1. 日期格式转化（参考）

select beg_time, end_time, extract(epoch from to_timestamp(end_time,'yyyy-mm-dd-HH24-MI-SS-US'))-extract(epoch from to_timestamp(beg_time,'yyyy-mm-dd-HH24-MI-SS-US')) from cdb_all_iu_data where beg_time > '2017-09-21'

注：beg_time, end_time以TEXT形式存储，求时间差时转化为时间戳再相减得到结果（s）

2. select * from (中间结果) t

select count(*) from (select chkid, count(*) from abc_table GROUP BY chkid) t

补充：自己写的postgreSQL查询语句

我就废话不多说了，大家还是直接看代码吧~

import psycopg2 class PostgreConn(): ''' 数据库连接类 ''' def __init__(self, database, user, password, host, port): self.conn = psycopg2.connect(database=database, user=user, password=password, host=host, port=port) print('数据库连接成功') self.cur = self.conn.cursor() self.rows = None def cur(self): return self.cur() def execute(self, sql, fetchone=0): self.cur.execute(sql) if fetchone: self.rows = self.cur.fetchone() else: self.rows = self.cur.fetchall() return self.rows def close(self): self.cur.close() self.conn.close() print('数据库连接关闭') def select_sql(table, keys, conditions, isdistinct=0): ''' 生成select的sql语句 @table，查询记录的表名 @key，需要查询的字段 @conditions,插入的数据，字典 @isdistinct,查询的数据是否不重复 ''' if isdistinct: sql = 'SELECT distinct %s ' % ",".join(keys) else: sql = 'SELECT %s ' % ",".join(keys) sql += ' from %s ' % table if conditions: sql += ' WHERE %s ' % dict_str_and(conditions) return sql def dict_str_and(dictin): ''' 将字典变成，key='value' and key='value'的形式 ''' tmplist = [] for k, v in dictin.items(): tmp = "%s='%s'" % (str(k), str(v)) tmplist.append(' ' + tmp + ' ') return ' and '.join(tmplist) def fSqlResult(r,key_list): ''' :param r: 数据库fetchall的结果 :param key_list: 查询字段的keys :return: format SQL Result 格式化数据库查询的结果，转化成包含多个字典的列表格式，即((1,2),(3,4))->[{"key1":1,"key2":2},{"key1":3,"key2":4}] 返回 @dict 查询结果 ''' mlist=[] l=len(key_list) if r: for item in r: tmp={} for i in range(l): tmp[key_list[i]]=str(item[i]) mlist.append(tmp) return mlist conn = PostgreConn('settle', 'admin', 'settle8', '123.57.285.89', '5432')key_list = ['user_id']sql = select_sql('st_user', key_list, {'phone': '138****'})print(sql)r = conn.execute(sql)re = fSqlResult(r, key_list)print(re)conn.close()

以上为个人经验，希望能给大家一个参考，也希望大家多多支持。如有错误或未考虑完全的地方，望不吝赐教。