From: | Dennis Björklund <db(at)zigo(dot)dhs(dot)org> |
---|---|
To: | pgsql-hackers(at)postgresql(dot)org |
Subject: | FROM clause omitted |
Date: | 2003-07-16 06:29:01 |
Message-ID: | Pine.LNX.4.44.0307160816420.9468-100000@zigo.dhs.org |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
I had a bug in one of my queries that wasn't detected by pg because if
filled in the from clause by itself. Take for example a querie like
select foo.a;
which I guess is transformed to
select foo.a
from foo;
Is this really a good thing to do? Is it part of the standard? Can it be
turned of? In my case it hid a bug and made my query work but produce the
wrong result.
Isn't this yet another case of "helpful" parsing that will only hurt in
the end? Look at how hard it is to parse html-pages because all browsers
accept broken code, but different broken code.
What about an example like this (the transformed code above but with alias
x added):
select foo.a
from foo x;
By adding the alias x the query still workes but gives a different result.
--
/Dennis
From | Date | Subject | |
---|---|---|---|
Next Message | Patrick Welche | 2003-07-16 09:07:55 | Re: backend/parser compile prob |
Previous Message | Tom Lane | 2003-07-16 04:27:21 | Re: perm question |