Re: Disable vacuuming to provide data history - Mailing list pgsql-hackers

From Vik Fearing
Subject Re: Disable vacuuming to provide data history
Date
Msg-id ad2cd43b-0e98-1683-99c8-e9d54bacdb26@postgresfriends.org
Whole thread Raw
In response to Re: Disable vacuuming to provide data history  (Corey Huinker <corey.huinker@gmail.com>)
Responses Re: Disable vacuuming to provide data history  (marekmosiewicz@gmail.com)
Re: Disable vacuuming to provide data history  (Hannu Krosing <hannuk@google.com>)
List pgsql-hackers
On 2/24/23 22:06, Corey Huinker wrote:
> On Thu, Feb 23, 2023 at 6:04 AM <marekmosiewicz@gmail.com> wrote:
> 
>   [1] some implementations don't use null, they use an end-timestamp set to
> a date implausibly far in the future ( 3999-12-31 for example ),

The specification is, "At any point in time, all rows that have their 
system-time period end column set to the highest value supported by the 
data type of that column are known as current system rows; all other 
rows are known as historical system rows."

I would like to see us use 'infinity' for this.

The main design blocker for me is how to handle dump/restore.  The 
standard does not bother thinking about that.
-- 
Vik Fearing




pgsql-hackers by date:

Previous
From: Jeff Davis
Date:
Subject: Re: Move defaults toward ICU in 16?
Next
From: Kirk Wolak
Date:
Subject: Re: Proposal: :SQL_EXEC_TIME (like :ROW_COUNT) Variable (psql)