This topic describes the general syntax of SQL statements, including the syntax of constants, character sets, data types, SQL properties, distributed properties, and storage properties.
Constants
INT_VALUE: matches integer constants such as123.DECIMAL_VALUE: matches fixed-point constants such as123.456.STR_VALUE: matches string constants such asabc.NULL: matchesNULLconstants.STORAGE_SIZE: matches storage size constants. The default unit of integer values is bytes, and the unit of string values can be set to M or G. For example, the size can be1024or'500M'.
const_value:
INT_VALUE
| DECIMAL_VALUE
| STR_VALUE
| NULL
| STORAGE_SIZE
STORAGE_SIZE:
INT_VALUE
| 'INT_VALUE[M|G]'
Character sets
charset:
default_charset
| column_charset
default_charset:
[DEFAULT] {CHARSET | CHARACTER SET} [=] {UTF8 | UTF8MB4 | BINARY}
column_charset:
{CHARSET | CHARACTER SET} {UTF8 | UTF8MB4 | BINARY}
Collations
collate:
default_collate
| column_collate
default_collate:
[DEFAULT] COLLATE [=] {UTF8MB4_GENERAL_CI | UTF8MB4_BIN | BINARY}
column_collate:
COLLATE {UTF8MB4_GENERAL_CI | UTF8MB4_BIN | BINARY}
Data types
data_type:
TINYINT[(precision)] [UNSIGNED] [ZEROFILL]
| SMALLINT[(precision)] [UNSIGNED] [ZEROFILL]
| MEDIUMINT[(precision)] [UNSIGNED] [ZEROFILL]
| INT[(precision)] [UNSIGNED] [ZEROFILL]
| INTEGER[(precision)] [UNSIGNED] [ZEROFILL]
| BIGINT[(precision)] [UNSIGNED] [ZEROFILL]
| FLOAT[(precision, scale)] [UNSIGNED] [ZEROFILL]
| DOUBLE[(precision, scale)] [UNSIGNED] [ZEROFILL]
| DECIMAL[(precision [, scale])] [UNSIGNED] [ZEROFILL]
| NUMERIC[(precision [, scale])] [UNSIGNED] [ZEROFILL]
| DATETIME[(scale)]
| TIMESTAMP[(scale)]
| DATE
| TIME[(scale)]
| YEAR
| VARCHAR(length) column_charset column_collate
| VARBINARY(length)
| CHAR[(length)] column_charset column_collate
| BINARY[(length)]
| TINYTEXT column_charset column_collate
| TINYLOB
| TEXT[(length)] column_charset column_collate
| BLOB[(length)]
| MEDIUMTEXT column_charset column_collate
| MEDIUMBLOB
| LONGTEXT column_charset column_collate
| LONGBLOB
precision | scale | length:
INT_VALUE
SQL properties
Object name
tenant_name | pool_name | unit_name | zone_name | region_name: STR_VALUE database_name | table_name | table_alias_name | column_name| column_alias_name | partition_name | subpartition_name: STR_VALUE index_name | view_name | object_name | constraint_name | tablegroup_name: STR_VALUE outline_name | user_name: STR_VALUE table_factor: [[database_name].] table_name column_factor: [table_factor.] column_nameExpressions
expression: const_value | column_factor | operator_expression | function_expressionComments
comment: COMMENT 'comment_text' comment_text: STR_VALUE
Distributed properties
PRIMARY_ZONE: specifies the leader distribution strategy for data.primary_zone: PRIMARY_ZONE [=] zone_nameZONE_LIST: specifies the resource distribution strategy for a tenant.zone_list: ZONE_LIST [=] (zone_name [, zone_name ...])REPLICA_NUM: specifies the number of replicas of data.replica_num: REPLICA_NUM [=] INT_VALUETABLEGROUP: specifies the leader distribution strategy for multiple data replicas.tablegroup: default_tablegroup | table_tablegroup default_tablegroup: DEFAULT TABLEGROUP [=] {tablegroup_name | NULL} table_tablegroup: TABLEGROUP [=] {tablegroup_name | NULL}
Storage properties
BLOCK_SIZE: specifies the microblock size for object storage.block_size: BLOCK_SIZE [=] INT_VALUECOMPRESSION: specifies the compression algorithm for object storage.compression: COMPRESSION [=] {NONE | LZ4_1.0 | LZ0_1.0 | SNAPPY_1.0 | ZLIB_1.0}PCTFREE: specifies the macroblock reservation percentage for object storage.pctfree: PCTFREE [=] INT_VALUETABLET_SIZE: specifies the minimum tablet size of a single task in a parallel compaction.tablet_size: TABLET_SIZE [=] INT_VALUE