ADD INDEX

ALTER TABLE.. ADD INDEX 语句用于在已有表中添加一个索引。在 TiDB 中，ADD INDEX 为在线操作，不会阻塞表中的数据读写。

警告：

  • 在升级 TiDB 集群的过程中，请勿执行 DDL 语句，否则可能会出现行为未定义的问题。
  • 集群中有 DDL 语句正在被执行时（通常为 ADD INDEX 和列类型变更等耗时较久的 DDL 语句），请勿进行升级操作。在升级前，建议使用 ADMIN SHOW DDL 命令查看集群中是否有正在进行的 DDL Job。如需升级，请等待 DDL 执行完成或使用 ADMIN CANCEL DDL 命令取消该 DDL Job 后再进行升级。

语法图

1. AlterTableStmt ::=
2. 'ALTER' IgnoreOptional 'TABLE' TableName ( AlterTableSpecListOpt AlterTablePartitionOpt | 'ANALYZE' 'PARTITION' PartitionNameList ( 'INDEX' IndexNameList )? AnalyzeOptionListOpt )

4. AlterTableSpec ::=
5. TableOptionList
6. |   'SET' 'TIFLASH' 'REPLICA' LengthNum LocationLabelList
7. |   'CONVERT' 'TO' CharsetKw ( CharsetName | 'DEFAULT' ) OptCollate
8. |   'ADD' ( ColumnKeywordOpt IfNotExists ( ColumnDef ColumnPosition | '(' TableElementList ')' ) | Constraint | 'PARTITION' IfNotExists NoWriteToBinLogAliasOpt ( PartitionDefinitionListOpt | 'PARTITIONS' NUM ) )
9. |   ( ( 'CHECK' | 'TRUNCATE' ) 'PARTITION' | ( 'OPTIMIZE' | 'REPAIR' | 'REBUILD' ) 'PARTITION' NoWriteToBinLogAliasOpt ) AllOrPartitionNameList
10. |   'COALESCE' 'PARTITION' NoWriteToBinLogAliasOpt NUM
11. |   'DROP' ( ColumnKeywordOpt IfExists ColumnName RestrictOrCascadeOpt | 'PRIMARY' 'KEY' | 'PARTITION' IfExists PartitionNameList | ( KeyOrIndex IfExists | 'CHECK' ) Identifier | 'FOREIGN' 'KEY' IfExists Symbol )
12. |   'EXCHANGE' 'PARTITION' Identifier 'WITH' 'TABLE' TableName WithValidationOpt
13. |   ( 'IMPORT' | 'DISCARD' ) ( 'PARTITION' AllOrPartitionNameList )? 'TABLESPACE'
14. |   'REORGANIZE' 'PARTITION' NoWriteToBinLogAliasOpt ReorganizePartitionRuleOpt
15. |   'ORDER' 'BY' AlterOrderItem ( ',' AlterOrderItem )*
16. |   ( 'DISABLE' | 'ENABLE' ) 'KEYS'
17. |   ( 'MODIFY' ColumnKeywordOpt IfExists | 'CHANGE' ColumnKeywordOpt IfExists ColumnName ) ColumnDef ColumnPosition
18. |   'ALTER' ( ColumnKeywordOpt ColumnName ( 'SET' 'DEFAULT' ( SignedLiteral | '(' Expression ')' ) | 'DROP' 'DEFAULT' ) | 'CHECK' Identifier EnforcedOrNot | 'INDEX' Identifier IndexInvisible )
19. |   'RENAME' ( ( 'COLUMN' | KeyOrIndex ) Identifier 'TO' Identifier | ( 'TO' | '='? | 'AS' ) TableName )
20. |   LockClause
21. |   AlgorithmClause
22. |   'FORCE'
23. |   ( 'WITH' | 'WITHOUT' ) 'VALIDATION'
24. |   'SECONDARY_LOAD'
25. |   'SECONDARY_UNLOAD'

27. Constraint ::=
28. ConstraintKeywordOpt ConstraintElem

30. ConstraintKeywordOpt ::=
31. ( 'CONSTRAINT' Symbol? )?

33. ConstraintElem ::=
34. ( ( 'PRIMARY' 'KEY' | KeyOrIndex IfNotExists | 'UNIQUE' KeyOrIndexOpt ) IndexNameAndTypeOpt | 'FULLTEXT' KeyOrIndexOpt IndexName ) '(' IndexPartSpecificationList ')' IndexOptionList
35. |   'FOREIGN' 'KEY' IfNotExists IndexName '(' IndexPartSpecificationList ')' ReferDef
36. |   'CHECK' '(' Expression ')' EnforcedOrNotOpt

38. IndexNameAndTypeOpt ::=
39. IndexName ( 'USING' IndexTypeName )?
40. |   Identifier 'TYPE' IndexTypeName

42. IndexPartSpecificationList ::=
43. IndexPartSpecification ( ',' IndexPartSpecification )*

45. IndexPartSpecification ::=
46. ( ColumnName OptFieldLen | '(' Expression ')' ) Order

48. IndexOptionList ::=
49. IndexOption*

51. IndexOption ::=
52. 'KEY_BLOCK_SIZE' '='? LengthNum
53. |   IndexType
54. |   'WITH' 'PARSER' Identifier
55. |   'COMMENT' stringLit
56. |   IndexInvisible

58. KeyOrIndex ::=
59. 'KEY'
60. |   'INDEX'

62. IndexKeyTypeOpt ::=
63. ( 'UNIQUE' | 'SPATIAL' | 'FULLTEXT' )?

65. IndexInvisible ::=
66. 'VISIBLE'
67. |   'INVISIBLE'

69. IndexTypeName ::=
70. 'BTREE'
71. |   'HASH'
72. |   'RTREE'

示例

1. CREATE TABLE t1 (id INT NOT NULL PRIMARY KEY AUTO_INCREMENT, c1 INT NOT NULL);

1. Query OK, 0 rows affected (0.11 sec)

1. INSERT INTO t1 (c1) VALUES (1),(2),(3),(4),(5);

1. Query OK, 5 rows affected (0.03 sec)
2. Records: 5  Duplicates: 0  Warnings: 0

1. EXPLAIN SELECT * FROM t1 WHERE c1 = 3;

1. +-------------------------+----------+-----------+---------------+--------------------------------+
2. | id                      | estRows  | task      | access object | operator info                  |
3. +-------------------------+----------+-----------+---------------+--------------------------------+
4. | TableReader_7           | 10.00    | root      |               | data:Selection_6               |
5. | └─Selection_6           | 10.00    | cop[tikv] |               | eq(test.t1.c1, 3)              |
6. |   └─TableFullScan_5     | 10000.00 | cop[tikv] | table:t1      | keep order:false, stats:pseudo |
7. +-------------------------+----------+-----------+---------------+--------------------------------+
8. 3 rows in set (0.00 sec)

1. ALTER TABLE t1 ADD INDEX (c1);

1. Query OK, 0 rows affected (0.30 sec)

1. EXPLAIN SELECT * FROM t1 WHERE c1 = 3;

1. +------------------------+---------+-----------+------------------------+---------------------------------------------+
2. | id                     | estRows | task      | access object          | operator info                               |
3. +------------------------+---------+-----------+------------------------+---------------------------------------------+
4. | IndexReader_6          | 0.01    | root      |                        | index:IndexRangeScan_5                      |
5. | └─IndexRangeScan_5     | 0.01    | cop[tikv] | table:t1, index:c1(c1) | range:[3,3], keep order:false, stats:pseudo |
6. +------------------------+---------+-----------+------------------------+---------------------------------------------+
7. 2 rows in set (0.00 sec)

MySQL 兼容性

  • 不支持 FULLTEXT，HASH 和 SPATIAL 索引。
  • 不支持降序索引（类似于 MySQL 5.7）。
  • 目前尚不支持在一条中同时添加多个索引。
  • 无法向表中添加 CLUSTERED 类型的 PRIMARY KEY。要了解关于 CLUSTERED 主键的详细信息，请参考聚簇索引。

另请参阅