Re: Add more information_schema columns - Mailing list pgsql-hackers

From Michael Paquier
Subject Re: Add more information_schema columns
Date
Msg-id 20180206071549.GB74355@paquier.xyz
Whole thread Raw
In response to Add more information_schema columns  (Peter Eisentraut <peter.eisentraut@2ndquadrant.com>)
Responses Re: Add more information_schema columns
Re: Add more information_schema columns
List pgsql-hackers
On Mon, Feb 05, 2018 at 08:59:31PM -0500, Peter Eisentraut wrote:
> Here is a patch that fills in a few more information schema columns, in
> particular those related to the trigger transition tables feature.

It is unfortunate that this cannot be backpatched.  Here are few
comments, the logic and theh definitions look correct to me.

> -           CAST(null AS cardinal_number) AS action_order,
>             -- XXX strange hacks follow
> +           CAST(rank() OVER (PARTITION BY n.oid, c.oid, em.num, (t.tgtype & 1 & 66) ORDER BY t.tgname) AS
cardinal_number)AS action_order,
 

Better to use parenthesis for (t.tgtype & 1 & 66) perhaps? You may want
to comment that this is to filter per row-statement first, and then with
after/before/instead of, which are what the 1 and the 66 are for.

> -           CAST(null AS sql_identifier) AS action_reference_old_table,
> -           CAST(null AS sql_identifier) AS action_reference_new_table,
> +           CAST(tgoldtable AS sql_identifier) AS action_reference_old_table,
> +           CAST(tgnewtable AS sql_identifier) AS action_reference_new_table,

> +SELECT trigger_name, event_manipulation, event_object_schema,
> event_object_table, action_order, action_condition,
> action_orientation, action_timing, action_reference_old_table,
> action_reference_new_table FROM information_schema.triggers ORDER BY
> 1, 2;

Writing those SQL queries across multiple lines would make them easier
to read...
--
Michael

Attachment

pgsql-hackers by date:

Previous
From: Rajkumar Raghuwanshi
Date:
Subject: Re: Query running for very long time (server hanged) with parallel append
Next
From: Kyotaro HORIGUCHI
Date:
Subject: Re: [HACKERS] Vacuum: allow usage of more than 1GB of work mem