On Friday 26 May 2006 09:45, Gurjeet Singh wrote:
> Something not supported should be stated as such through an ERROR. Except
> for count(), none of the following make any sense:
>
> The transcipt:
>
> test=# \d t1
> Table "public.t1"
> Column | Type | Modifiers
> --------+---------+-----------
> a | integer | not null
> Indexes:
> "t1_pkey" PRIMARY KEY, btree (a)
>
> test=# select * from t1;
> a
> ---
> 1
> 2
> 3
> 4
> 5
> (5 rows)
>
given:
pagila=# select 1 from t1;
?column?
----------
1
1
1
1
1
(5 rows)
> test=# select count(*) from t1;
> count
> -------
> 5
> (1 row)
>
this makes sense
> test=# select count(1) from t1;
> count
> -------
> 5
> (1 row)
>
and so does this
> test=# select max(*) from t1;
> max
> -----
> 1
> (1 row)
>
not so much
> test=# select max(1) from t1;
> max
> -----
> 1
> (1 row)
>
but this does
> test=# select min(*) from t1;
> min
> -----
> 1
> (1 row)
>
not here though
> test=# select avg(*) from t1;
> avg
> ------------------------
> 1.00000000000000000000
> (1 row)
>
nor here
> test=# select sum(*) from t1;
> sum
> -----
> 5
> (1 row)
>
or here
> test=# select sum(1) from t1;
> sum
> -----
> 5 <--- this is correct
> (1 row)
>
> test=#
yep... but really most aggregates are ok with a 1
--
Robert Treat
Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL
---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster
