Thread: Named parameter notation in Postgresql 9.0

Hi,


I was testing the new named parameter notation for calling functions in 9.0

http://developer.postgresql.org/pgdocs/postgres/sql-syntax-calling-funcs.html

Named parameter notation is really nice, however, it is still enforced
that all parameters after the first parameter with a default value
must also have a default value in the function declaration.

so the following function will fail to compile:

FUNCTION foo(a INTEGER DEFAULT 0, b INTEGER)


ERROR:  input parameters after one with a default value must also have defaults


As the developer documentation states (35.4.6):
---------------
Although the use of named argument notation could allow this
restriction to be relaxed,  it's still enforced so that positional
argument notation works sensibly.
---------------


One of the main benefits of named notation is that it offers clarity
when you have a large number of parameters. When a function has a
large number of parameters, then it is often desirable to logically
group the parameters together at the function declaration, instead of
having to move some of them to the end because they have a default
value.

Wouldn't it be better to relax this constraint, and instead check that
when a function is being called using positional notation, that all
parameters must have been provided until (including) the last
parameter that has no default value.


Kind regards,

Ingmar Brouns

On 4/22/10 9:27 AM, Ingmar Brouns wrote:
> Wouldn't it be better to relax this constraint, and instead check that
> when a function is being called using positional notation, that all
> parameters must have been provided until (including) the last
> parameter that has no default value.

The problem is that you can't know how the function is going to be
called in the future.  That is, regardless of how you create the proc,
it could be called either named or positionally.  Checking it at calling
time doesn't work because we need to create the "fingerprint" of a proc
when it's created, not when it's called.

So, potentially solvable but will require a lot more mucking around with
function call code to make it work.  You're welcome to hack on it.

--
                                  -- Josh Berkus
                                     PostgreSQL Experts Inc.
                                     http://www.pgexperts.com