Re: serial sequence problem with existing keys
От | Stephan Szabo |
---|---|
Тема | Re: serial sequence problem with existing keys |
Дата | |
Msg-id | 20011031083109.L18528-100000@megazone23.bigpanda.com обсуждение исходный текст |
Ответ на | serial sequence problem with existing keys (Randall Perry <rgp@systame.com>) |
Ответы |
Re: serial sequence problem with existing keys
|
Список | pgsql-general |
On Wed, 31 Oct 2001, Randall Perry wrote: > I've had to modify a parent table by recreating it (I substituted integer > foreign key values from lookup tables for text in 2 fields. As postgres > doesn't have a DROP COLUMN command I couldn't revise the table structure > without rebuilding it) . > > It has a serial data type for the primary key. I'm importing the data back > in and need to keep the original primary key values, as they're used in > several child tables. > > But, if I put in my own key values in a serial field the auto-increment > sequence uses the next number from where it left off, and not the max value, > and complains that it can't create a duplicate key. > > Is there any way around this, or am I going to have to programmatically > increment the key and change serial to int? Check out setval('<sequence name>', <number>) to change the sequence's current value.
В списке pgsql-general по дате отправления: