Skip to content
Merged
Changes from 10 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 14 additions & 2 deletions sql-statements/sql-statement-modify-column.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,12 +8,24 @@ aliases: ['/docs-cn/dev/sql-statements/sql-statement-modify-column/','/docs-cn/d

`ALTER TABLE .. MODIFY COLUMN` 语句用于修改已有表上的列,包括列的数据类型和属性。若要同时重命名,可改用 [`CHANGE COLUMN`](/sql-statements/sql-statement-change-column.md) 语句。

从 v5.1.0 版本起,TiDB 开始支持 Reorg 类型变更,包括但不限于:
从 v5.1.0 起,TiDB 支持需要 Reorg-Data 的列类型变更。这类变更在执行过程中会重建表中现有的所有数据,包括读取原始数据、按照新的数据格式进行处理,以及重新写入表中。由于需要处理全表数据,Reorg-Data 操作通常较为耗时,其执行时间与表中的数据量成正比。
Copy link
Collaborator

@qiancai qiancai Jan 6, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

#reorg-data-change 里只有一个 INT -> VARCHAR 的示例,不能成为 Reorg-Data 的定义,或者可以把 “需要 Reorg-Data 的列类型变更” 修改为 “需要 Reorg-Data(重新表数据)的列类型变更"?


常见的需要 Reorg-Data 的列类型变更包括(但不限于):


- 从 `VARCHAR` 转换为 `BIGINT`
- `DECIMAL` 精度修改
- 从 `VARCHAR(10)` 到 `VARCHAR(5)` 的长度压缩

从 v8.5.5 和 v9.0.0 起,针对部分原本需要 Reorg-Data 的列类型变更,TiDB 优化了执行效率。如果当前会话的 [SQL 模式](/sql-mode.md)为严格模式(即 `sql_mode` 值包含 `STRICT_TRANS_TABLES` 或 `STRICT_ALL_TABLES`),TiDB 在进行以下类型变更时,会预先检查类型转换过程中是否存在数据截断风险,若不存在数据截断风险,TiDB 将不再进行表数据重建,仅重建受影响的索引:

- 整数类型之间的变更(例如 `BIGINT` 到 `INT`)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这个场景下,是不是不需要重建索引?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

不需要。之前需要

- 字符串类型之间的变更(例如 `VARCHAR(200)` 到 `VARCHAR(100)`)

> **注意:**
>
> 当从 `VARCHAR` 转换为 `CHAR` 时,要求所有原数据的末尾均不包含空格;若存在不满足该条件的数据,TiDB 仍会执行 Reorg-Data,以确保转换后的数据符合 `CHAR` 类型的填充规则。

## 语法图

```ebnf+diagram
Expand Down Expand Up @@ -168,7 +180,7 @@ CREATE TABLE `t1` (
> ERROR 1406 (22001): Data Too Long, field len 4, data len 5
> ```
>
> - 由于和 Async Commit 功能兼容,DDL 在开始进入到 Reorg Data 前会有一定时间(约 2.5s)的等待处理:
> - 由于和 Async Commit 功能兼容,在关闭[元数据锁](/metadata-lock.md)的情况下,DDL 在开始进入到 Reorg-Data 前会有一定时间(约 2.5 秒)的等待处理:
>
> ```
> Query OK, 0 rows affected (2.52 sec)
Expand Down