Date: Thu, 13 Jun 2024 10:20:22 -0700 (PDT)
From: Rich Shepard <rshepard@appl-ecosys.com>
Reply-To: Rich Shepard <rshepard@appl-ecosys.com>
To: pgsql-general@lists.postgresql.org
Subject: Reset sequence to current maximum value of rows
Message-ID: <387fa6c2-5de-f24-d522-46282597703@appl-ecosys.com>
MIME-Version: 1.0
Content-Type: text/plain; format=flowed; charset=US-ASCII
Archived-At: <https://www.postgresql.org/message-id/387fa6c2-5de-f24-d522-46282597703%40appl-ecosys.com>
Precedence: bulk

Two tables have a sequence for the PK. Over time I manually entered the PK
numbers not being aware of applying DEFAULT to generate the next number.

I just tried to set one table's PK sequence to the current max(PK) value
using this expression from a stackexchange thread:
SELECT setval('<sequence_name>', <current_max_number>, true);  -- next value will be max(PK) + 1

Needing to add a new row to a table for a specific industry table (with 52
rows) I set the PK as DEFAULT in the INSERT INTO expression. To my surprise
and disappointment all 52 rows now have the company_name column as the newly
inserted name. Feh! I need to restore all the correct names for each PK.

There's an alternate expression in that SE thread that I didn't try:
ALTER SEQUENCE <sequence_name> RESTART WITH <next_number>;

I want to avoid this same situation when resetting the second table's PK
sequence number and would like to understand why the SELECT expression
changed all column values rather than adding a new row with its attributes.
And how to I reset sequences to ignore all current values while adding the
next higher value to the end when a new row is INSERTed.

TIA,

Rich