Re: setLastTid() and currtid()

Поиск
Список
Период
Сортировка
От Andres Freund
Тема Re: setLastTid() and currtid()
Дата
Msg-id 20190411180525.sqrh7m64r3kudrtk@alap3.anarazel.de
обсуждение исходный текст
Ответ на Re: setLastTid() and currtid()  (Alvaro Herrera <alvherre@2ndquadrant.com>)
Список pgsql-hackers
Hi,

On 2019-04-11 13:52:08 -0400, Alvaro Herrera wrote:
> As I understand, this setLastTid stuff would cause trouble if used
> with a non-core table AM.

I'm not sure there'd actually be trouble. I mean, what it does for heap
is basically meaningless already, so it's not going to be meaningfully
worse for any other table AM.  It's an undocumented odd interface, whose
implementation is also ugly, and that'd be a fair reason on its own to
rip it out though.

Greetings,

Andres Freund



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Peter Billen
Дата:
Сообщение: Re: serializable transaction: exclude constraint violation (backed byGIST index) instead of ssi conflict
Следующее
От: Bruce Momjian
Дата:
Сообщение: Re: Retronym: s/magnetic disk/main data/