新增推荐脚本

databaseConn/__init__.py

@@ -16,6 +16,7 @@ mysql_info={
         "host":"172.24.17.119",
         "port":"15307"
     },
+
     "220":{
         "name":"recommender_user_rw",
         "pwd":"rkwnPEG4",
@@ -51,6 +52,13 @@ mysql_info={
         "pwd":"3^4f=GZeA$",
         "host":"recommend-nearline-prod-5-7.cr22rydq7l0o.rds.cn-north-1.amazonaws.com.cn",
         "port":"3306"
+    },
+    "usergroup":{
+        "name":"tjzimuali_test",
+        "pwd":"bYut032QgkJ7ElMXwW",
+        "host":"aliyun-test.mongodb.rds.aliyuncs.com:3717,dds-2ze3634f3b850e642.mongodb.rds.aliyuncs.com",
+        "port":"3717"
+        # mongodb://tjzimuali_test:bYut032QgkJ7ElMXwW@aliyun-test.mongodb.rds.aliyuncs.com:3717,dds-2ze3634f3b850e642.mongodb.rds.aliyuncs.com:3717/admin?replicaSet=mgset-58488296
     }
 }
 

databaseConn/mongoOperation.py

@@ -9,14 +9,22 @@ def mongo_universal(namespace):
     print("mongodb连接信息:",temp)
     return temp
 
-def mongoClient(namespace,db="app_vcc2"):
+def personmongoClient(namespace,db="app_vcc2"):
     '''
     :param namespace:
     :param db:
     :return: 自动连接该db
     '''
+    #mongo_info=mysqlInfo(namespace,dbtype='mongodb')
     temp=mongo_universal(namespace)
-    return eval("MongoClient(temp).{db}".format(db=db))
+    #temp_db=eval("MongoClient(temp)")# .format(db=db))
+    temp_db=eval("MongoClient(temp).{db}".format(db=db))
+    try:
+        #if temp_db.authenticate(mongo_info.get('name'),mongo_info.get('pwd')):
+        return temp_db
+    except:
+        print('moongo连接失败',type(temp_db))
+        traceback.print_exc(2)
 
 def tableSearch(namespace,tablename,**kwargs):
     '''
@@ -27,7 +35,7 @@ def tableSearch(namespace,tablename,**kwargs):
     :return: 返回find结果
     '''
     temp=''
-    db=mongoClient(namespace,kwargs.get('db'))
+    db=personmongoClient(namespace,kwargs.get('db'))
     tables=[i['name'] for i in db.list_collections()]
     if tablename not in tables:
         raise Exception('table[{}]不在databse[{}]中,所有的table如下:'.format(tablename,kwargs.get('db')),tables)
@@ -43,12 +51,13 @@ def tableSearch(namespace,tablename,**kwargs):
 
 
 if __name__ == '__main__':
-    key={"db":"ocrDb","query":[{
-       #,"userId":"70549608"
-    },{"_id":1}]}
-    temp=tableSearch('qa','system.version',**key)
-    result=[i for i in temp]
-    print(len(result))
-    for i in temp:
-        print('===',i)
-    print(temp)
\ No newline at end of file
+    # key={"db":"ocrDb","query":[{
+    #    #,"userId":"70549608"
+    # },{"_id":1}]}
+    # temp=tableSearch('usergroup','ocrVerifyBaseInfo',**key)
+    # result=[i for i in temp]
+    # print(len(result))
+    # for i in temp:
+    #     print('===',i)
+    # print(temp)
+    print(personmongoClient('usergroup'))

databaseConn/mysqlOperation.py

@@ -72,4 +72,5 @@ def concatSql(sql,**kwargs):
     return sql+' where '+temp[:-4]
 
 if __name__ == '__main__':
-    print(execmysl('qa',"select sku_name from kdsp.t_sku_info where  sku_name like '%%口红%%'"))
\ No newline at end of file
+    print(mysqlInfo('test1'))
+    #print(execmysl('qa',"select sku_name from kdsp.t_sku_info where  sku_name like '%%口红%%'"))
\ No newline at end of file

databaseConn/redisOperation.py

@@ -54,7 +54,7 @@ def getRedisValue(key,conntype=None,oper='select'):
         redis_db_conn=dapan_redis_db_conn
     else:
         redis_db_conn=redis_db_conn_119
-    uuid_redis_result=redis_db_conn.keys(key)
+    uuid_redis_result=redis_db_conn.keys(key.strip())
     print('模糊匹配到的所有key:',uuid_redis_result)
     if uuid_redis_result and oper=='delete':
         for i in uuid_redis_result:
@@ -68,7 +68,7 @@ def getRedisValue(key,conntype=None,oper='select'):
     for i in uuid_redis_result:
         redis_key_type=redis_db_conn.type(i)
         if redis_key_type == 'list':
-            redis_value[i]=redis_db_conn.lrange(i,0,100)
+            redis_value[i]=redis_db_conn.lrange(i,0,300)
             # uuid_redis_result = redis_db_conn.lrange(i)
         elif redis_key_type in ('dict','hash'):
             redis_value[i]=redis_db_conn.hgetall(i)
@@ -84,7 +84,8 @@ def getRedisValue(key,conntype=None,oper='select'):
 
 if __name__ == '__main__':
     userUuid='00003a93-2a32-4501-b338-755b6cb1ec49'
-    t=getRedisValue("same_product_batch_uuid")
+    #t=getRedisValue("order_product_graph:179240378044417")
+    t=getRedisValue("pay_product_graph:179240378044417")
     print(t)
 
