Horton

Re: generating function default settings from pg_proc.dat

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andrew Dunstan <andrew@dunslane.net>
Cc: Andres Freund <andres@anarazel.de>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2026-02-17T01:36:25Z
Lists: pgsql-hackers

Attachments

I wrote:
> After a little bit of thought, I propose the following sketch
> for what to do in the bootstrap backend:

Here is a draft patch along those lines.  I've verified that
the initial contents of pg_proc are exactly as before,
except that json[b]_strip_nulls gain the correct provolatile
value, and a few proargdefaults entries no longer involve
cast functions.

The changes in system_functions.sql and pg_proc.dat are nearly
verbatim from your 0002 patch, except that I had to adjust the
quoting conventions in some places because array_in does it
differently.

			regards, tom lane

Commits

  1. Simplify creation of built-in functions with non-default ACLs.

  2. Be more wary of false matches in initdb's replace_token().

  3. Simplify creation of built-in functions with default arguments.