Distinct SQL when many to many - sql

Question

3 tables sto_Product, sto_ProductXCategory and sto_Category
sto_Product has all the product info (bvin=Primary Key)
sto_Category has all the Categoryinfo (bvin=Primary Key)
sto_ProductXCategory (ProductId and CategoryId are PK and FK)
currently it results in duplicate products because some products are duplicate in sto_ProductXCategory (with different CategoryId's).
this SQL statement is part of a larger stored procedure so I'm trying to keep it similar.
This is the statement I'm trying to fix
DECLARE #Sortorder int =6;
with product as
(
select distinct p.*
from sto_Product p
inner join sto_ProductXCategory px on(p.bvin = px.productid)
where p.parentID = ''
and p.Status = 1
and p.siteprice >=0
and p.siteprice <= 2147483647
and (
(p.manufacturerid = '8036ab2c-641e-487b-a577-76c09c5c3cc9'
)
)
),
AllProducts AS
(
SELECT
RowNum =
CASE
WHEN #Sortorder = 0
THEN ROW_NUMBER() OVER (ORDER BY px.SortOrder)
WHEN #Sortorder = 1
THEN ROW_NUMBER() OVER (ORDER BY px.SortOrder)
WHEN #Sortorder = 2
THEN ROW_NUMBER() OVER (ORDER BY ProductName)
WHEN #Sortorder = 3
THEN ROW_NUMBER() OVER (ORDER BY SitePrice)
WHEN #Sortorder = 4
THEN ROW_NUMBER() OVER (ORDER BY SitePrice DESC)
WHEN #Sortorder = 6
THEN ROW_NUMBER() OVER (ORDER BY ProductName DESC)
ELSE ROW_NUMBER() OVER (ORDER BY px.SortOrder)
END,
p.*
from product p
inner join sto_ProductXCategory px on(p.bvin = px.productid)
)
SELECT AllProducts.RowNum,*,
(SELECT COUNT(*) FROM AllProducts) As TotalRowCount
FROM AllProducts
WHERE RowNum BETWEEN (1) and (12 )
ORDER BY AllProducts.RowNum
Sample Dataset
sto_Product
bvin,sku,ProductName,SitePrice
0001,001,AProductName,100
0002,002,BProductName,50
0003,003,CProductName,75
sto_Category
bvin,CategoryName
1000,ACategoryName
2000,BCategoryName
3000,CCategoryName
sto_ProductXCategory
ProductId,CategoryId,SortOrder
0001,1000,1
0001,2000,2
0001,3000,3
0002,2000,4
0003,2000,5
0003,3000,6
thanks in advance

Answer 1

You've stated that
...currently it results in duplicate products because some products are duplicate in sto_ProductXCategory (with different CategoryId's).
But you havent said what it is you want to happen.
For my purposes here, I'll assume you want only one row per product, ignoring the different values of CategoryID appearing in sto_ProductXCategory
In AllProducts you have:
from product p inner join sto_ProductXCategory px on(p.bvin = px.productid)
this needs to be changed to filter the duplicate rows in sto_ProductXCategory that you do not want in you output dataset;
from product p join (
select productid, min(categoryid) as Category
from sto_ProductXCategory
group by productid
) x on p.bvin = x.productid
join sto_ProductXCategory px on p.bvin = px.Productid and x.Category = px.CategoryID
How's that?