From: | Robert Haas <robertmhaas(at)gmail(dot)com> |
---|---|
To: | Michael Paquier <michael(dot)paquier(at)gmail(dot)com> |
Cc: | PostgreSQL mailing lists <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Refactoring of replication commands using printsimple |
Date: | 2017-01-31 14:59:35 |
Message-ID: | CA+TgmoYnrAwZtuesC4dxyBTvsSAGzNzz_OZcQ+8gJ2DEr1ED3g@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, Jan 31, 2017 at 12:19 AM, Michael Paquier
<michael(dot)paquier(at)gmail(dot)com> wrote:
> This is a follow up of the refactoring that has been discussed in the
> thread to increase the default size of WAL segments:
> https://www.postgresql.org/message-id/CAB7nPqQ4hyNrLq+W1JrrYVSySOxuQa40PYb2Uw5UQkKaG4hong@mail.gmail.com
>
> The discussion has resulted in the creation of a84069d9 that has
> introduced a new DestReceiver method called printsimple that does not
> need any catalog access. After some investigation, I have noticed that
> a couple of messages used in the replication protocol could be
> refactored as well:
> - IDENTIFY_SYSTEM
> - TIMELINE_HISTORY
> - CREATE_REPLICATION_SLOT
> This results in the following code reduction:
> 3 files changed, 115 insertions(+), 162 deletions(-)
>
> A commit fest entry has been created:
> https://commitfest.postgresql.org/13/978/
Sorry, I have a little more nitpicking. How about having
printsimple() use pq_sendcountedtext() instead of pq_sendint()
followed by pq_sendbytes(), as it does for TEXTOID?
Other than that, this looks fine to me now.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Haas | 2017-01-31 15:02:37 | Re: Parallel tuplesort (for parallel B-Tree index creation) |
Previous Message | Alvaro Herrera | 2017-01-31 14:07:56 | Re: Patch: Write Amplification Reduction Method (WARM) |