Re: COPY formatting
От | Karel Zak |
---|---|
Тема | Re: COPY formatting |
Дата | |
Msg-id | 20040318103136.GD19093@zf.jcu.cz обсуждение исходный текст |
Ответ на | COPY formatting (Lee Kindness <lkindness@csl.co.uk>) |
Список | pgsql-hackers |
On Thu, Mar 18, 2004 at 09:29:03AM +0000, Lee Kindness wrote: > To be honest this idea strikes me as overkill - over > engineering. It was suggestion, maybe you're right :-) > While i have done a lot of messing around reading/writing the binary > format (and been stung by changes in that format) if you are using > this format then you're 99% likely to be in control of the > incoming/outgoing data and thus able to format to your wishes outwith > COPY. I partly agree.But.. there is possible write directly final file by backend withoutdata transfer to client. If we wantto support this feature we needcontrol output format by server... And.. I can image format that is use for BE/FE data transfer only andnot for some final data presentation. For example compression of datastream from/to BE without PostgreSQL protocol change. > Something else in the TODO regarding COPY is XML import/export, and > for this to be supported in your proposed implementation the function > would need to be passed in a heap more information. Yes, very probably some struct with all COPY information and formatspecific stuff. Tom was right that in this caseit will C functionsonly. As I said I will try implement it without user defined functioncall for format conversion, but I will do it modular and in future wecan create some interface for user defined formats. Karel -- Karel Zak <zakkr@zf.jcu.cz>http://home.zf.jcu.cz/~zakkr/
В списке pgsql-hackers по дате отправления: