Re: [HACKERS] distinct. Is this the correct behaviour? - Mailing list pgsql-hackers

From Damond Walker
Subject Re: [HACKERS] distinct. Is this the correct behaviour?
Date
Msg-id 00ef01bf1be5$567e4600$af63a8c0@walkers.org
Whole thread Raw
Responses Re: [HACKERS] distinct. Is this the correct behaviour?  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
----- Original Message -----
> Hmph, so sybase hasn't thought through the implications of ORDER BY on
> a hidden column vs. DISTINCT either.  Can anyone try it on some other
> DBMSes?
>

Using the following script...
   create table t1(f1 int, f2 int);   insert into t1(f1, f2) values(1,1);   insert into t1(f1, f2) values(1,2);
insertinto t1(f1, f2) values(1,3);   insert into t1(f1, f2) values(2,4);   select distinct f1 from t1 order by f2;
 


Returned the following message under Oracle8 on NT:   ORA-01791: not a SELECTed expression

Returned the following message under MS SQL Server 7.0:   ORDER BY items must appear in the select list if SELECT
DISTINCTis
 
specified.

I could try it on Oracle8i but I suspect the result will be the same.



pgsql-hackers by date:

Previous
From: "Ansley, Michael"
Date:
Subject: RE: [HACKERS] distinct. Is this the correct behaviour?
Next
From: "Paul Becker"
Date:
Subject: RE: [DOCS] Re: [HACKERS] Outline for PostgreSQL book