On Wed, Sep 16, 2020 at 1:48 AM Ashutosh Sharma <ashu.coek88@gmail.com> wrote:
> Attached is the patch with the changes suggested here. I've tried to
> use a normal heap table with (autovacuum = off) wherever possible.
> Please have a look and let me know for any other issues.
I think the comment needs some wordsmithing -- "unlike other cases" is
not that informative, and "we get a stable vacuum results" isn't
either very clear or all that grammatical. If we're going to add a
comment add here, why not just "use a temp table, so autovacuum can't
interfere"?
Tom, I know that you often have strong feelings about the exact
wording and details of this kind of stuff, so if you feel moved to
commit something that is fine with me. If not, I will take my best
shot at it.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company