Re: Per-function search_path => per-function GUC settings

On 9/2/07, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> I thought about ways to include GUC settings directly into CREATE
> FUNCTION, but it seemed pretty ugly and inconsistent with the
> existing syntax.  So I'm thinking of supporting only the above
> syntaxes, meaning it'll take at least two commands to create a secure
> SECURITY DEFINER function.

There's a niceness to being able to tell Postgres everything it needs
to know about a function in the one CREATE FUNCTION command.

So if we integrated the GUC settings into CREATE FUNCTION, we'd end up
writing something like

CREATE FUNCTION foo(int) RETURNS int AS $$
...
$$LANGUAGE plpgsqlSTABLESTRICTSECURITY DEFINERRESET search_pathSET regex_flavor = 'cinnamon';

That doesn't seem especially horrible.  In what way do you feel it is
inconsistent with existing syntax?

And ... although I'll admit this is a paranoid thing to mention, if
you have to fix the search_path setting *after* creating a function as
SECURITY DEFINER, then there is necessarily a short period of time
where the function exists and is insecure.

Cheers,
BJ