Re: SQL-standard function body

On 31.03.21 12:12, Julien Rouhaud wrote:
> On Tue, Mar 23, 2021 at 11:28:55PM -0500, Jaime Casanova wrote:
>> On Fri, Mar 19, 2021 at 8:49 AM Peter Eisentraut
>> <peter(dot)eisentraut(at)2ndquadrant(dot)com> wrote:
>>>
>>> Right. Here is a new patch with that fix added and a small conflict
>>> resolved.
>>
>> Great.
>>
>> It seems print_function_sqlbody() is not protected to avoid receiving
>> a function that hasn't a standard sql body in
>> src/backend/utils/adt/ruleutils.c:3292, but instead it has an assert
>> that gets hit with something like this:
>>
>> CREATE FUNCTION foo() RETURNS int LANGUAGE SQL AS $$ SELECT 1 $$;
>> SELECT pg_get_function_sqlbody('foo'::regproc);

fixed

> It would also be good to add a regression test checking that we can't define a
> function with both a prosrc and a prosqlbody.

done

> @@ -76,6 +77,7 @@ ProcedureCreate(const char *procedureName,
> Oid languageValidator,
> const char *prosrc,
> const char *probin,
> + Node *prosqlbody,
> char prokind,
> bool security_definer,
> bool isLeakProof,
> @@ -119,8 +121,6 @@ ProcedureCreate(const char *procedureName,
> /*
> * sanity checks
> */
> - Assert(PointerIsValid(prosrc));
> -
> parameterCount = parameterTypes->dim1;
>
>
> Shouldn't we still assert that we either have a valid procsrc or valid
> prosqlbody?

fixed

New patch attached.