2010/9/9 Kevin Grittner <[email protected]>:
> Tom Lane <[email protected]> wrote:
>
>> I'm with Robert: this would be a huge extra complication for a
>> remarkably small amount of benefit.
>
> This is probably heresy, but unless it's required by the standard or
> drop-dead simple to allow, I'd be fine with *not* supporting
> overloading of stored procedure names based on argument types at
> all. I can see the need for to support it for functions; I can't
> think where it would be all that useful for stored procedures. If
> unique stored procedure names were required, it seems we might be
> able to allow assignment casts on parameters, which might be more
> useful.
>
> I'm probably missing some good use case....
for example - value transformation from / to bytea
CREATE FUNCTION send(int);
CREATE FUNCTION send(text);
CREATE FUNCTION recv(int);
CREATE FUNCTION recv(text)
then you can write
BEGIN
send('ahoj');
send(10);
recv(textvar);
recv(numvar);
Regards
Pavel Stehule
>
> -Kevin
>
--
Sent via pgsql-hackers mailing list ([email protected])
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
