From: | Peter Eisentraut <peter_e(at)gmx(dot)net> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | Andrew McMillan <Andrew(at)catalyst(dot)net(dot)nz>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Proposal for fixing numeric type-resolution issues |
Date: | 2000-05-15 18:56:51 |
Message-ID: | Pine.LNX.4.21.0005142037410.915-100000@localhost.localdomain |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Tom Lane writes:
> I expect what you are after is the ability to produce an 0-or-1
> numeric value from a bool field, so that you could do things like
> sum(boolfield::int) to count the number of true values in a column.
> I agree that we need such an operator (and I'm surprised no one's
> gotten round to contributing one).
Let's contribute one now ...
select count(*) from test4 where a = true;
select sum( case when a then 1 else 0 end ) from test4;
> But I don't agree that there
> should be an implicit, automatic conversion from bool to int; that
> defeats half of the error-checking value of having a separate type for
> truth values in the first place.
Definitely.
--
Peter Eisentraut Sernanders väg 10:115
peter_e(at)gmx(dot)net 75262 Uppsala
http://yi.org/peter-e/ Sweden
From | Date | Subject | |
---|---|---|---|
Next Message | Peter Eisentraut | 2000-05-15 18:57:27 | Re: Casting, again |
Previous Message | Jim Mercer | 2000-05-15 18:44:26 | broken links on http://www.postgresql.org/doxlist.html |