From: | "bille(at)npphotonics (Bill Eaton)" <bille(at)npphotonics(dot)com> |
---|---|
To: | <pgsql-novice(at)postgresql(dot)org> |
Subject: | getting jiggy with AS clauses |
Date: | 2002-10-23 22:44:28 |
Message-ID: | 001401c27ae5$bf7fd300$1a64a8c0@bille |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-novice |
It would be really nice to be able to get column labels from
within the database. Or in other words, trick the AS clause
into returning info from a SELECT query, i.e.
SELECT header AS (SELECT header from labheaders where
dataset=1) FROM ...
Let's say you've got several columns of data from the
laboratory. It would be nice if I could store the column
headers in one place (LabHeaders) and store the data in
another place (LabData)
LabHeaders LabData
---------- --------
dataset int8 <--> dataset int8
headers text[] row int8
labrawdata float8[]
The tricky part is constructing a query that returns
something like
row MyColumnDataFromLabHeadersTable
---- ---------------------------------
1 3.1415927
2 2.7128
3 5.3
Why not just used static column names, you ask? Well, we can
pretty much rest assured that the things that are measured
will occasionally change. And it's pretty silly to add a new
table every time a change is made.
I've played a bit with writing functions that return SETOF,
but with little success.
Ideas?
Bill Eaton
mailto://bill(at)npphotonics(dot)com
From | Date | Subject | |
---|---|---|---|
Next Message | Andrew McMillan | 2002-10-24 00:02:15 | Re: [BUGS] Optimization disaster |
Previous Message | Radhesh Mohandas | 2002-10-23 21:15:36 | Re: [ANNOUNCE] San Francisco Bay Area PostgreSQL User's Group |