Re: [HACKERS] Date/Time Flaw in pg_dump ?
От | Thomas Lockhart |
---|---|
Тема | Re: [HACKERS] Date/Time Flaw in pg_dump ? |
Дата | |
Msg-id | 37346C8F.D31A4535@alumni.caltech.edu обсуждение исходный текст |
Ответ на | Re: [HACKERS] pg_dump problem? (Bruce Momjian <maillist@candle.pha.pa.us>) |
Список | pgsql-hackers |
> Any datetime fields are different. I think it's a timezone problem. > The dump includes the timezone as part of the dump, so I'm guessing that > the problem is on the part of psql not noticing that. I'm using the > Australian "EST" zone if that's useful. > Is there an immediate work-around? Yeah, move to the east coast of the US :) EST is the US-standard designation for "Eastern Standard Time" (5 hours off of GMT). If you compile your backend with the flag -DUSE_AUSTRALIAN_RULES=1 you will instead get this to match the Australian convention, but will no longer handle the US timezone of course. This is used in backend/utils/adt/dt.c, and is done with an #if rather than an #ifdef. Perhaps I should change that... btw, Australia has by far the largest "timezone space" I've ever seen! There are 17 Australia-specific timezones supported by the Postgres backend. I know it's a big place, but the "timezone per capita" leads the world ;) - Tom -- Thomas Lockhart lockhart@alumni.caltech.edu South Pasadena, California
В списке pgsql-hackers по дате отправления: