Question

我有3个查询

select uid, sum(call_duration) as cTime from privacy_call_history 
where event_date >= '2015-11-20 01:00:00' and event_date < '2015-11-20 01:00:00' + interval 1 hour
group by uid having sum(call_duration) > 0;

select uid, sum(run_time) as rTime from app_finish_history
where event_date >= '2015-11-20 01:00:00' and event_date < '2015-11-20 01:00:00' + interval 1 hour
group by uid having sum(run_time) > 0;

select uid, sum(chat_time) as wcTime, sum(msg_avg_time) as maTime from whisper_chat_history
where event_date >= '2015-11-20 01:00:00' and event_date < '2015-11-20 01:00:00' + interval 1 hour
group by uid having sum(chat_time) > 0 and sum(msg_avg_time) > 0;

我想将它们组合成一个结果集，如下所示：

＆＃13;

uid	     cTime		      rTime		     wcTime	            maTime
1		2			0			0			0
2		0			3			0			0
3		0			0			4			0
4		0			0			0			5

＆＃13;

如何进行查询？

Answer 1

您可以将查询作为子查询合并到一个结果集中。但是，您将需要用户表，其中包含完整的uid列表，或者您需要在mysql中模拟完整的外部联接。我会使用用户表（或其他任何名称）。

select user.uid, ifnull(t1.cTime,0) cTime, ifnull(t2.rTime,0) rTime, ifnull(t3.wcTime,0) wcTime, ifnull(t3.maTime,0) maTime
from user left join
(select uid, sum(call_duration) as cTime from privacy_call_history 
where event_date >= '2015-11-20 01:00:00' and event_date < '2015-11-20 01:00:00' + interval 1 hour
group by uid having sum(call_duration) > 0) t1 on user.uid=t1.uid
left join
(select uid, sum(run_time) as rTime from app_finish_history
where event_date >= '2015-11-20 01:00:00' and event_date < '2015-11-20 01:00:00' + interval 1 hour
group by uid having sum(run_time) > 0) t2 on user.uid=t2.uid
left join
(select uid, sum(chat_time) as wcTime, sum(msg_avg_time) as maTime from whisper_chat_history
where event_date >= '2015-11-20 01:00:00' and event_date < '2015-11-20 01:00:00' + interval 1 hour
group by uid having sum(chat_time) > 0 and sum(msg_avg_time) > 0) t3 on user.uid=t3.uid

如果你想过滤掉那些没有出现在任何子选择中的用户，那么添加一个citeria需要至少1个非空值的地方：coalesce（cTime，rTime，wcTime，maTime）不为null

如何从mysql中多个不同的表

1 个答案: