From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Paul Condon <pecondon(at)quiknet(dot)com> |
Cc: | matt(at)ctlno(dot)com, pgsql-general <pgsql-general(at)hub(dot)org> |
Subject: | Re: Re: Crosstab SQL Question |
Date: | 2000-06-16 06:14:31 |
Message-ID: | 5814.961136071@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Paul Condon <pecondon(at)quiknet(dot)com> writes:
> There are lots of issues with how one displays a crosstab which are not
> easily addressed within the context of RDBMS. I suggest the PostgreSQL
> team not plan to add crosstabs to PostgreSQL, ...
As near as I could see, the data that was being asked for was exactly
what you could get from a GROUP BY with multiple group columns:
SELECT salesman, month, SUM(netprice) FROM salesorders
GROUP BY salesman, month;
However the raw SQL output would look like
Jim Jan 1100
Jim Feb 1500
...
Joe Jan 120
Joe Feb 1200
...
so what's really wanted is a frontend to reformat this into a nicely
presented table. I agree with Paul that that's not the task SQL is
designed for, whereas there are lots of data-visualization tools that
*are* designed for it.
> I have found a nice implementation of crosstabs in R, which is a
> system/language for statistical analysis, data display, and data
> plotting. <www.r-project.org>
Yup, R is cool (and free). Anyone up for making an R-to-PGSQL
interface?
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2000-06-16 06:49:23 | Re: Performance for indexes on functions |
Previous Message | Michael Meskes | 2000-06-16 06:09:00 | Re: Crosstab SQL Question |