Query to perform like Minus statement

Question

Hi Guys,
I have an issue to ask. I need to extract the final issued parts from a table. Table stores the issue and unissue parts.
e.g lets say 4 items have been issued and 3 items have been un-issued from a table. so in last 1 items should be resulted. while I am executing the following code its not working correctly.
Table data:
REPAIRED_ITEM_ID      ISSUED_REMOVED_ PART_ID     Oper_ID     QTY
122013187                      1323938                                       308             1
122013187                      1323938                                       308             1
122013187                      1323938                                       309             1
122013187                      1323940                                       308             1
122013187                   1323940                                       308             1
122013187                      1323940                                       309             1
122013187                      1323940                                       309             1
SELECT * FROM WC1.ISSUED_REMOVED_ITEM IRI
    WHERE
    IRI.ORDER_ITEM_OPER_ID = 308
    AND IRI.SES_CUSTOMER_ID =1
    AND IRI.REPAIRED_ITEM_ID =  122013187 
    AND 0 = (SELECT COUNT(1)
           FROM WC1.ISSUED_REMOVED_ITEM U
           WHERE U.REPAIRED_ITEM_ID =  IRI.REPAIRED_ITEM_ID
           AND U.ORDER_ITEM_OPER_ID = 309
           AND U.SES_CUSTOMER_ID= 1
           )But the above query is not showing any result, while it should show 1 record of 1323938 , Can anyone please explain me how to get this right result?     
Note: Repair Item id is unique of MAIN UNIT. which can contains multiple issue and un-issue of dependent units.
Many Thanks,
M.C.
Edited by: BluShadow on 06-Sep-2012 13:19
added {noformat}{noformat} tags to help readability.  Please read {message:id=9360002} and learn to post questions with format in future.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   

Answer 1

Hi,
Mark Cooper wrote:
Eg. If a unit Serial no '354879019900009' has a part (1015268) issued 8 times and then unissued 4 times so finally the part was issued 4 times. so I need 4 rows to show for each qty 1 for that part and unit serial number. Please find script below.
All the letters have to be lower-case in \ tags.
-- ITL Table
Create table ITL_TEST ( ...Thanks for posting the CREATE TABLE and INSERT statments.
Don't forget to post the exact results you want from the given sample data, and which version of Oracle you're using (e.g. 11.2.0.2.0).
COMMIT;CREATE TABLE ... AS is a DDL command.  DDL commands automatically COMMIT, so you don't need a separate COMMIT statement here.  (It's not doing any harm, of course; I just thought you'd like to know.)
If you need to reference individual rows of the issued_removed_item table, then you might want to change Stew's query to use the analytic SUM function, not the aggregate SUM.  Either way, compute issued_qty in a sub-query, and wait until you're in a super-query to test for issued_qty>0, so you don't have to repeat the DECODE expression.
I'm guessing you want something like this:WITH     got_issued_qty          AS
     SELECT repaired_item_id
     ,     issued_part_id
--     ,     ...               -- any other columns you want
     ,     SUM ( DECODE ( oper_id
               , 308     , +issued_removed_quantity
               , 309     , -issued_removed_quantity
          )     OVER ( PARTITION BY repaired_item_id
               ,     issued_part_id
               )          AS issued_qty
     FROM issued_removed_item
     WHERE     repaired_item_id     IN (122013187)     -- Easy to add its, if needed
,     cntr     AS
     SELECT     LEVEL     AS n
     FROM     (
               SELECT MAX (issued_qty)     AS max_issued_qty
               FROM got_issued_qty
     CONNECT BY     LEVEL     <= max_issued_qty
SELECT     t.item_serial_no, t.item_bcn     -- or t.*, or whatever columns you want
,     q.repaired_item_id, q.issued_part_id     -- or q.*, or whatever columns you want
FROM          itl_test     t
LEFT JOIN     got_issued_qty     q ON q.repaired_item_id     = t.item_id
                    AND     q.issued_qty          > 0
LEFT JOIN     cntr          c ON q.issued_qty          <= c.n
ORDER BY q.repaired_item_id
,      q.issued_part_id
Again, this is just a guess.  Until you post the results you want, all I can do is guess.  Guessing isn't the best way to solve problems.
I displayed only a few columns, just to make the output more readable.  Adding whatever columns you want later will be trivial.
The output I get is 74 rows, starting with:ITEM_SERIAL_NO ITEM_BCN REPAIRED_ITEM_ID ISSUED_PART_ID
354879019900009 BCN141290167 122013187 1015268
354879019900009 BCN141290167 122013187 1015268
354879019900009 BCN141290167 122013187 1015268
354879019900009 BCN141290167 122013187 1015268
354879019900009 BCN141290167 122013187 1015268
354879019900009 BCN141290167 122013187 1015268
354879019900009 BCN141290167 122013187 1015268
354879019900009 BCN141290167 122013187 1015268 ...
For testing and debugging, you may may to change the columns that are displayed.
I don't know if 74 rows is right for this sample data.  I'm certain that it's too many for initial testing.  If you could devise some sample data such that the desired output was only 20 or 25 rows, that would be a lot easier to test and to understand.