Interface OnExecutionGenerator

    • Method Detail

      • referenceColumnsInSql

        boolean referenceColumnsInSql​(Dialect dialect)
        Determines if the columns whose values are generated are included in the column list of the SQL insert or update statement. For example, this method should return:
        Returns:
        true if the column is included in the column list of the SQL statement.
      • writePropertyValue

        boolean writePropertyValue()
        Determines if the property values are written to JDBC as the argument of a JDBC ? parameter.
      • getReferencedColumnValues

        String[] getReferencedColumnValues​(Dialect dialect)
        A SQL expression indicating how to calculate the generated values when the mapped columns are included in the SQL statement. The SQL expressions might be:
        • function calls like current_timestamp or nextval('mysequence'), or
        • syntactic markers like default.
        Parameters:
        dialect - The SQL dialect, allowing generation of an expression in dialect-specific SQL.
        Returns:
        The column value to be used in the generated SQL statement.
      • getGeneratedIdentifierDelegate

        @Incubating
        default InsertGeneratedIdentifierDelegate getGeneratedIdentifierDelegate​(PostInsertIdentityPersister persister)
        The InsertGeneratedIdentifierDelegate used to retrieve the generated value if this object is an identifier generator.

        This is ignored by GeneratedValuesProcessor, which handles multiple generators at once. So if this object is not an identifier generator, this method is never called.

        Note that this method arguably breaks the separation of concerns between the generator and coordinating code, by specifying how the generated value should be retrieved.

        The problem solved here is: we want to obtain an insert-generated primary key. But, sadly, without already knowing the primary key, there's no completely-generic way to locate the just-inserted row to obtain it.

        We need one of the following things:

        Alternatively, if the generated id is an identity/"autoincrement" column, we can take advantage of special platform-specific functionality to retrieve it. Taking advantage of the specialness of identity columns is the job of one particular implementation: IdentityGenerator. And the need for customized behavior for identity columns is the reason why this layer-breaking method exists.

      • getUniqueKeyPropertyNames

        @Incubating
        default String[] getUniqueKeyPropertyNames​(EntityPersister persister)
        The name of a property of the entity which may be used to locate the just-inserted row containing the generated value. Of course, the columns mapped by this property should form a unique key of the entity.

        The default implementation uses the @NaturalId property, if there is one.

      • generatedOnExecution

        default boolean generatedOnExecution()
        Description copied from interface: Generator
        Determines if the property value is generated when a row is written to the database, or in Java code that executes before the row is written.
        • Generators which only implement BeforeExecutionGenerator must result false.
        • Generators which only implement OnExecutionGenerator must result true.
        • Generators which implement both subinterfaces may decide at runtime what value to return.
        Specified by:
        generatedOnExecution in interface Generator
        Returns:
        true if the value is generated by the database as a side effect of the execution of an insert or update statement, or false if it is generated in Java code before the statement is executed via JDBC.