| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com> |
| Cc: | James Coleman <jtc331(at)gmail(dot)com>, Robert Haas <robertmhaas(at)gmail(dot)com>, Andrew Dunstan <andrew(at)dunslane(dot)net>, "Bossart, Nathan" <bossartn(at)amazon(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Document atthasmissing default optimization avoids verification table scan |
| Date: | 2022-03-26 23:36:23 |
| Message-ID: | 183570.1648337783@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
"David G. Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com> writes:
> Or, we can leave it where things are and make sure the reader understands
> there are two paths to having a NOT NULL constraint on the newly added
> column. Something like:
> "If you plan on having a NOT NULL constraint on the newly added column you
> should add it as a column constraint during the ADD COLUMN command. If you
> add it later via ALTER COLUMN SET NOT NULL the table will have to be
> completely scanned in order to ensure that no null values were inserted."
The first way also requires having a non-null DEFAULT, of course, and
then also that default value must be a constant (else you end up with
a table rewrite which is even worse). This sort of interaction
between features is why I feel that a separate unified discussion
is the only reasonable solution.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Daniel Gustafsson | 2022-03-26 23:37:27 | Re: Add LZ4 compression in pg_dump |
| Previous Message | Tom Lane | 2022-03-26 23:28:07 | Re: pgsql: Add 'basebackup_to_shell' contrib module. |