On Wed, 16 Nov 2011 09:17:45 +0100
Thomas Kellerer <spam_eater@gmx.net> wrote:
> Ivan Sergio Borgonovo, 16.11.2011 01:01:
> > test=# begin;
> > create or replace function process_table (
> > action TEXT, v_table_name varchar(100)
> > ) RETURNS BOOLEAN
> > AS $$
> > DECLARE
> >
> > BEGIN
> > return true;
> > END;
> > $$ LANGUAGE plpgsql;
> >
> > drop function process_table (
> > action TEXT, v_table_name varchar(100)
> > );
> > commit;
> > BEGIN
> > CREATE FUNCTION
> > DROP FUNCTION
> > COMMIT
> > test=#
> >
> > Repeat just the input parameters.
>
> You don't have to include the parameter names though
>
> drop function process_table (TEXT, varchar(100));
> is just as good and bit less typing ;)
In psql/pgadmin you've tab completion. It will complete without the
parameters name.
If you're writing more durable code generally you can just cut&paste
the creation code.
I admit I haven't spent enough time to see if I can have tab
completion inside my IDE/editor.
When you're refactoring the function most probably you'll have to
refactor the drop code too.
I tend to refactor much more frequently the number/type of parameters
rather than the names, so skipping the names is anticipating some
work that I'll seldom take advantage of.
I admit I drop functions much more frequently in psql rather than in
my IDE/editor, but still I generally have the creation code handy.
--
Ivan Sergio Borgonovo
http://www.webthatworks.it
