Question

我目前在这个设计中有一个表

CREATE TABLE [dbo].[test_all_DIVISION](
    [all_id] [int] NOT NULL,
    [division_id] [int] NOT NULL,
    [some_value] [tinyint] NULL,
 CONSTRAINT [PK_test_all_DIVISION] PRIMARY KEY CLUSTERED 
(
    [all_id] ASC,
    [division_id] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON, FILLFACTOR = 100) ON [PRIMARY]
) ON [PRIMARY]

其中all_id和division_id是引用其他表的外键因此，按照

的顺序插入数据时

all_id  division_id   some_value
94619   1153          24
94619   1133          25
94619   1002          22
94619   1156          23

根据division_id对其进行排序，并按以下方式插入

all_id  division_id   some_value
94619   1002          22
94619   1133          25
94619   1153          24
94619   1156          23

现在我已将该约束更改为非群集约束。像这样

ALTER TABLE [dbo].[test_all_DIVISION]
DROP CONSTRAINT [PK_test_all_DIVISION]
GO

ALTER TABLE [dbo].[test_all_DIVISION]
ADD CONSTRAINT [PK_test_all_DIVISION] PRIMARY KEY NONCLUSTERED 
(
    [all_id] ASC,
    [division_id] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON, FILLFACTOR = 100) ON [PRIMARY]
GO

仍然行为没有改变，并且表在进行表插入时仍然在排序，并且表现得像聚簇索引。所以我做错了什么。请帮忙。

Answer 1

插入堆时数据未排序。您能否澄清为什么需要按特定顺序插入数据？您可以使用订单来选择数据时控制订单。

如果必须控制插入顺序，则可以添加IDENTITY列和群集。

- 编辑 -

保证结果顺序的唯一方法是在select语句中。聚集索引只会影响数据的存储顺序。

CREATE TABLE [dbo].[test_all_DIVISION](
    [id] [int] IDENTITY(1, 1) NOT NULL,
    [all_id] [int] NOT NULL,
    [division_id] [int] NOT NULL,
    [some_value] [tinyint] NULL) ON [PRIMARY];

然后在select语句中控制顺序。

SELECT
    [all_id],
    [division_id],
    [some_value]
FROM
    [dbo].[test_all_DIVISION]
ORDER BY
    [id];

数据在插入非聚集索引表时进行排序

1 个答案: