Re: 8.0.0beta4: "copy" and "client_encoding"
От | Kris Jurka |
---|---|
Тема | Re: 8.0.0beta4: "copy" and "client_encoding" |
Дата | |
Msg-id | Pine.BSO.4.56.0411051519360.28172@leary.csoft.net обсуждение исходный текст |
Ответ на | Re: 8.0.0beta4: "copy" and "client_encoding" (Markus Schaber <schabios@logi-track.com>) |
Ответы |
Re: 8.0.0beta4: "copy" and "client_encoding"
|
Список | pgsql-jdbc |
On Fri, 5 Nov 2004, Markus Schaber wrote: > I think you're right. Except COPY from STDIN, client encoding should be > independent from COPY encoding. So I suggest that Adrian requests > backend support for this. > Well that's not going to get in for 8.0, so we'll need to do something especially since the driver is supposed to offer backward compatibility. From my perspective we've got three options, just drop the check, Oliver's add a URL parameter to turn off the check, or add an API so that they don't issue the COPY command directly, but something like issueServerCopy(table, file, encoding) where the driver could turn off the check, switch the encoding, do the copy, switch the encoding back, and turn on the check. I think the third option is too much work at the moment, when we do offer a copy API in the driver we could fold that in, but not now. As to the first two options, I don't really care. Either is fine with me, so since Oliver seems to like a URL parameter, that's what we'll probably go with. Kris Jurka
В списке pgsql-jdbc по дате отправления: