Question on using Distinct operator and avoiding ORA-00936 error

Question

I may have answered my own question, by searching the OTN forums, but I just want to confirm what I've learned.
I'm trying to use a distinct on one of my columns, where I am pulling multiple columns, and trying to avoid duplicate entries. From most of my textbook example, when an example is using a distinct, they were only pulling one column.
From my example below, when I try to use a distinct, I get the following error: ORA-00936: missing expression:
select p.list_id, distinct(p.associate_id), d.first, d.last, c.street1, c.street2, c.city, c.state, c.country, c.postal_code, c.email
from sf.profiles p, demographics d, contact_info c
where p.entity_id = d.entity_id
and p.entity_id = c.entity_id
and p.list_id = 111
and associate_id in (insert associate_id in here
);Once I searched OTN, I saw some examples, where they had the distinct operator as the first column. So I switched it with p.list_id, and the query ran find.
select distinct p.associate_id, p.list_id, d.first, d.last, c.street1, c.street2, c.city, c.state, c.country, c.postal_code, c.email
from sf.profiles p, demographics d, contact_info c
where p.entity_id = d.entity_id
and p.entity_id = c.entity_id
and p.list_id = 111
and associate_id in (insert associate_id in here);So, does the distinct operator have to go first, if using multiple columns?
thanks
Message was edited by:
cmmiller

Answer 1

With or without parenthesis, it's a distinct clause for all columns in select statement:
SQL> select * from tt1;
        ID T
     10203 1
     10203 2
SQL> select distinct(id), text from tt1;
        ID T
     10203 1
     10203 2
SQL> select id, text from tt1;
        ID T
     10203 1
     10203 2
SQL> Nicolas.