Re: Could regexp_matches be immutable?

Поиск
Список
Период
Сортировка
От Tom Lane
Тема Re: Could regexp_matches be immutable?
Дата
Msg-id 22087.1255583096@sss.pgh.pa.us
обсуждение исходный текст
Ответ на Re: Could regexp_matches be immutable?  (David Fetter <david@fetter.org>)
Ответы Re: Could regexp_matches be immutable?  (Josh Berkus <josh@agliodbs.com>)
Список pgsql-hackers
David Fetter <david@fetter.org> writes:
> On Wed, Oct 14, 2009 at 11:51:13PM -0400, Andrew Dunstan wrote:
>> That's not the worst of it :-( See
>> <http://openacs.org/xowiki/How_to_install_in_Postgres_8.x>

> This just illustrates the fact that at least as far as PostgreSQL is
> concerned, OpenACS is a dead project.  It's been at least 3 major
> versions since they even attempted to keep compatible with PostgreSQL.

That seems pretty harsh, considering that there's plenty of discussion
of how to use OpenACS with PG on their website.

What it looks like to me is that the recommendation about regex_flavor
might be just cargo-cult programming.  Somebody had some trouble with an
updated PG version, turned on every backwards-compatibility option he
could find, and when it worked he didn't inquire too closely into which
settings were actually important.  And ever since then that's been the
received wisdom about how to make OpenACS run with Postgres.  It would
be interesting to try turning off these options one at a time to see
which ones do matter.

(I'd bet lunch that the one about add_missing_from is bogus, too,
or could easily be made so.  mysql isn't forgiving about missing
FROM items, so it's hard to believe that they have a lot of such
things no matter how little they care about Postgres.)
        regards, tom lane


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Itagaki Takahiro
Дата:
Сообщение: Re: Skip WAL in ALTER TABLE
Следующее
От: Heikki Linnakangas
Дата:
Сообщение: Re: [PATCH] Largeobject access controls