Re: ALTER TABLE ... TO ... to change related names

Поиск
Список
Период
Сортировка
Искать
От
Rod Taylor
Тема
Re: ALTER TABLE ... TO ... to change related names
Дата
Msg-id
1062285120.83670.50.camel@jester
Ответ на
Список
Дерево обсуждения
ALTER TABLE ... TO ... to change related names Jonathan Gardner <jgardner@jonathangardner.net>
Re: ALTER TABLE ... TO ... to change related names Andrew Dunstan <andrew@dunslane.net>
Re: ALTER TABLE ... TO ... to change related names Jonathan Gardner <jgardner@jonathangardner.net>
Re: ALTER TABLE ... TO ... to change related names Tom Lane <tgl@sss.pgh.pa.us>
Re: ALTER TABLE ... TO ... to change related names Jonathan Gardner <jgardner@jonathangardner.net>
Re: ALTER TABLE ... TO ... to change related names Dennis Björklund <db@zigo.dhs.org>
Re: ALTER TABLE ... TO ... to change related names Tom Lane <tgl@sss.pgh.pa.us>
Re: ALTER TABLE ... TO ... to change related names Andrew Dunstan <andrew@dunslane.net>
Re: ALTER TABLE ... TO ... to change related names Dennis Björklund <db@zigo.dhs.org>
Re: ALTER TABLE ... TO ... to change related names Rod Taylor <rbt@rbt.ca>
Re: ALTER TABLE ... TO ... to change related names Thomas Swan <tswan@idigx.com>
> > Backwards compatibility, if nothing else.  Are you prepared to break
> > every existing dump file that has
> > 	CREATE TABLE ser (f1 serial);
> > 	SELECT pg_catalog.setval('ser_f1_seq', 1, false);
> 
> I think breaking backward compability not always bad. If

Replacing SERIAL with a GENERATOR make sense as the definition of an
identity (including current count) is treated as an attribute of the
column for create table.

SQL2003 GENERATOR / IDENTITIES (based on DB2 and MSSQL, I believe) have
greater functionality than SERIAL.
В списке pgsql-hackers по дате отправления
От: Dennis Björklund
Дата:
От: Gaetano Mendola
Дата:
Сообщение: Re: SetVariable
FAQ