Question

我正在尝试执行一个查询，这个查询会让我知道我的表格的结构如下：

RunnerID,   State,   Timestamp_race
R1          start     2017-04-11 12:15:15.722415
R2          start     2017-04-11 13:15:15.722415
R1          finish    2017-04-11 15:15:15.722415
R1          finish    2017-04-11 17:15:15.722415

跑步者在不同时间开始并在不同时间结束，想法是计算整个比赛的平均时间

Answer 1

您可以使用extract(epoch)获取每场比赛的秒数，然后取平均值：

select avg(seconds) as avg_seconds
from (select runnerid, extract(epoch from max(Timestamp_race) - min(Timestamp_race)) as seconds
      from t
      group by runnerid
     ) r;

这假设最小时间是开始，最大值是结束。

编辑：

如果您希望将平均值作为间隔，则可以执行以下操作：

select avg(diff) as avg_diff
from (select runnerid, (max(Timestamp_race) - min(Timestamp_race)) as diff
      from t
      group by runnerid
     ) r;

Answer 2

这是一种更好的格式：

select to_char(avg(seconds),'DD HH24:MI:SS.MS') as avg_time
from (select runnerid, max(Timestamp_race) - min(Timestamp_race) as seconds
      from race t
      group by runnerid
     ) r;

|        avg_time |
|-----------------|
| 00 03:30:00.000 |

如何计算Postgresql的平均时间

2 个答案: