On Sat, 7 Feb 2009 15:49:49 -0600 "Adam Rich" <adam.r@xxxxxxxxxxxxx> wrote: > > > > I have the following sql: > > > > SELECT * from table order by dato asc limit 20 offset 0 > > > > > > This gives me different rows than the 20 first rows when running the > > following sql: > > > > SELECT * from table order by dato asc > > > > > > Shouldn't the 20 first rows in the second sql statment be the same 20 > > rows that is returned in the first statement > > or am I missing something? > > > > Not necessarily. In your example query, if "dato" was not a unique > column, and there were some duplicates, the "top 20" values is not > a defined set. Adding the offset clause might cause a different > query plan, resulting in a different ordering of the duplicate values. > > I found out that if I created an index on the dato field, then I got the same 20 rows... Does that make sense or is it just a coincedense? BTJ -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
