Help with joins

Question

Hi,
I am not sure if I am doing something wrong here, but would appreciate if someone point out my mistake.
I have two table item and item_status
item_status do not contain all the items but only the ones which have specific status
Question(1):
select count(*) from item
and
select count(*) from item left join item_status on item.itemid = item_status.item_id
Result of these two queries dont match. Can you'll plz help me understand it.
Question(2):
I want count of items from table item by location (location is another field in table item) but I want to exclude items which have status lost in the item_status table.
This is the query I used:
select location, count(*) from item left join item_status on item.itemid = item_status.item_id where item_status.status = 'lost'
is this the correct way, coz when I tried the figures where way to low,
Thanks,
Raja

Answer 1

Hi, Raja,
Raja Gopalan wrote:
Hi,
I am not sure if I am doing something wrong here, but would appreciate if someone point out my mistake.
I have two table item and item_status
item_status do not contain all the items but only the ones which have specific status
Question(1):
select count(*) from item
and
select count(*) from item left join item_status on item.itemid = item_status.item_id
Result of these two queries dont match. Can you'll plz help me understand it.If there can be many rows in item_status that all match the same row in item, then the 2nd query will have a higher COUNT than the first.
If you'd post your tables, I could show you in detail.
Since you didn't I'll use tables from the scott schema.
SELECT  d.deptno
FROM     scott.dept     d;produces 4 rows:
.   DEPTNO
        10
        20
        30
        40Only 3 of these rows have a match in the mep table, but the ones that match have many matches.
SELECT     d.deptno
,     e.ename
FROM          scott.dept     d
LEFT OUTER JOIN     scott.emp     e     ON     d.deptno     = e.deptno;produces 15 rows
.   DEPTNO ENAME
        20 SMITH
        30 ALLEN
        30 WARD
        20 JONES
        30 MARTIN
        30 BLAKE
        10 CLARK
        20 SCOTT
        10 KING
        30 TURNER
        20 ADAMS
        30 JAMES
        20 FORD
        10 MILLER
        401 row for each row in emp that has a match in dept (that is all 14 rows in emp) plus 1 row for every row in dept that does not have a match in emp (1 row).
Question(2):
I want count of items from table item by location (location is another field in table item) but I want to exclude items which have status lost in the item_status table.
This is the query I used:
select location, count(*) from item left join item_status on item.itemid = item_status.item_id where item_status.status = 'lost'
is this the correct way, coz when I tried the figures where way to low,If you want a count by location, then you should say "GROUP BY location".
You probably need a sub-query to check for 'lost'. Without seeing your tables and data, I can't be sure.
Maybe something like
SELECT  location
,     COUNT (*)
FROM     item
WHERE     itemid     NOT IN     (     -- Begin subquery to find bad item_ids
                   SELECT  item_id
               FROM     item_status
               WHERE     status     = 'lost'
                   )     -- End subquery to find bad item_ids
GROUP BY  location
;