Re: Why are stored procedures looked on so negatively? - Mailing list pgsql-general

From Merlin Moncure
Subject Re: Why are stored procedures looked on so negatively?
Date
Msg-id CAHyXU0yRr=_k-aJj2pwETbt+yYZtpSwXeF4av_e_B-OwW28Kog@mail.gmail.com
Whole thread Raw
In response to Re: Why are stored procedures looked on so negatively?  (Chris Travers <chris.travers@gmail.com>)
Responses Re: Why are stored procedures looked on so negatively?  (Craig Ringer <craig@2ndquadrant.com>)
List pgsql-general
On Fri, Aug 2, 2013 at 1:49 AM, Chris Travers <chris.travers@gmail.com> wrote:
> Here's my $0.02
>
> Stored procedures have a bunch of problems historically.  Part of this is
> because the interface traditionally is pretty spartan, and partly because
> some people take them too far.
>
> The first issue is that if you have a stored procedure which takes 2
> arguments and you need to extend it to three, then you have to change every
> call in the calling application.  This can create a maintenance problem.
> Variadic functions help somewhat but there are limits to what a variadic
> function can do here.

This is true of most popular languages.  The other defenses are
default arguments (use very sparingly), overloading, and named
parameter arguments.  If you're writing library routines that need to
accommodate a lot of behaviors, named arguments + use of defaults is a
pretty neat way to go.

merlin


pgsql-general by date:

Previous
From: Tim Bowden
Date:
Subject: Installing 9.2 on Ubuntu from packages: what is the current recommendation?
Next
From: Adrian Klaver
Date:
Subject: Re: Add a NOT NULL column with default only during add