Thanks for your comment.

> 8.3. Character Types:
> The name type exists only for the storage of identifiers in the internal 
> system catalogs

I didn't know the policy about data types. Thank you.
But I think the column names should match pg_replication_slots.
The attached patch changes only the column names and macros.

Regards,
Noriyoshi Shinoda

-----Original Message-----
From: Masahiko Sawada [mailto:[email protected]] 
Sent: Tuesday, October 13, 2020 9:11 AM
To: Shinoda, Noriyoshi (PN Japan A&PS Delivery) <[email protected]>
Cc: Amit Kapila <[email protected]>; Dilip Kumar <[email protected]>; 
Magnus Hagander <[email protected]>; Tomas Vondra 
<[email protected]>; PostgreSQL Hackers 
<[email protected]>; Ajin Cherian <[email protected]>
Subject: Re: Resetting spilled txn statistics in pg_stat_replication

(Please avoid top-posting)

On Mon, 12 Oct 2020 at 23:45, Shinoda, Noriyoshi (PN Japan A&PS
Delivery) <[email protected]> wrote:
>
> Sawada-san, Thank you your comments.
>
> The attached patch reflects the comment.
> I also made a fix for the regression test.
>
> Regards,
> Noriyoshi Shinoda
>
> -----Original Message-----
> From: Masahiko Sawada [mailto:[email protected]]
> Sent: Monday, October 12, 2020 8:12 PM
> To: Shinoda, Noriyoshi (PN Japan A&PS Delivery) 
> <[email protected]>
> Cc: Amit Kapila <[email protected]>; Dilip Kumar 
> <[email protected]>; Magnus Hagander <[email protected]>; Tomas 
> Vondra <[email protected]>; PostgreSQL Hackers 
> <[email protected]>; Ajin Cherian <[email protected]>
> Subject: Re: Resetting spilled txn statistics in pg_stat_replication
>
> On Mon, 12 Oct 2020 at 18:29, Shinoda, Noriyoshi (PN Japan A&PS
> Delivery) <[email protected]> wrote:
> >
> > Hi, thank you for the awesome feature.
> >
>
> Thank you for reporting!
>
> > As it may have been discussed, I think the 'name' column in 
> > pg_stat_replication_slots is more consistent with the column name and data 
> > type matched to the pg_replication_slots catalog.
> > The attached patch changes the name and data type of the 'name' column to 
> > slot_name and 'name' type, respectively.
>
> It seems a good idea to me. In other system views, we use the name data type 
> for object name. When I wrote the first patch, I borrowed the code for 
> pg_stat_slru which uses text data for the name but I think it's an oversight.

Hmm, my above observation is wrong. All other statistics use text data type and 
internally use char[NAMEDATALEN]. So I think renaming to 'slot_name' would be a 
good idea but probably we don’t need to change the internally used data type. 
For the data type of slot_name of pg_stat_replication_slots view, given that 
the doc says the following[1], I think we can keep it too as this view is not a 
system catalog. What do you think?

8.3. Character Types:
The name type exists only for the storage of identifiers in the internal system 
catalogs

[1] https://www.postgresql.org/docs/devel/datatype-character.html 

Regards,

--
Masahiko Sawada            http://www.2ndQuadrant.com/ 
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Attachment: pg_stat_replication_slots_v3.diff
Description: pg_stat_replication_slots_v3.diff

Reply via email to