Horton

Re: A doubt about a newly added errdetail

Alvaro Herrera <alvherre@alvh.no-ip.org>

View thread View on postgr.es
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: Amit Kapila <amit.kapila16@gmail.com>
Cc: "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>, Kyotaro Horiguchi <horikyota.ntt@gmail.com>, "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>
Date: 2022-09-27T12:42:49Z
Lists: pgsql-hackers

Attachments

While reading this code, I noticed that function expr_allowed_in_node()
has a very strange API: it doesn't have any return convention at all
other than "if we didn't modify errdetail_str then all is good".  I was
tempted to add an "Assert(*errdetail_msg == NULL)" at the start of it,
just to make sure that it is not called if a message is already set.

I think it would be much saner to inline the few lines of that function
in its sole caller, as in the attached.

-- 
Álvaro Herrera        Breisgau, Deutschland  —  https://www.EnterpriseDB.com/
"E pur si muove" (Galileo Galilei)

Commits

  1. Remove publicationcmds.c's expr_allowed_in_node as a function

  2. Improve some publication-related error messages