Question

我需要计算列中重复项的数量，这样做完成后会这样：

SELECT CorrelationId, Count(CorrelationId) as total
FROM VoipDetailsView
WHERE ResponseCode = 200
GROUP BY CorrelationId

但是，当我尝试将其他列添加到结果

时

SELECT SessionIdTime, CorrelationId, Count(CorrelationId) as total
FROM VoipDetailsView
WHERE ResponseCode = 200
GROUP BY CorrelationId, SessionIdTime
ORDER BY SessionIdTime

即使（或可能因为）现在显示重复值，所有计数现在都是0或1。我不介意重复，但我需要最后一列包含每个选定的重复CorrelationIds的总数。（例如，如果值存在两次，则这两行的计数将为2）

我需要做什么样的查询？

Answer 1

如SessionIdTime

在min()列上使用聚合函数怎么样？

SELECT min(SessionIdTime), CorrelationId, Count(CorrelationId) as total
FROM VoipDetailsView
WHERE ResponseCode = 200
GROUP BY CorrelationId
ORDER BY min(SessionIdTime)

Answer 2

您的第二个查询为您提供特定SessionIdTime中重复项CorrelationId的数量，这是您想要的吗？ SessionIdTime是什么意思？

是的，你可以这样做：

SELECT CorrelationId, Count(*) as total
FROM VoipDetailsView
WHERE ResponseCode = 200
GROUP BY CorrelationId;

它有点优化

Answer 3

您需要将CorrelationId的聚合分成单独的查询，并将其与您的表连接以获取其他列，例如：

SELECT v.SessionIdTime, v.CorrelationId, total
From VoipDetailsView v inner join
(SELECT CorrelationId, Count(CorrelationId) as total
FROM VoipDetailsView
WHERE ResponseCode = 200
GROUP BY CorrelationId) agg
on v.CorrelationId = agg.CorrelationId

将列添加到group by子句会扰乱计数

3 个答案: