Re: SERIAL type feature request - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: SERIAL type feature request
Date
Msg-id 200512042129.jB4LTEi15896@candle.pha.pa.us
Whole thread Raw
In response to Re: SERIAL type feature request  (Zoltan Boszormenyi <zboszor@dunaweb.hu>)
List pgsql-hackers
Zoltan Boszormenyi wrote:
> Peter Eisentraut ?rta:
> 
> >Josh Berkus wrote:
> >  
> >
> >>I believe that our SERIAL/SEQUENCE stuff is already in compliance
> >>with the SQL standard for sequences (in SQL03).   Why would we change
> >>it?
> >>    
> >>
> >
> >Because your belief is wrong, but Zoltan's proposal is not getting is 
> >closer.
> >
> >  
> >
> OK, what does the standard say on SERIAL for specifying the start value?
> And about this:
> 
> <last serial value was e.g. 307>
> insert into mytable (serial_id, ...) values (500, ...);
> delete from mytable where serial_id = 500;
> 
> In Informix, this is a way to start the next serial value at 501.

This seems much stranger than a simple setval(), which get got from
Oracle.

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
359-1001+  If your life is a hard drive,     |  13 Roberts Road +  Christ can be your backup.        |  Newtown Square,
Pennsylvania19073
 


pgsql-hackers by date:

Previous
From: Andrew Dunstan
Date:
Subject: Re: pg_restore [archiver] file offset in dump file is too
Next
From: Andrew Dunstan
Date:
Subject: Re: SERIAL type feature request