基于列更改的选择SQL中的自动编号行

Question

我使用全选语句从表A中检索所有值。表示例如下：

+---+----+---+
| a | 23 | X | 
+---+----+---+
| a | 23 | Y |
+---+----+---+
| a | 24 | X |
+---+----+---+
| a | 24 | Y |
+---+----+---+
| b | 24 | X |
+---+----+---+
| b | 24 | Y |
+---+----+---+
| b | 25 | X |
+---+----+---+
| b | 25 | Y |
+---+----+---+
| b | 25 | Z | 
+---+----+---+

出于此查询稍后的目的，我想为第1列和第2列的每个唯一组合都有一个记录号。例如：

+---+----+---+---+
| a | 23 | X | 1 |
+---+----+---+---+
| a | 23 | Y | 2 |
+---+----+---+---+
| a | 24 | X | 1 |
+---+----+---+---+
| a | 24 | Y | 2 |
+---+----+---+---+
| b | 24 | X | 1 |
+---+----+---+---+
| b | 24 | Y | 2 |
+---+----+---+---+
| b | 25 | X | 1 |
+---+----+---+---+
| b | 25 | Y | 2 |
+---+----+---+---+
| b | 25 | Z | 3 |
+---+----+---+---+

这可能与SQL有关，怎么做？

Answer 1

我相信您所需要的只是

选择不同ROW_NUMBER（）OVER（ORDER BY Col1 ASC，Col2 ASC）AS Row＃，Col1，Col2从表1

Answer 2

对问题的描述将使用dense_rank()：

select t.*, dense_rank() over (order by col1, col2)
from t;

您的样本数据显示dense_rank()和partitition by：

select t.*,
       dense_rank() over (partition by col1, col2 order by col3) as seqnum
from t;