     ## search_fm_offline_feature:{md5(搜索词)}_query_offline

recommend/cid3brandname_group_recommend.py

+from recommend import *
+def appTopic():
+    temp=defaultdict(list)
+    cid3_cate_brand_heat_rank_change_sql=cate_brand_heat_rank_sql+' where show_type=1 order by rank asc limit 30'
+    brandname_cate_brand_heat_rank_change_sql=cate_brand_heat_rank_sql+' where show_type=2 order by rank asc limit 30'
+    cid3_df=execmysl(119,cid3_cate_brand_heat_rank_change_sql)
+    brandname_df=execmysl(119,brandname_cate_brand_heat_rank_change_sql)
+    temp['apptopic_cid3']=cid3_df['show_id'].to_list() if not cid3_df.empty else []
+    temp['apptopic_brandname']=cid3_df['show_id'].to_list() if not cid3_df.empty else []
+    return temp
+
+def clickTopic(skuno,cid3=None,brandid=None):
+    temp=defaultdict()
+    from recommend.publicFunc import skuinfo
+    sku_infos=skuinfo(skuno)
+    if cid3 and brandid:
+        sku_infos['cid3']=cid3
+        sku_infos['brand_id']=brandid
+
+    cid3_intention_change_sql=concatSql(cate_brand_intention_score_sql,
+                                        **{"show_id":[sku_infos.get('cid3'),sku_infos.get('cid2'),sku_infos.get('cid1')]})
+    cid3_intention_change_sql+=" or brand_id='{}'".format(sku_infos.get('brand_id'))
+    print('cicktopic 转换后的sql:',cid3_intention_change_sql)
+    cid3brandname_df=execmysl(119,cid3_intention_change_sql)#.sort_values(by=['rank'])['']
+    print('cicktopic的结果:',cid3brandname_df.head(1))
+    ##cid3+brandid
+    temp['clicktopic_cid3']=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid3'))
+                                             & (cid3brandname_df['show_type']==1) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                    .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##cid2+brandid
+    temp['clicktopic_cid3']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid2'))
+                                              & (cid3brandname_df['show_type']==1) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##cid1+brandid
+    temp['clicktopic_cid3']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid1'))
+                                              & (cid3brandname_df['show_type']==1) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##cid3+无名品牌分
+    temp['clicktopic_cid3']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid3'))
+                                              & (cid3brandname_df['show_type']==1) &(cid3brandname_df['brand_id']==-1)] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##cid2+无名品牌分
+    temp['clicktopic_cid3']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid2'))
+                                              & (cid3brandname_df['show_type']==1) &(cid3brandname_df['brand_id']==-1)] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##cid1+无名品牌分
+    temp['clicktopic_cid3']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid1'))
+                                              & (cid3brandname_df['show_type']==1) &(cid3brandname_df['brand_id']==-1)] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##无品类分+品牌分
+    temp['clicktopic_cid3']+=cid3brandname_df[(cid3brandname_df['show_id']==-1)
+                                              & (cid3brandname_df['show_type']==1) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+
+    temp['clicktopic_cid3']=removeRepeat(temp['clicktopic_cid3'])[:30]
+    ##cid3+brandid
+    temp['clicktopic_brandname']=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid3'))
+                                             & (cid3brandname_df['show_type']==2) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                                .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##cid2+brandid
+    temp['clicktopic_brandname']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid2'))
+                                              & (cid3brandname_df['show_type']==2) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##cid1+brandid
+    temp['clicktopic_brandname']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid1'))
+                                              & (cid3brandname_df['show_type']==2) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##cid3+无名品牌分
+    temp['clicktopic_brandname']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid3'))
+                                              & (cid3brandname_df['show_type']==2) &(cid3brandname_df['brand_id']==-1)] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##cid2+无名品牌分
+    temp['clicktopic_brandname']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid2'))
+                                              & (cid3brandname_df['show_type']==2) &(cid3brandname_df['brand_id']==-1)] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##cid1+无名品牌分
+    temp['clicktopic_brandname']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid1'))
+                                              & (cid3brandname_df['show_type']==2) &(cid3brandname_df['brand_id']==-1)] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    ##无品类分+品牌分
+    temp['clicktopic_brandname']+=cid3brandname_df[(cid3brandname_df['show_id']==-1)
+                                              & (cid3brandname_df['show_type']==2) &(cid3brandname_df['show_id']==sku_infos.get('brand_id'))] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+
+    temp['clicktopic_brandname']=removeRepeat(temp['clicktopic_brandname'])[:30]
+
+    return temp
+
+
+def searchTopic():
+    temp=defaultdict(list)
+    from recommend.searchTopic import getseed,step1
+    top10_sku,_=step1()
+    sku_infos=getseed(top10_sku)
+    cid3_intention_change_sql=concatSql(cate_brand_intention_score_sql,
+                                        **{"show_id":sku_infos.get('cid3')})
+    cid3_intention_change_sql+=" or brand_id='{}'".format(sku_infos.get('brand_id'))
+    cid3brandname_df=execmysl(119,cid3_intention_change_sql)
+    temp['searchtopic_cid3']=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid3'))
+                                             & (cid3brandname_df['show_type']==1) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                                .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+
+    temp['searchtopic_cid3']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid3'))
+                                              & (cid3brandname_df['show_type']==1) &(cid3brandname_df['brand_id']==-1)] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+
+    temp['searchtopic_cid3']+=cid3brandname_df[(cid3brandname_df['show_id']==-1)
+                                                   & (cid3brandname_df['show_type']==1) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                                      .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+
+    temp['searchtopic_brandname']=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid3'))
+                                              & (cid3brandname_df['show_type']==2) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                                 .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+
+    temp['searchtopic_brandname']+=cid3brandname_df[(cid3brandname_df['show_id']==sku_infos.get('cid3'))
+                                               & (cid3brandname_df['show_type']==2) &(cid3brandname_df['brand_id']==-1)] \
+                                  .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+
+    temp['searchtopic_brandname']+=cid3brandname_df[(cid3brandname_df['show_id']==-1)
+                                               & (cid3brandname_df['show_type']==2) &(cid3brandname_df['brand_id']==sku_infos.get('brand_id'))] \
+                                  .sort_values(by=['rank'])['show_id'].to_list()[:30]  if not cid3brandname_df.empty else []
+    temp['searchtopic_cid3']=removeRepeat(temp['searchtopic_cid3'])[:30]
+    temp['searchtopic_brandname']=removeRepeat(temp['searchtopic_brandname'])[:30]
+    return temp
+
+
+
+
+
+
+
+if __name__ == '__main__':
+    print(clickTopic('179240378044417'))
+
+
+
+
+

recommend/payTopic.py

+from recommend import *
+
+
+def dealData(t):
+    print('处理成json的数据:',t)
+    tt=json.loads(t)
+    temp=[]
+    for i in tt:
+        temp.append(i.get('sku_id'))
+    return temp
+
+
+def payRecall(skuno,num=100):
+    temp=defaultdict(list)
+    print("支付图关联表召回start")
+    pay_product_change_redis=pay_product_redis.strip().format(sku_no=skuno)
+    #print(pay_product_change_redis)
+    pay_result=getRedisValue(pay_product_change_redis)
+    from recommend.publicFunc import skuinfo
+    skuno_info=skuinfo(skuno)
+    pay_skus=dealData(pay_result.get(pay_product_change_redis))
+    if pay_result :
+        skuss_change_sql=concatSql(skuinfo_sql,**{"sku_no":pay_skus}) +" and cid3!='{}'".format(skuno_info.get('cid3'))
+        skdss=execmysl('nearly',skuss_change_sql)
+        if not skdss.empty :
+            final_skus=skdss['sku_no'].to_list()
+            temp['70011']=final_skus[:num]
+
+            print("支付图关联表召回stop",len(final_skus),skuss_change_sql)
+            #return temp
+
+    print("下单图关联表召回start",'**'*50)
+    order_product_change_redis=order_product_redis.format(sku_no=skuno).strip()
+    order_result=getRedisValue(order_product_change_redis)
+    order_skus=dealData(order_result.get(order_product_change_redis))
+    if order_result :
+        skuss_change_sql=concatSql(skuinfo_sql,**{"sku_no":order_skus}) +' and cid3!={}'.format(skuno_info.get('cid3'))
+        skdss=execmysl('nearly',skuss_change_sql)
+        if not skdss.empty :
+            final_skus=skdss['sku_no'].to_list()
+            chaji=subtraction(final_skus,temp.get('70011'))
+            temp['70012']=chaji[:num]
+            print("下单图关联表召回stop",len(final_skus),skuss_change_sql)
+            #return temp
+    print("加购图关联表召回start",'**'*50)
+    product_change_redis=product_redis.format(sku_no=skuno).strip()
+    product_result=getRedisValue(product_change_redis)
+    if product_result :#and len(product_result.get(product_change_redis))>=num:
+        skuss_change_sql=concatSql(skuinfo_sql,**{"sku_no":product_result.get(product_change_redis)}) +' and cid3!={}'.format(skuno_info.get('cid3'))
+        skdss=execmysl('nearly',skuss_change_sql)
+        if not skdss.empty :#and skdss.shape>num:
+            final_skus=skdss['sku_no'].to_list()
+            chaji=subtraction(subtraction(final_skus,temp.get('70011')),temp.get('70012'))
+            temp['70013']=chaji
+            print("加购图关联表召回stop")
+            return temp
+
+    print("ALS相关表召回start",'**'*50)
+    als_correlation_change_redis=als_correlation_redis.format(sku_no=skuno).strip()
+    als_correlation_result=getRedisValue(als_correlation_change_redis)
+
+    if als_correlation_result :#and len(als_correlation_result.get(als_correlation_change_redis))>=num:
+        skuss_change_sql=concatSql(skuinfo_sql,**{"sku_no":als_correlation_result.get(als_correlation_change_redis)}) +' and cid3!={}'.format(skuno_info.get('cid3'))
+        skdss=execmysl('nearly',skuss_change_sql)
+        if not skdss.empty: #and skdss.shape>num:
+            final_skus=skdss['sku_no'].to_list()
+            chaji=subtraction(subtraction(subtraction(final_skus,temp.get('70011')),temp.get('70012')),temp.get('70013'))
+            temp['70014']=chaji
+            print("ALS相关表召回stop",len(skdss),skuss_change_sql)
+            #return temp
+
+    dapan_change_sql=concatSql(dapan_sql,**{"cid3":skuno_info.get('cid3')})+" and brand_name!='{}'".format(skuno_info.get('brand_name'))
+
+    dapan_df=execmysl('nearly',dapan_change_sql)
+    #if dapan_df.shape[0]>=50:
+    temp['70015_1']=dapan_df['sku_no'].to_list()[:50]
+
+    #return temp
+
+    dapan_change_sql=concatSql(dapan_sql,**{"brand_name":skuno_info.get('brand_name')})+' and cid3!={}'.format(skuno_info.get('cid3'))
+    dapan_df=execmysl('nearly',dapan_change_sql)
+    temp['70015_2']=dapan_df['sku_no'].to_list()[:50]
+    return temp
+
+def backRecall(uuid):
+    back_recall_change_sql=nearly_batch_sql.format(uuid)+' and recall_reason!=50000'
+    back_recall_result=execmysl('nearly',back_recall_change_sql)
+    return {"7002":back_recall_result['sku_no'].to_list()} if not back_recall_result.empty else {"7002":[]}
+
+def payDapanbu(skuno):
+    try:
+        from recommend.publicFunc import dapanBu
+        dapan_result=dapanBu()
+        return {"7003":dapan_result[:300]}
+    except :
+        traceback.print_exc(2)
+        return {"7003":[]}
+
+
+def all(skuno,uuid):
+    temp=defaultdict()
+    sku_temp=[]
+    #back_recall_skus=backRecall(uuid)
+    #temp.append(back_recall_skus)
+    pay_recall_skus=payRecall(skuno,num=100)
+    temp.update(pay_recall_skus)
+
+    dapan_result=payDapanbu(skuno)
+    for k,v in temp.items():
+        sku_temp+=v
+    print('支付召回数据',sku_temp[:300])
+
+    if len(removeRepeat(mergelist(sku_temp)))<300:
+        temp.update(dapan_result)
+
+    return temp
+
+def breakUp(temp):
+    pass
+
+
+
+
+def checkresult():
+    a=['100012355342', '11050632353793', '11102968880129', '179229279393281', '188701955858944', '214857903777280', '214864564329984', '216227226849793', '219236606151681', '219717743418368', '219717760191488', '220405299154944', '221238765822976', '221242506611712', '221326317721601', '226455405730817', '2302639605249', '230741925957633', '232141556812289', '238057538721280', '239217926999553', '241458842967559', '241462055803909', '2428586170369', '243851634153473', '244342367719425', '245090111202307', '246531249415169', '246532608630785', '246615437743105', '257383818600960', '257401358391296', '261007085148162', '266081478647298', '275001899225600', '279850356055553', '281430559826433', '281529478289409', '282035714270724', '29169639759873', '298224007783425', '298225467657729', '301741544312332', '302492895152129', '307494551820801', '307565317857793', '311153452910081', '316840577339904', '316975868289025', '319861666615296', '320721197213185', '322042655152128', '322740226627585', '326397760381953', '328040374869504', '328575677900289', '340438528036359', '340913373319168', '340970164461576', '343082273351169', '343204705611264', '343222237008385', '343594322895873', '343989417936385', '346650937922049', '346845511949325', '346848867125256', '346936259644929', '349035157262337', '351699319917057', '351737404724737', '351824041478657', '353012455837185', '353229235030529', '355068805384705', '356766476148225', '356795642291201', '359044015728131', '361900941516801', '361913264637441', '367038149960712', '367681053661185', '369193007195649', '369201421750785', '371346178315265', '372730534042113', '372781200972325', '373380391116801', '374333605684738', '374340064647694', '377681608183297', '377843289688065', '37992240793601', '37993935289345', '383811482819591', '64833806673409', '794422', '384196679049746', '393791694966273']
+    sql="""
+    select  -- a.*,
+distinct b.sku_no,
+b.recall_reason 
+-- count(1)
+from 
+(select id,recall_batch_uuid,event_type from nearline_recommend_test.recommend_product_batch 
+where user_uuid='1e51cf78-e12c-4f17-9d5b-b25404303a66-3' order by id desc limit 1)a
+join nearline_recommend_test.recommend_product_record b on a.recall_batch_uuid=b.recall_batch_uuid
+where -- b.sku_no='37993381637121'
+recall_reason=70011
+    """
+    df=execmysl('nearly',sql)['sku_no'].to_list()
+    print(subtraction(a,df))
+
+
+if __name__ == '__main__':
+    skuno='179240378044417'
+    uuid='1e51cf78-e12c-4f17-9d5b-b25404303a66-5'
+    t=all(skuno,uuid)
+    print(t)
+    #print(payDapanbu(skuno))
+    #print(deal_all())
+    #print(checkresult())
+
+
+
+

recommend/publicFunc.py

@@ -42,7 +42,7 @@ def minPriceFill(skus,num=100):
 def skuinfo(sku):
     change_sql=concatSql(skuinfo_sql,**{"sku_no":sku})
     sku_df=execmysl(119,change_sql)
-    sku_info=sku_df[['cid1','cid2','cid3','brand_name']].to_dict(orient='records')[0]
+    sku_info=sku_df[['cid1','cid2','cid3','brand_name','brand_id']].to_dict(orient='records')[0]
     return sku_info
 
 

recommend/publicSql.py

@@ -14,11 +14,11 @@ similar_redis='product_similarity:{skuno}'
 ##商品关联度
 correlation_redis='product_correlation:${skuNo}'
 
-##本批次+上一批次
+##上一批次
 nearly_batch_sql="""
 select a.*,b.sku_no,b.recall_reason from 
-(select id,recall_batch_uuid,event_type from nearline_recommend.recommend_product_batch where user_uuid='{}' order by id desc limit 2)a
-join nearline_recommend.recommend_product_record b on a.recall_batch_uuid=b.recall_batch_uuid
+(select id,recall_batch_uuid,event_type from nearline_recommend_test.recommend_product_batch where user_uuid='{}' order by id desc limit 1)a
+join nearline_recommend_test.recommend_product_record b on a.recall_batch_uuid=b.recall_batch_uuid
 """
 
 cur_batch_sql="""
@@ -77,3 +77,30 @@ xyqb_recommender_system.kdsp_same_ranked_list
 back_batch_redis="""
 recommend_recall:{uuid}:{sku_no}
 """
+
+##支付图关联表,有序列表
+pay_product_redis="""pay_product_graph:{sku_no}"""
+##下单图关联表,有序列表
+order_product_redis="""
+order_product_graph:{sku_no}
+"""
+##加购图关联表,字符串
+product_redis="""
+product_graph:{sku_no}
+"""
+##ALS关联度表，有序列表
+als_correlation_redis="""
+product_correlation:{sku_no}
+"""
+
+###more-热度表（大盘热度表）
+cate_brand_heat_rank_sql="""
+select * from 
+offline_recommend.cate_brand_heat_rank 
+"""
+
+###more-用户意图得分表
+cate_brand_intention_score_sql="""
+select * from 
+offline_recommend.cate_brand_intention_score 
+"""

recommend/searchTopic.py

@@ -23,8 +23,10 @@ def getseed(top10_sku):
     df['cid3']=df['cid3'].astype('string')
     cid3=df.groupby(by=['cid3']).groups.__repr__()#['cid3']#.max()
     brandname=df.groupby(by=['brand_name']).groups.__repr__()
+    brandid=df.groupby(by=['brand_id']).groups.__repr__()
     result['cid3']=maxdict(**json.loads(cid3.replace("'",'"')))
     result['brand_name']=maxdict(**json.loads(brandname.replace("'",'"')))
+    result['brand_id']=maxdict(**json.loads(brandid.replace("'",'"')))
     result['avg_price']=df['price'].mean()
     result['skuinfo']=df.to_dict(orient='records')
     return result

tools/listOperation.py

@@ -27,6 +27,12 @@ def removeRepeat(a):
     temp.sort(key=a.index)
     return temp
 
+def subtraction(a,b):
+    if a and  b:
+        return list(set(a).difference(set(b)))
+    return b if b else a
+
+
 def mergelist(a):
     '''
     :param a:list
@@ -52,7 +58,19 @@ def listGroup(tt,size=3):
             break
     return temp
 
+def listSubgroup(tt,size=6):
+    temp=[]
+    for i in range(0,len(tt),size):
+        print(tt[i:i+size])
+        temp.append(tt[i:i+size])
+    return temp
+
 
 if __name__ == '__main__':
     a=[1,2,4,3,5,6,7,8]
-    print(listGroup(a))
+    #print(listGroup(a))
+
+    #a=[[1,2],[2,3]]
+    #print(mergelist(a))
+
+    print(listSubgroup(a,6))