From: | Craig Ringer <craig(at)postnewspapers(dot)com(dot)au> |
---|---|
To: | Albretch Mueller <lbrtchx(at)gmail(dot)com> |
Cc: | pgsql-jdbc(at)postgresql(dot)org |
Subject: | Re: making postgresql "source" its data feed from external csv file ... |
Date: | 2011-01-04 00:14:10 |
Message-ID: | 4D226652.9040507@postnewspapers.com.au |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-jdbc |
On 01/04/2011 09:55 AM, Albretch Mueller wrote:
> As I understand postgresql "COPY FROM" does in effect copy the data
> to the DBMS, but what I need is for the DBMS to "source" its data feed
> with the outer file and synchronize it as it changes without having to
> run "COPY TO". Is this possible? This way I can keep the data in a way
> that I can try on any DBMS
No, that's not currently possible. What you describe is covered by
SQL/MED. There's ongoing work to implement SQL/MED for 9.1 / 9.2, but
there is no support for it in any current release version of PostgreSQL.
SQL/MED is intended for a lot more than just accessing flat files, but
should be capable of that.
There's a reason relational databases don't store their backing data as
CSV files - because they're *SLOW* to work with, in particular to
update. You can expect horribly slow write performance with any scheme
like this even if you do get it working.
--
Craig Ringer
From | Date | Subject | |
---|---|---|---|
Next Message | Maciek Sakrejda | 2011-01-04 00:38:29 | Re: making postgresql "source" its data feed from external csv file ... |
Previous Message | Albretch Mueller | 2011-01-03 22:55:03 | making postgresql "source" its data feed from external csv file ... |