util.get_current_time import current_time import threading import json m1 = KnightUtil().mysql_() tk = tk_(2397).get('EPAL_WEB') # 创建线程锁 mutex = threading.Lock() def get_user_id_list(): headers = { "_tk_": tk, "content-type": "application/json", "user-agent": "Mozilla / 5.0(Windows NT 10.0;Win64;x64)" } data = {"limit": 30, "likeUserIds": [], "dislikeUserIds": [], "excludeUserIds": []} url = "https://test-community.gamesegirl.com/user/meets/match/v-3.6" response = requests.post(url=url, json=data, headers=headers).json() # response = json.loads(response_) if response['status'] == 'OK': response_length = len(response['content']['data']) userid_list = [response['content']['data'][i]['userId'] for i in range(response_length)] print(f"总共卡片数：{len(userid_list)}") print(f"接口返回的所有meet卡片数据{userid_list}") return userid_list def get_t_user_id(): """获取t_user表里过去14天没有登录的用户id""" start_time = current_time("sub_time", 14) print(f"查询开始时间：{start_time}") end_time = current_time("add_time", 1) print(f"查询结束时间：{end_time}") # sql = '''SELECT id FROM community.t_user t1 WHERE DATE_FORMAT(t1.last_access_time,'%Y-%m-%d %H:%i:%S') BETWEEN '2020-12-30 00:00:00' AND '2021-01-13 23:59:59';''' sql = '''SELECT id FROM community.t_user t1 WHERE DATE_FORMAT(t1.last_access_time,'%Y-%m-%d %H:%i:%S') BETWEEN '{}' AND '{}';'''.format( start_time, end_time) # 锁定 mutex.acquire() m1.execute(sql) # 释放 mutex.release() id_list = [data[0] for data in m1.fetchall()] print("在去14天登录的用户数据(含今天)") print(id_list) return id_list def get_last_access_time(id_): """获取最后一次访问时间""" sql = '''SELECT t1.last_access_time FROM community.t_user t1 WHERE id={}'''.format(id_) mutex.acquire() m1.execute(sql) mutex.release() for data in m1.fetchall(): # print(data[0]) return data[0] def run(): db_list = get_t_user_id() # 数据库数据 api_list = get_user_id_list() # 接口数据 # 没在数据库里面返回的数据 not_in_list = [id_ for id_ in api_list if id_ not in db_list] # 在数据库里面返回的数据 in_list = [id_ for id_ in api_list if id_ in db_list] expire_user_dict = {} for j in in_list: expire_user_dict.update({f"在过去14天登录的用户ID：{j})": f"最后一次访问时间{get_last_access_time(j)}"}) print(f"过去14天登录过的用户{in_list}") print(expire_user_dict) write_log(expire_user_dict) last_access_time_dict = {} for i in not_in_list: last_access_time_dict.update({f"在过去14天没有登录的用户ID{i}": f"最后一次访问时间{get_last_access_time(i)}"}) # 利用索引找到卡片位置，因索引从0开始，所以返回结果需要加1 for x in not_in_list: print(f"卡片出现位置在：{api_list.index(x) + 1}处") write_log(f"卡片出现位置在：{api_list.index(x) + 1}处") print(not_in_list) print(last_access_time_dict) write_log(last_access_time_dict) if __name__ == '__main__': run() # lock = threading.Lock() # for i in range(5): # t1 = threading.Thread(target=run) # # t1.setDaemon(True) # t1.start()
需求：每30张卡片最多包含2张欢迎度小于等于10%的用户，尽量不要出现在前15张
# 需求：每30张卡片最多包含2张欢迎度小于等于10%的用户，尽量不要出现在前15张 from knight_util.get_db_token import tk_ import requests from knight_util.Util import KnightUtil m1 = KnightUtil().mysql_() tk = tk_(2397).get('EPAL_WEB') def get_user_id_list(): headers = { "_tk_": tk, "content-type": "application/json", "user-agent": "Mozilla / 5.0(Windows NT 10.0;Win64;x64)" } data = {"limit": 30, "likeUserIds": [], "dislikeUserIds": [], "excludeUserIds": []} url = "https://test-community.gamesegirl.com/user/meets/match/v-3.6" response = requests.post(url=url, json=data, headers=headers).json() if response['status'] == 'OK': response_length = len(response['content']['data']) userid_list = [response['content']['data'][i]['userId'] for i in range(response_length)] print(f"总共卡片数：{len(userid_list)}") print(f"接口返回的所有meet卡片数据{userid_list}") return userid_list def get_t_user_id(): """获取欢迎度小于10%的""" sql = '''SELECT user_id,popularity FROM community.user_meets um WHERE um.popularity <=0.10;''' m1.execute(sql) id_list = [data[0] for data in m1.fetchall()] print(f"欢迎度小于10%的所有数据{id_list}") return id_list def get_popularity(user_id): sql = '''SELECT popularity FROM community.user_meets WHERE user_id={};'''.format(user_id) m1.execute(sql) for data in m1.fetchall(): return data[0] def get_region_name(user_id): sql = '''SELECT country,region_name FROM egirl.t_dict_area_info WHERE user_id={};'''.format(user_id) m1.execute(sql) for data in m1.fetchall(): print(data) return data def run(): db_list = get_t_user_id() # 数据库数据 api_list = get_user_id_list() # 接口数据 # 欢迎度大于10 popularity_greater = [id_ for id_ in api_list if id_ not in db_list] print(f"接口返回欢迎度大于10%的数据{popularity_greater}") # 欢迎度小于10 popularity_less = [id_ for id_ in api_list if id_ in db_list] print(f"接口返回欢迎度小于10%的数据{popularity_less}") popularity_dict = {} for j in popularity_less: popularity_dict.update({f"欢迎度userID是：{j}": get_popularity(j)}) print(popularity_dict) # 利用索引找到卡片位置，因索引从0开始，所以返回结果需要加1 for x in popularity_less: print(f"卡片出现位置在：{api_list.index(x) + 1}处") # 查是否同城 for t in popularity_less: get_region_name(t) if __name__ == '__main__': run()
每30张卡片最多包含2张非真人图片，尽量不要出现在前15张
# 每30张卡片最多包含2张非真人图片，尽量不要出现在前15张 from knight_util.get_db_token import tk_ import requests from knight_util.Util import KnightUtil from knight_util.write_txt import write_log m1 = KnightUtil().mysql_() tk = tk_(2397).get('EPAL_WEB') def get_user_id_list(): headers = { "_tk_": tk, "content-type": "application/json", "user-agent": "Mozilla / 5.0(Windows NT 10.0;Win64;x64)" } data = {"limit": 30, "likeUserIds": [], "dislikeUserIds": [], "excludeUserIds": []} url = "https://test-community.gamesegirl.com/user/meets/match/v-3.6" response = requests.post(url=url, json=data, headers=headers).json() if response['status'] == 'OK': response_length = len(response['content']['data']) userid_list = [] for i in range(response_length): userid_list.append(response['content']['data'][i]['userId']) print(userid_list) print(f"总共卡片数：{len(userid_list)}") return userid_list def is_person(user_id): """是否真人""" sql = f"SELECT real_person_img FROM community.user_meets WHERE user_id={user_id} AND user_img_status='PASS' AND status='ENABLED';" m1.execute(sql) for data in m1.fetchall(): return data[0] def get_real_person_code(user_id): """获取是否是真人图片状态码""" sql = '''SELECT real_person_img FROM community.user_meets WHERE user_id={} '''.format(user_id) m1.execute(sql) for data in m1.fetchall(): return data[0] def run(): userid_dict = {} id_list = get_user_id_list() # 是真人 real_people = [] # 不是真人 not_real_people = [] for i in id_list: if is_person(i) == 0: real_people.append(i) # userid_dict.update({f"user_id：{i}": f"是真人"}) elif is_person(i) == 1: not_real_people.append(i) # userid_dict.update({f"user_id：{i}": f"不是真人"}) print(f"是真人图片的userid：{real_people}") print(f"不是真人图片的userid：{not_real_people}") dic = {} for i in not_real_people: dic.update({i: get_real_person_code(i)}) print(f"不是真人图片的数据：{dic}") for x in not_real_people: print(f"卡片出现位置在：{id_list.index(x) + 1}处") if __name__ == '__main__': run()
需求如图：
代码如下：
# 智能图片正向场景测试 from knight_util.MysqlUtil import MysqlPool import requests from multiprocessing import Pool from knight_util.Util import KnightUtil from knight_util.sso_token import query_meet_token import random from knight_util.write_txt import write_log mysql = MysqlPool() redis = KnightUtil().redis_() # TODO 全局变量user_id user_id = [3801] token_list = query_meet_token(user_id[0]) def query_redis_exposureCount(): """查询曝光度""" redis_data = redis.get("meets:image:calculate:{}".format(user_id[0])) img_dict = {} for i in range(len(eval(redis_data))): img_dict.update({eval(redis_data)[i]['userImg']: eval(redis_data)[i]['exposureCount']}) write_log(eval(redis_data)[i]) count_list = [j for j in img_dict.values()] return sum(count_list) def batch_like(userid_list=user_id): """批量喜欢""" # 查计算中的图片 sql = "SELECT user_img FROM community.user_meets_image WHERE user_id={} AND calculate_status='CALCULATING';".format( userid_list[0]) data = mysql.fetch_all(sql) # 计算中的图片 # user_img_list = [i['user_img'] for i in data] user_img_list = ['https://community-oss.epal.gg/data/community/meet/pt/1611841764302.jpeg'] # 删除数据触发列表 triggering_condition_list = [20, 80, 130, 180, 230, 270] like_num = 0 while True: for tk in token_list: like_num += 1 print(f"第{like_num}次请求开始") write_log(f"第{like_num}次请求开始") # 喜欢状态 like_status_list = ["like", "dislike"] # like_status_list = ["like", "dislike", "superlike"] # 随机喜欢不喜欢 status = random.choice(like_status_list) # 随机正在计算中的图片 img_url = random.choice(user_img_list) # 请求url, token, 参数 headers = { "_tk_": tk } url = "https://test-community.gamesegirl.com/user/meets/likeOrDislike" data1 = {"likeUserIds": userid_list, "dislikeUserIds": [], "likeUserIdAndImg": [{"userId": userid_list[0], "imgUrl": img_url}], "disLikeUserIdAndImg": []} data2 = {"likeUserIds": [], "dislikeUserIds": userid_list, "likeUserIdAndImg": [], "disLikeUserIdAndImg": [{"userId": userid_list[0], "imgUrl": img_url}]} if status == "like": # 喜欢 try: response1 = requests.post(url=url, json=data1, headers=headers).json() print(response1) if response1['status'] == 'OK': print(f"喜欢成功：{img_url}") print(f"曝光总次数：{query_redis_exposureCount()}") # 总曝光数等于300时结束运行 if query_redis_exposureCount() == 299: print("结束计算") return # 满足条件时，删除meet关系数据 if like_num in triggering_condition_list: del_sql = "DELETE FROM community.user_relation WHERE user_id={} OR to_user_id={};".format( userid_list[0], userid_list[0]) print(f"共删除{mysql.delect(del_sql)}条数据") except Exception as e: raise e # 超级喜欢 elif status == "superlike": try: super_url = "https://test-community.gamesegirl.com/user/meets/super/like" data3 = {"userId": userid_list[0], "imgUrl": img_url} response3 = requests.post(url=super_url, json=data3, headers=headers).json() print(response3) if response3['status'] == 'OK': print(f"超级喜欢成功：{img_url}") print(f"曝光总次数：{query_redis_exposureCount()}") # # 删除超级喜欢数据 list_keys = redis.keys("1002:test:user:member:superLike:*") if list_keys: for key in list_keys: print(redis.delete(key)) print("删除超级喜欢会员数据成功") # 总曝光数等于298时结束运行 if query_redis_exposureCount() == 299: print("结束计算") return # 满足条件时，删除meet关系数据 if like_num in triggering_condition_list: del_sql = "DELETE FROM community.user_relation WHERE user_id={} OR to_user_id={};".format( userid_list[0], userid_list[0]) print(f"共删除{mysql.delect(del_sql)}条数据") except Exception as e: raise e else: # 不喜欢 try: response2 = requests.post(url=url, json=data2, headers=headers).json() print(response2) if response2['status'] == 'OK': print(f"不喜欢成功：{img_url}") print(f"曝光总次数：{query_redis_exposureCount()}") # 总曝光数等于298时结束运行 if query_redis_exposureCount() == 299: print("结束计算") # print("开始写入最终结果") return # 满足条件时，删除meet关系数据 if like_num in triggering_condition_list: del_sql = "DELETE FROM community.user_relation WHERE user_id={} OR to_user_id={};".format( userid_list[0], userid_list[0]) print(f"共删除{mysql.delect(del_sql)}条数据") except Exception as e: raise e if __name__ == '__main__': pool = Pool(6) pool.apply_async(batch_like) pool.close() pool.join() # a = query_redis_exposureCount() # print(a)