Re: [HACKERS] generated columns - Mailing list pgsql-hackers

From Corey Huinker
Subject Re: [HACKERS] generated columns
Date
Msg-id CADkLM=d+Jiz8fGxw+HaYE8zjZvpZrRvVtke-77p5G2Q0v7DSZg@mail.gmail.com
Whole thread Raw
In response to Re: [HACKERS] generated columns  (Michael Paquier <michael@paquier.xyz>)
List pgsql-hackers
> 3. Radical alternative: Collapse everything into one new column.  We
> could combine atthasdef and attgenerated and even attidentity into a new
> column.  (Only one of the three can be the case.)  This would give
> client code a clean break, which may or may not be good.  The
> implementation would be uglier than #1 but probably cleaner than #2.  We
> could also get 4 bytes back per pg_attribute row.
>
> I'm happy with the current choice #1, but it's worth thinking about.

#3 looks very appealing in my opinion as those columns have no overlap,
so it would take five possible values:

Could the removed columns live on...as generated-always columns?

pgsql-hackers by date:

Previous
From: Alvaro Herrera
Date:
Subject: Re: pgsql: Add flag values in WAL description to all heap records
Next
From: Robert Haas
Date:
Subject: Re: wal_dump output on CREATE DATABASE