An alternative would be to do the the work at the application level, i.e.
the point where the data is collected. You could use any language you like
for that.
> -----Original Message-----
> From: Soma Interesting [SMTP:dfunct@telus.net]
> Sent: Friday, March 30, 2001 3:59 PM
> To: Jeff Eckermann; pgsql-general@postgresql.org
> Subject: RE: [GENERAL] dynamic field names in a function.
>
> At 03:43 PM 3/30/2001 -0600, you wrote:
> >I think I am beginning to understand your question. You want to know
> >whether a field name can be referenced using a variable as all or part of
> >the field name? I don't know for sure, but I can't make it work, and
> >apparently you can't either.
> >Your earlier message is still out on the wire, so someone more
> knowledgeable
> >may still answer.
> >If you expect the maximum number of question/answer pairs to be
> manageably
> >small, you could do a series of actions for each field by name,
> incrementing
> >a counter for each pair, and comparing the counter value to 'qty'.
> >Horrible, I know, but the best I can suggest right now.
>
> I'm going to quote your rephrasing of my question to the list, maybe your
> words will carry more weight as "valid" question.
>
> Another option would be to use Tcl or Pearl, would they be able to use
> variables as all or part of the field name?