Re: [HACKERS] COPY problems with psql / libpq
От | Patrick Welche |
---|---|
Тема | Re: [HACKERS] COPY problems with psql / libpq |
Дата | |
Msg-id | 20000120172659.L12966@quartz.newn.cam.ac.uk обсуждение исходный текст |
Ответ на | Re: [HACKERS] COPY problems with psql / libpq (Alfred Perlstein <bright@wintelcom.net>) |
Ответы |
Re: [HACKERS] COPY problems with psql / libpq
|
Список | pgsql-hackers |
On Thu, Jan 20, 2000 at 09:22:16AM -0800, Alfred Perlstein wrote: > * Patrick Welche <prlw1@newn.cam.ac.uk> [000120 09:10] wrote: > > On Thu, Jan 20, 2000 at 11:02:31AM -0500, Tom Lane wrote: > > > > > > It sure sounds like psql is failing to recognize the trailing \. > > > of the COPY data. > > > > Precisely what I saw yesterday (cf Subject: pg_dump disaster) - but what > > does one do about it? > > Is this with a recent snapshot or 6.5.3 using libpq? For me, it's using yesterday's cvs'd source - but I obviously can't speak for Oliver. > Either way, you should check the contents of the send buffer, please let > me know if there is data queued in it. You can include the 'internal' > header for libpq (libpq-int.h?) to get at the send buffer. That will take a while. In the meantime, just pg_dumpall something and try to read the output back in. I do have ^M's in some of the text columns if that matters. Cheers, Patrick
В списке pgsql-hackers по дате отправления: