How to compare rows in two identical tables and return result set of missing rows?

Question

I've looked through all the related questions, but none of them does exactly what I need. I have two identically-structured tables (id, VoucherNbr, BalanceInit) - one is from our current production system (tableA), and the other is from a client's legacy system (tableB). Table A has records a,b,c,d,e (for example), and table B has a,b,c,d,e,f,g. Table B will always have all the values that exist in table A, but also has extra rows. I need to return a resultset that contains all the rows in table B that don't exist in table A (f and g, using the example.) How can I do this?

EDIT:

TABLE A

ID | VoucherNbr         | BalanceInit
=============================================
1  | 1111111111111111   | 25.00
2  | 2222222222222222   | 50.00
3  | 3333333333333333   | 10.00

TABLE B

ID | VoucherNbr         | BalanceInit
=============================================
15 | 1111111111111111   | 25.00
17 | 1212121212121212   | 15.00
22 | 2222222222222222   | 50.00
34 | 3333333333333333   | 25.00
41 | 3232323232323232   | 75.00

What I need to get back is a resultset containing only the rows that don't exist in Table A, where it's the value in the VoucherNbr field that doesn't exist, not the ID field.

ID | VoucherNbr         | BalanceInit
=============================================
17 | 1212121212121212   | 15.00
41 | 3232323232323232   | 75.00

Answer 1

Note: Assuming id is the PK for both tables.

SELECT B.* FROM TableB B 
LEFT OUTER JOIN TableA A
ON B.id = A.id
WHERE A.id IS NULL

EDIT : Based on further commments from OP, since Id's are different for the same rows in the two tables, assuming that the matching criteria is based on comparing the 2 column values

SELECT B.* FROM TableB B 
LEFT OUTER JOIN TableA A
ON B.VoucherNbr = A.VoucherNbr
AND B.BalanceInit = A.BalanceInit
WHERE A.VoucherNbr IS NULL 

EDIT2 : If only voucher number itself is sufficient, then balanceInit too can be discarded

SELECT B.* FROM TableB B 
LEFT OUTER JOIN TableA A
ON B.VoucherNbr = A.VoucherNbr
WHERE A.VoucherNbr IS NULL 

Answer 2

Some databases offer this feature directly using a SUBTRACT or MINUS operation. MySQL doesn't seem to have any such operation.

I would probably do this:

SELECT  B.id, B.voucherNbr, B.balanceInit FROM tableB B
   WHERE NOT EXISTS 
     (SELECT * FROM tableA A 
      WHERE A.voucherNbr = B.voucherNbr AND A.balanceInit = B.balanceInit)

However, here are a couple of other MySQL solutions (I googled them) you can test for speed:

SELECT  id, voucherNbr, balanceInit FROM tableB 
   WHERE (voucherNbr, balanceInit) NOT IN 
      (SELECT voucherNbr, balanceInit FROM tableA);

SELECT  id, voucherNbr, balanceInit
   FROM tableB LEFT JOIN tableA USING (voucherNbr, balanceInit)
      WHERE tableA.voucherNbr IS NULL

Honestly, I like the first googled solution better than my own if the execution time is similar or better.

Note: I've modified these from my original answer to reflect the fact that you state in a comment to another answer that the ids vary in the two tables.