Multiple column Union Query without duplicates - sql

Question

I'm trying to write a Union Query with multiple columns from two different talbes (duh), but for some reason the second column of the second Select statement isn't showing up in the output. I don't know if that painted the picture properly but here is my code:
Select empno, job
From EMP
Where job = 'MANAGER'
Union
Select empno, empstate
From EMPADDRESS
Where empstate = 'NY'
Order By empno
The output looks like:
EMPNO JOB
4600 NY
5300 MANAGER
5300 NY
7566 MANAGER
7698 MANAGER
7782 MANAGER
7782 NY
7934 NY
9873 NY
Instead of 5300 and 7782 appearing twice, I thought empstate would appear next to job in the output. For all other empno's I thought the values in the fields would be (null). Am I not understanding Unions correctly, or is this how they are supposed to work?
Thanks for any help in advance.

Answer 1

If you want the data in a separate column you will want a JOIN not a UNION:
Select e.empno, e.job, a.empstate
From EMP e
left join EMPADDRESS a
on e.empno = a.empno
Where job = 'MANAGER'
AND empstate = 'NY'
Order By e.empno
A UNION combines the two results into a single set but the data is listed in the same columns. So basically they are placed on top of one another:
select col1, col2, 'table1' as src
from table1
union all
select col1, col2, 'table2' as src
from table2
Will result in:
col1 | col2 | src
t1 | t1 | table1
t2 | t2 | table2
If you want to have the data in a separate column which is sounds like you do then you will use a join of the tables.

Answer 2

Bluefeet has the correct answer.
Think of joins as combining tables horizontally - you're adding more columns to the original query with each table you join.
Think of unions as stacking record sets vertically - you're adding extra rows to the same set of columns.

Answer 3

You need a JOIN for this..
Select e.empno, e.job, ea.empstate
From EMP e LEFT OUTER JOIN EMPADDRESS ea ON e.empno = ea.empno
Where e.job = 'MANAGER'
And ea.empstate = 'NY'
Order By e.empno
UNION is for taking 2 result sets with the same column names and merging them into one. In your example, its lumping column 2 (job and empstate) together, and taking the name from the first select.

Answer 4

i think you meant to write is as a join instead?
ie if you wanted empstate to be null for those employee's not in NY.
select empno, job, empstate
from emp e
left outer join empaddress a
on a.empno = e.empno
and e.empstate = 'NY'
where e.job = 'MANAGER';

Answer 5

this one works in oracle..by using union ..here inner query will fetch out the all the columns after that grouping with empno and rest of the columns is string concatenated
select EMPNO
,wm_concat(job) job
,wm_concat(EMPSTATE) EMPSTATE
from
( select EMPNO,job,'' as EMPSTATE from EMP Where job ='MANAGER'
union select EMPNO,'' as job, EMPSTATE from EMPADDRESS Where empstate ='NY'
)
group by EMPNO order by 1