Re: [16Beta1][doc] pgstat: Track time of the last scan of a relation - Mailing list pgsql-hackers

From David Rowley
Subject Re: [16Beta1][doc] pgstat: Track time of the last scan of a relation
Date
Msg-id CAApHDvr2WOUdnvJemheTXtONC6KdWrBwSxLCN8b687+vLFbKyQ@mail.gmail.com
Whole thread Raw
In response to [16Beta1][doc] pgstat: Track time of the last scan of a relation  ("Shinoda, Noriyoshi (PN Japan FSIP)" <noriyoshi.shinoda@hpe.com>)
Responses RE: [16Beta1][doc] pgstat: Track time of the last scan of a relation
List pgsql-hackers
On Wed, 31 May 2023 at 15:57, Shinoda, Noriyoshi (PN Japan FSIP)
<noriyoshi.shinoda@hpe.com> wrote:
> According to the documentation [2], the data type of the columns added to these views is 'timestamptz'.
> However, columns of the same data type in pg_stat_all_tables.last_vacuum, last_analyze and other tables are unified
to'timestamp with time zone'. The attached patch changes the data type of the added column from timestamptz to
timestampwith time zone.
 

I agree that it would be good to make those consistently use timestamp
with time zone for all columns of that type in the docs for
pg_stat_all_tables.

More generally, it might be good if we did it for the entire docs:

doc $ git grep "<type>timestamptz</type>" | wc -l
17
doc $ git grep "<type>timestamp with time zone</type>" | wc -l
74

Clearly "timestamp with time zone" is much more commonly used.

The bar is probably set a bit higher for changing the
longer-established ones, however.

David



pgsql-hackers by date:

Previous
From: David Rowley
Date:
Subject: Re: PG 16 draft release notes ready
Next
From: Richard Guo
Date:
Subject: Re: ERROR: wrong varnullingrels (b 3) (expected (b)) for Var 2/1