SQL still have duplicates after Group By - sql

Question

I'm trying to get a grand total of the qty_on_hand between our 2 warehouses combined but after I do a group by the totals don't combine. I'm not sure what I am doing wrong.
SELECT
sku_master.warehouse,
sku_master.sku
, sku_master.min_on_hand
, sku_master.max_on_hand
, x.total_qty_on_hand
FROM [wms].dbo.[sku_master]
left join
(
SELECT
sku_master.sku, sum(location_inventory.qty_on_hand) as total_qty_on_hand
FROM [wms].[dbo].[location_inventory]
JOIN [wms].dbo.[sku_master] ON location_inventory.sku = sku_master.sku
GROUP BY sku_master.sku) x on sku_master.sku = x.sku
Results
Expected Results

Answer 1

Looks like you want sum all qty_on_hand for each sku. Try something like this using OUTER APPLY
SELECT m.sku,
m.min_on_hand,
m.max_on_hand,
oa.total_qty_on_hand
FROM (SELECT sku,
Min(min_on_hand) min_on_hand,
Max(max_on_hand) max_on_hand
FROM [wms].dbo.[sku_master] m
GROUP BY sku) m
OUTER Apply (SELECT Sum(l.qty_on_hand) AS total_qty_on_hand
FROM [wms].[dbo].[location_inventory] l
WHERE l.sku = m.sku) oa

Answer 2

If you need only the grand total then don't join or just for sample add with a union this
SELECT
sku_master.warehouse,
sku_master.sku
, sku_master.min_on_hand
, sku_master.max_on_hand
, x.total_qty_on_hand
FROM [wms].dbo.[sku_master]
left join
(
SELECT
sku_master.sku, sum(location_inventory.qty_on_hand) as total_qty_on_hand
FROM [wms].[dbo].[location_inventory]
JOIN [wms].dbo.[sku_master] ON location_inventory.sku = sku_master.sku
GROUP BY sku_master.sku) x on sku_master.sku = x.sku
union
SELECT null, null, nullm null, sum(location_inventory.qty_on_hand)
FROM [wms].[dbo].[location_inventory]
JOIN [wms].dbo.[sku_master] ON location_inventory.sku = sku_master.sku
or do the fact you have sku_master.warehouse not grouped and repeated .. remove it form select
SELECT
sku_master.sku
, sku_master.min_on_hand
, sku_master.max_on_hand
, x.total_qty_on_hand
FROM [wms].dbo.[sku_master]
left join
(
SELECT
sku_master.sku, sum(location_inventory.qty_on_hand) as total_qty_on_hand
FROM [wms].[dbo].[location_inventory]
JOIN [wms].dbo.[sku_master] ON location_inventory.sku = sku_master.sku
GROUP BY sku_master.sku) x on sku_master.sku = x.sku

Answer 3

you can use this simple query
SELECT
t.sku,
min(t.min_on_hand) as min_qty_on_Hand,
max(t.max_on_hand) as max_on_hand,
sum(t.total_qty_on_hand) + sum(t2.total_qty_on_hand) as total_qty_on_hand
FROM table1 t
inner join table2 t2
on t.sku = t2.sku
group by t.sku
the attached image shows the process including the result
Note: the result here is same as you mentioned in your question and what you need here just change the table names into your table name

Answer 4

In your request's comments you explain that you prefer one warehouse over the other. In this case you should have a table containing the warehouses with a column for the priority. Thus the query wouldn't have to know about warehouse priorities. (Just imagine you add a warehouse some day - you'd have to change all your queries.)
My query aggregates sku_master records per SKU and warehouse and then picks the best warehouse per SKU. It then joins the aggregated inventory rows by SKU.
Here is the query without a warehouse table.
select
m.warehouse,
m.sku,
m.total_min_on_hand,
m.total_max_on_hand,
li.total_qty_on_hand
from
(
select
sku,
warehouse,
sum(min_on_hand) as total_min_on_hand,
sum(max_on_hand) as total_max_on_hand,
row_number() over (partition by sku
order by case when warehouse = 'XDGM' then 1 else 2 end) as rn
from sku_master
group by sku
) m
join
(
select sku, sum(qty_on_hand) as total_qty_on_hand
from location_inventory
group by sku
) li on li.sku = m.sku
where m.rn = 1; -- only the better warehouse when there is more than one for the SKU
With a warehouse table the from clause would slightly change to:
from
(
select
sku,
warehouse,
sum(min_on_hand) as total_min_on_hand,
sum(max_on_hand) as total_max_on_hand,
row_number() over (partition by sku order by w.priority) as rn
from sku_master sm
join warehouse w on w.warehouse = sm.warehouse
group by sku
) m