From: | Adrian Klaver <aklaver(at)comcast(dot)net> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Cc: | "Albretch Mueller" <lbrtchx(at)gmail(dot)com> |
Subject: | Re: ERROR: relation . . . does not exist |
Date: | 2008-08-28 23:50:55 |
Message-ID: | 200808281650.55565.aklaver@comcast.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Thursday 28 August 2008 3:14:01 pm Albretch Mueller wrote:
> On Thu, Aug 28, 2008 at 5:40 PM, Raymond O'Donnell <rod(at)iol(dot)ie> wrote:
> > You have to create the table first before you can COPY into it.
>
> ~
> Well, based on how the statement reads I made the wrong assumption of
> thinking that PG would be smart enough to make ends-meat with the
> data that it gets once you indicate:
> ~
> 1) the header names (to be mapped to column names)
> ~
> 2) the data from which PG could easily figure out the types for each
> column, since it is CSV data
Define easily. A CSV column has fields with 1's and 0's should that be an
integer field or a boolean field? A CSV column has a series of text entries
none of which are greater then 25 characters long. Does that mean a datatype
of varchar(25) or is it just an artifact of the data to date? You have a CSV
column with text strings of format 'YYYY/MM/DD' . Do you want them stored as
strings or dates? I could go on, but the point is that table data types
require some thought on the part of the DBA.
> ~
> That may/should become a RFE
> ~
> Thanks
> lbrtchx
> ~
--
Adrian Klaver
aklaver(at)comcast(dot)net
From | Date | Subject | |
---|---|---|---|
Next Message | Matthew Dennis | 2008-08-29 00:45:00 | Re: indexes on functions and create or replace function |
Previous Message | Tom Lane | 2008-08-28 23:22:06 | Re: indexes on functions and create or replace function |