Re: [HACKERS] DROP TABLE inside transaction block

Поиск
Список
Период
Сортировка
От Leon
Тема Re: [HACKERS] DROP TABLE inside transaction block
Дата
Msg-id 37D3AAD2.C0AA4D4A@udmnet.ru
обсуждение исходный текст
Ответ на Re: [HACKERS] DROP TABLE inside transaction block  (Tom Lane <tgl@sss.pgh.pa.us>)
Ответы Re: [HACKERS] DROP TABLE inside transaction block  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-hackers
Tom Lane wrote:

> 
> > Cant you just rename to a unique name, maybee in another directory,
> 
> Not if other backends are also accessing the table.  Remember that to
> make this really work, the DROP would have to be invisible to other
> backends until commit.
> 

Is that really needed? Remember that table's creation is not transparent
to other users - when someone attempts to create a table, others,
though can't see that table, cannot create a table with the same name.
So you can simply issue a draconian-level lock on a table being deleted.
But in any case it would need postponing real killing until transaction
commit.

> The amount of work needed seems vastly more than the feature is worth...

I personally have a project in development which extensively uses
that feature. It is meant to be database restructuring 'on the fly'.
If you break that, it would be a big drawback to me. And I assume, not
only to me, because it would break an idea of trasaction itself. 
Database restructuring by software, not by hand, will be seriously
damaged.

-- 
Leon.
-------
He knows he'll never have to answer for any of his theories actually 
being put to test. If they were, they would be contaminated by reality.



В списке pgsql-hackers по дате отправления:

Предыдущее
От: "Ansley, Michael"
Дата:
Сообщение: RE: [HACKERS] Postgres' lexer
Следующее
От: Tom Lane
Дата:
Сообщение: Re: [HACKERS] md.c is feeling much better now, thank you