postgres按日期（长）选择时区

Question

这个查询给出了重复但是我很难理解为什么？我在trading.delayed中每个id都有很多行，但只有一个具有完全相同的时间戳

有人能解释我的错过吗？

      SELECT t1.contract,t1.date FROM trading.delayed AS t1, 
(SELECT contract, max(date) AS date FROM trading.delayed GROUP BY contract) AS t2 
WHERE t1.date =t2.date order BY contract DESC;

结果：合同，日期

32;"2017-06-07 02:04:11.797+07"
32;"2017-06-07 02:04:14.489+07"
31;"2017-06-07 02:04:12.04+07"
30;"2017-06-07 02:03:54.182+07"
30;"2017-06-07 00:20:27.812+07"
30;"2017-06-07 02:03:51.177+07"
29;"2017-06-07 00:20:27.812+07"
28;"2017-06-07 01:45:53.129+07"
27;"2017-06-07 01:58:02.974+07"

Answer 1

这不是答案，但此版本有效

SELECT t1.contract,close 
FROM trading.delayed AS t1, 
    ( 
      SELECT max(id) as id, 
             contract, 
             max(date) AS date 
      FROM trading.delayed 
      GROUP BY contract
   ) AS t2 
WHERE t1.id =t2.id 
order by contract desc;