Re: SQL-standard function body

Some conflicts have emerged, so here is an updated patch.

I have implemented/fixed the inlining of set-returning functions written
in the new style, which was previously marked TODO in the patch.

On 2020-08-28 07:33, Peter Eisentraut wrote:
> On 2020-06-30 19:49, Peter Eisentraut wrote:
>> This adds support for writing CREATE FUNCTION and CREATE PROCEDURE
>> statements for language SQL with a function body that conforms to the
>> SQL standard and is portable to other implementations.
>
> Here is a new patch. The only significant change is that pg_dump
> support is now fixed. Per the discussion in [0], I have introduced a
> new function pg_get_function_sqlbody() that just produces the formatted
> SQL body, not the whole function definition. All the tests now pass.
> As mentioned before, more tests are probably needed, so if reviewers
> just want to play with this and find things that don't work, those could
> be put into test cases, for example.
>
> As a thought, a couple of things could probably be separated from this
> patch and considered separately:
>
> 1. making LANGUAGE SQL the default
>
> 2. the RETURN statement
>
> If reviewers think that would be sensible, I can prepare separate
> patches for those.
>
>
> [0]:
> https://www.postgresql.org/message-id/flat/9df8a3d3-13d2-116d-26ab-6a273c1ed38c%402ndquadrant.com
>

--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services