CREATE AGGREGATE name ( BASETYPE = input_data_type, SFUNC = sfunc, STYPE = state_type [ , FINALFUNC = ffunc ] [ , INITCOND = initial_condition ] )
The name of an aggregate function to create.
The input data type on which this aggregate function operates. This can be specified as ANY for an aggregate that does not examine its input values (an example is count(*)).
The name of the state transition function to be called for each input data value. This is normally a function of two arguments, the first being of type state_type and the second of type input_data_type. Alternatively, for an aggregate that does not examine its input values, the function takes just one argument of type state_type. In either case the function must return a value of type state_type. This function takes the current state value and the current input data item, and returns the next state value.
The data type for the aggregate's state value.
The name of the final function called to compute the aggregate's result after all input data has been traversed. The function must take a single argument of type state_type. The output data type of the aggregate is defined as the return type of this function. If ffunc is not specified, then the ending state value is used as the aggregate's result, and the output type is state_type.
The initial setting for the state value. This must be a literal constant in the form accepted for the data type state_type. If not specified, the state value starts out NULL.
CREATE AGGREGATE allows a user or programmer to extend Postgres functionality by defining new aggregate functions. Some aggregate functions for base types such as min(integer) and avg(double precision) are already provided in the base distribution. If one defines new types or needs an aggregate function not already provided, then CREATE AGGREGATE can be used to provide the desired features.
An aggregate function is identified by its name and input data type. Two aggregates can have the same name if they operate on different input types. To avoid confusion, do not make an ordinary function of the same name and input data type as an aggregate.
An aggregate function is made from one or two ordinary functions: a state transition function sfunc, and an optional final calculation function ffunc. These are used as follows:
sfunc( internal-state, next-data-item ) ---> next-internal-state ffunc( internal-state ) ---> aggregate-value
Postgres creates a temporary variable of data type stype to hold the current internal state of the aggregate. At each input data item, the state transition function is invoked to calculate a new internal state value. After all the data has been processed, the final function is invoked once to calculate the aggregate's output value. If there is no final function then the ending state value is returned as-is.
An aggregate function may provide an initial condition, that is, an initial value for the internal state value. This is specified and stored in the database as a field of type text, but it must be a valid external representation of a constant of the state value data type. If it is not supplied then the state value starts out NULL.
If the state transition function is declared "strict" in pg_proc, then it cannot be called with NULL inputs. With such a transition function, aggregate execution behaves as follows. NULL input values are ignored (the function is not called and the previous state value is retained). If the initial state value is NULL, then the first non-NULL input value replaces the state value, and the transition function is invoked beginning with the second non-NULL input value. This is handy for implementing aggregates like max. Note that this behavior is only available when state_type is the same as input_data_type. When these types are different, you must supply a non-NULL initial condition or use a non-strict transition function.
If the state transition function is not strict, then it will be called unconditionally at each input value, and must deal with NULL inputs and NULL transition values for itself. This allows the aggregate author to have full control over the aggregate's handling of NULLs.
If the final function is declared "strict", then it will not be called when the ending state value is NULL; instead a NULL result will be output automatically. (Of course this is just the normal behavior of strict functions.) In any case the final function has the option of returning NULL. For example, the final function for avg returns NULL when it sees there were zero input tuples.
Refer to the chapter on aggregate functions in the PostgreSQL Programmer's Guide for complete examples of usage.