RE: Yet another SQL newbie question

UNION removes duplicates and UNION ALL preserves duplicates.

-----Original Message-----
Subject: RE: Yet another SQL newbie question

A join is not the same as a union.

For example 
select * from jhhl1
union 
select * from jhhl2
returns
....+....1....
MYKEY  MYCHAR 
  A    Test A 
  B    Test B 
  C    Test C 
  D    Test D 
  E    Test E 
  F    Test F 
********  End 

While
select a.*, b.*  from jhhl1 a 
join jhhl2 b on a.mykey=b.mykey
returns nothing
....+....1....+....2....+....3....+.
MYKEY  MYCHAR      MYKEY  MYCHAR 
********  End of data  ******** 

Summary:  A union returns separate rows from each table while a join 
combines rows from multiple tables into a single row.


Hey, I noticed that if I change the contents of jhhl3 from
....+....1...
MYKEY  MYCHAR
  C    Test 1
  F    Test 2
********  End
to
....+....1...
MYKEY  MYCHAR
  C    Test C
  F    Test F
********  End

Then I can do
select * from jhhl1
union 
select * from jhhl2
except 
select * from jhhl3

and get
....+....1...
MYKEY  MYCHAR
  A    Test A
  B    Test B
  D    Test D
  E    Test E
********  End

However, if I leave MYCHAR back at their original values then I get six 
rows.  EXCEPT and INTERSECT are in the section "What's new for V5R3 in the 
SQL Reference Book".  Looks like you'll have to use my earlier suggestion 
for obsolete versions of the operating system.

Rob Berendt