Re: Improving the notation for ecpg.addons rules

Michael Paquier <michael@paquier.xyz>

From: Michael Paquier <michael@paquier.xyz>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-hackers@lists.postgresql.org
Date: 2024-08-23T00:22:22Z
Lists: pgsql-hackers
On Tue, Aug 20, 2024 at 02:33:23PM -0400, Tom Lane wrote:
> I wrote:
>> Yeah, I was wondering about that.  I wouldn't do it exactly like
>> that, but with a check that the entry gets matched somewhere.
> 
> Here's a patch for that (again based on the other patch series).
> This did not turn up anything interesting, but it's probably
> worth keeping.

Okay, I see where you're going with this one.  It does not seem like
this is going to cost much in long-term maintenance while catching
unfortunate issues, so +1 from me.

The patch does not apply on HEAD due to the dependency with the other
things you are proposing, and I would have hardcoded failures to check
that the reports are correct, but that looks neat on read.
--
Michael

Commits

  1. ecpg: invent a saner syntax for ecpg.addons entries.

  2. ecpg: add cross-checks to parse.pl for usage of internal tables.