From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Greg Stark <gsstark(at)mit(dot)edu> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: minor view creation weirdness |
Date: | 2003-10-03 17:26:07 |
Message-ID: | 28239.1065201967@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Greg Stark <gsstark(at)mit(dot)edu> writes:
> Uhm, doesn't the spec have anything to say about this?
> I mean, the view sure looks like standard SQL on its face.
Well, you might read the spec as requiring the view column to have
datatype CHAR(n) where n is the length of the unknown literal.
I see in SQL92:
9) The data type of a <character string literal> is fixed-length
character string. The length of a <character string literal>
is the number of <character representation>s that it contains.
...
Note: <character string literal>s are allowed to be zero-length
strings (i.e., to contain no characters) even though it is
not permitted to declare a <data type> that is CHARACTER with
<length> zero.
The NOTE would appear to imply that
CREATE VIEW v AS SELECT '';
is illegal, which is not a conclusion I care to follow blindly. In any
case, in Postgres I would think we would want to take the type as "text"
not "CHAR(n)", spec or no spec.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Andrew Dunstan | 2003-10-03 18:32:56 | initdb |
Previous Message | Gottfried F. Zojer | 2003-10-03 17:22:39 | Re: Question regarding coopting Database Engine |