Naming Strategy
When mapping our entities to database tables and columns, their names will be defined by naming strategy. There are 3 basic naming strategies we can choose from:
UnderscoreNamingStrategy
- default of all SQL driversMongoNamingStrategy
- default ofMongoDriver
EntityCaseNamingStrategy
- uses unchanged entity and property names
You can override this when initializing ORM. You can also provide our own naming strategy, just implement NamingStrategy
interface and provide our implementation when bootstrapping ORM:
class MyCustomNamingStrategy implements NamingStrategy {
...
}
const orm = await MikroORM.init({
...
namingStrategy: MyCustomNamingStrategy,
...
});
You can also extend
AbstractNamingStrategy
which implements one method for we -getClassName()
that is used to map entity file name to class name.
Naming Strategy in mongo driver
MongoNamingStrategy
will simply use all field names as they are defined. Collection names will be translated into lower-cased dashed form:
MyCoolEntity
will be translated into my-cool-entity
collection name.
Naming Strategy in SQL drivers
MySqlDriver
defaults to UnderscoreNamingStrategy
, which means our all our database tables and columns will be lower-cased and words divided by underscored:
CREATE TABLE `author` (
`id` int(11) unsigned NOT NULL AUTO_INCREMENT,
`created_at` datetime(3) DEFAULT NULL,
`updated_at` datetime(3) DEFAULT NULL,
`terms_accepted` tinyint(1) DEFAULT NULL,
`name` varchar(255) DEFAULT NULL,
`email` varchar(255) DEFAULT NULL,
`born` datetime DEFAULT NULL,
`favourite_book_id` int(11) DEFAULT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
NamingStrategy API
NamingStrategy.getClassName(file: string, separator?: string): string
Return a name of the class based on its file name.
NamingStrategy.classToTableName(entityName: string): string
Return a table name for an entity class.
NamingStrategy.getEntityName(tableName: string, schemaName?: string): string
Return a name of the entity class based on database table name (used in EntityGenerator
).
Default implementation ignores the schema name, but when duplicates are detected, the name will be prefixed automatically.
NamingStrategy.propertyToColumnName(propertyName: string): string
Return a column name for a property (used in EntityGenerator
).
NamingStrategy.getEnumClassName(columnName: string, tableName: string, schemaName?: string): string
Return an enum class name for a column (used in EntityGenerator
).
NamingStrategy.enumValueToEnumProperty(enumValue: string, columnName: string, tableName: string, schemaName?: string): string
Return an enum property name for an enum value (used in EntityGenerator
).
NamingStrategy.referenceColumnName(): string
Return the default reference column name.
NamingStrategy.joinColumnName(propertyName: string): string
Return a join column name for a property.
NamingStrategy.joinTableName(sourceEntity: string, targetEntity: string, propertyName: string): string
Return a join table name. This is used as default value for pivotTable
.
NamingStrategy.joinKeyColumnName(entityName: string, referencedColumnName?: string): string
Return the foreign key column name for the given parameters.
NamingStrategy.indexName(tableName: string, columns: string[], type: 'primary' | 'foreign' | 'unique' | 'index' | 'sequence'): string
Returns key/constraint name for given type. Some drivers might not support all the types (e.g. mysql and sqlite enforce the PK name).
NamingStrategy.aliasName(entityName: string, index: number): string
Returns alias name for given entity. The alias needs to be unique across the query, which is by default ensured via appended index parameter. It is optional to use it as long as we ensure it will be unique.
NamingStrategy.inverseSideName(entityName: string, propertyName: string, kind: ReferenceKind): string
Returns the name of the inverse side property. Used in the EntityGenerator
with bidirectionalRelations
option. The default implementation will vary based on the property kind:
- M:N relations will be named as
${propertyName}Inverse
(the property name is inferred from pivot table name). - Other relation kinds will use the target entity name, with first character lowercased, and
Collection
appended in case it's a 1:M collection.
This behavior changed in v6.3, before that, all the properties were named with the
Inverse
suffix as the M:N relations are now.