需求:
我这边要实现每天点击最多的n个用户,通过kylin界面写sql,不支持max,所以改为调用kylin接口来实现,代码如下:
#!/usr/bin/env python # coding=utf-8 import sqlalchemy as sa def kylin_query_show(conn_str, query_sql): res_dic = {} kylin_engine = sa.create_engine(conn_str) results = kylin_engine.execute(query_sql) for line in results: userid = str(line[0]).strip() datetime = line[1].strip() per_count = line[2] if datetime not in res_dic: res_dic[datetime] = [] res_dic[datetime].append((userid, per_count)) else: res_dic[datetime].append((userid, per_count)) return res_dic if __name__ == "__main__": conn_str = "kylin://username:password@ip:7070/database?version=v1" query_sql = "select userid, datetime, count(*) c from xxx_report group by userid, datetime" res_dic = kylin_query_show(conn_str, query_sql) tmp_list = [] for k, v in res_dic.items(): final_userid_count_list = sorted(v, key = lambda x: x[1], reverse=True) final_userid_count_list = final_userid_count_list[0:5] tmp_list.append((k, ','.join([':'.join([u_c[0], str(u_c[1])]) for u_c in final_userid_count_list]))) res = sorted(tmp_list, key = lambda x:x[0], reverse=True) for i in res: print "\t".join(i)