Question

我有一个类似于以下内容的数据集

|---------------------|------------------|------------------|
|      RowID          |     UserID       |      Code        |
|---------------------|------------------|------------------|
|       1             |         123      |        0         |
|---------------------|------------------|------------------|
|       2             |         123      |        0         |
|---------------------|------------------|------------------|
|       3             |         123      |        50        |
|---------------------|------------------|------------------|
|       4             |         456      |        0         |
|---------------------|------------------|------------------|
|       5             |         456      |        100       |
|---------------------|------------------|------------------|

我想将每个UserID的0更新为非0代码。有人可以为此提供协助吗？

Answer 1

一个选项使用可更新的公用表表达式和窗口函数：

with cte as (
    select code, max(code) over(partition by userID) max_code
    from mytable
)
update cte set code = max_code where code = 0

Answer 2

您可以为此使用一个简单的子查询：

update MyTable set
  /* If its possible that a user might have multiple Codes which are non-zero you adjust the sub-query to return the correct one */
  Code = (select top 1 Code from MyTable T2 where Code != 0 and T2.UserID = MyTable.UserId)
where Code = 0

更新缺失值

2 个答案: