RE: User defined data types in Logical Replication

From: Huong Dangminh <huo-dangminh(at)ys(dot)jp(dot)nec(dot)com>
To: Petr Jelinek <petr(dot)jelinek(at)2ndquadrant(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org>
Cc: Hiroshi Yanagisawa <hir-yanagisawa(at)ut(dot)jp(dot)nec(dot)com>
Subject: RE: User defined data types in Logical Replication
Date: 2017-12-18 14:25:11
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

Hi Petr Jelineks, Sawada-san

> I think the changes make sense in terms of how it all works now.
> That said I don't think the renaming idea is a good one, the naming was
> chosen to be future proof because eventually we'll need to map types to
> local oid (and possibly more) where the local info is cached so that we
> can interpret binary representation of replicated data (which we'll add
> at some point since it's big performance boost).
> So I am afraid that if we do the rename of typmap to remotetype in this
> patch it will a) make backports of fixes in the related code harder, b)
> force us to rename it back again in the future.

Thanks for your comment.

> I'd keep your general approach but keep using typmap naming.

I update the patch as Petr Jelineks mention, keep using typmap naming.

Thanks and best regards,
Dang Minh Huong
NEC Solution Innovators, Ltd.

Attachment Content-Type Size
fix_slot_store_error_callback_v8.patch application/octet-stream 7.2 KB

In response to


Browse pgsql-hackers by date

  From Date Subject
Next Message Alvaro Herrera 2017-12-18 14:25:47 Re: [HACKERS] Add support for tuple routing to foreign partitions
Previous Message Stephen Frost 2017-12-18 13:48:27 Re: GSoC 2018