Question

我有一个SQL查询，用于计算ID在日期范围内的表中出现的次数

    SELECT *,COUNT( id) AS member_count FROM members_history 
where date_registered > '2018-09-01' 
AND date_registered < '2018-12-31' 
GROUP BY id ORDER BY last_name,first_name ASC

现在的问题是，如果一个成员同时注册了今年和下一年，则该成员又对其进行了两次计数（应该如此），但在这种情况下，我想按事件列过滤掉并说出它们是否为同一事件注册的帐户请勿将其计算两次。这有意义吗？抱歉，我是新来的。让我知道我可以如何改善。

我想我想做的是

SELECT *,
       COUNT( id) AS member_count 
FROM members_history 
where date_registered > '2018-09-01' AND 
      date_registered < '2018-12-31' AND 
      event!= event 
GROUP BY id 
ORDER BY last_name,first_name ASC

谢谢

Answer 1

您只需要在event_number组中的id上Count(Distinct...)。此外，Where子句中的event != event将始终返回false，因此您不会得到任何结果。

此外，请阅读：Why is SELECT * considered harmful?

然后，SELECT list is not in GROUP BY clause and contains nonaggregated column .... incompatible with sql_mode=only_full_group_by

尝试以下操作：

SELECT id, 
       firstname, 
       lastname, 
       COUNT(DISTINCT event_number) AS member_count 
FROM members_history 
where date_registered > '2018-09-01' AND 
      date_registered < '2018-12-31' 
GROUP BY id, firstname, lastname  
ORDER BY last_name,first_name ASC

Answer 2

您可以将DISTINCT应用于“ MemberId”列

SELECT COUNT( DISTINCT  MemberId) AS member_count 
FROM members_history 
where date_registered > '2018-09-01' AND 
      date_registered < '2018-12-31' AND 
      event!= event 
GROUP BY id

Answer 3

我猜测id是成员的ID。而您想计数事件。这表明：

select id, last_name, first_name, count(distinct event_id) AS num_events 
from members_history 
where date_registered > '2018-09-01' and date_registered < '2018-12-31' 
group by id, last_name, first_name
order by last_name, first_name asc;

请注意，我固定了group by和select，以使各列兼容。

Answer 4

我认为这是

 SELECT id,event,COUNT(*) AS member_count 
    FROM members_history 
    where date_registered > '2018-09-01' AND 
          date_registered < '2018-12-31' AND 

    GROUP BY id,event 
    having count (event)=1

SQL计数排除相似的行

4 个答案: