Postgresql trigger function with parameters

Question

You do not need to pass the NEW and OLD as parameters to the trigger function. They are automagically available there:

http://www.postgresql.org/docs/9.1/interactive/trigger-definition.html :

The trigger function must be declared as a function taking no arguments and returning type trigger. (The trigger function receives its input through a specially-passed TriggerData structure, not in the form of ordinary function arguments.)

About the records passed to the trigger procedure, please see http://www.postgresql.org/docs/9.1/interactive/plpgsql-trigger.html :

When a PL/pgSQL function is called as a trigger, several special variables are created automatically in the top-level block. They are: [...] NEW, [...] OLD [...]

As SeldomNeedy pointed in the comment below, you can still pass and use parameters to the trigger function. You declare the function as taking no parameters, but when defining the trigger (by CREATE TRIGGER), you may add some.

They will be available for the trigger as TG_NARG (the number of such parameters), and TG_ARGV[] (an array of text values).

As Greg stated, trigger functions can take arguments, but the functions themselves cannot have declared parameters. Here's a simple example in plpgsql:

CREATE TABLE my_table ( ID SERIAL PRIMARY KEY ); -- onelined for compactness

CREATE OR REPLACE FUNCTION raise_a_notice() RETURNS TRIGGER AS
$$
DECLARE
    arg TEXT;
BEGIN
    FOREACH arg IN ARRAY TG_ARGV LOOP
        RAISE NOTICE 'Why would you pass in ''%''?',arg;
    END LOOP;
    RETURN NEW; -- in plpgsql you must return OLD, NEW, or another record of table's type
END;
$$ LANGUAGE plpgsql;

CREATE TRIGGER no_inserts_without_notices BEFORE INSERT ON my_table
FOR EACH ROW EXECUTE PROCEDURE raise_a_notice('spoiled fish','stunned parrots');

INSERT INTO my_table DEFAULT VALUES;

-- the above kicks out the following:
--
-- NOTICE:  Why would you pass in 'spoiled fish'?
-- NOTICE:  Why would you pass in 'stunned parrots'?
--

There are a few other goodies such as TG_NARGS (to know how many args you got without looping through them) discussed in the docs. There's also information there about how to get the name of the triggering table in case you have mostly-but-not-quite-shared logic for one trigger-function that spans a number of tables.

The trigger function can have parameters, but, you can't have those parameters passed like a normal function (e.g. arguments in the function definition). You can get the same result... In python you get access to the OLD and NEW data as the answer above describes. For example, I can use TD['new']['column_name'] in python to reference the new data for column_name. You also have access to the special variable TD['args']. So, if you like:

create function te() returns trigger language plpython2u as $function$
    plpy.log("argument passed 1:%s 2:%s" %(TD['args'][0], TD['args'][1], ))
$function$

create constraint trigger ta after update of ttable
for each for execute procedure te('myarg1','myarg2');

Granted, these arguments are static, but, they are useful when calling a common trigger function from multiple trigger declarations. I am pretty sure that the same variables are available for other stored procedure languages. (sorry if the code doesn't work verbatim, but, I do practice this technique, so I know you can pass arguments!).