Here is a small simplified snipped of my data
OrderID QTY ItemID ActualQTY(does not exist in database)
1 2 1
2 1 2
3 1 1
4 5 3
Now I need a query that will fill in the ActualQTY based on the ItemID's. So summing total QTY for ItemID 1 = 3, and for ItemID 2 = 1, and last for ItemID 3 = 5
It should look like this
OrderID QTY ItemID ActualQTY
1 2 1 3
2 1 2 1
3 1 1 3
4 5 3 5
The problem is I am new to TSQL and I can't figure out a good way to do this.
--EDIT--
Someone else helped me with this problem and gave this solution which seems like the most efficient solution to me. However this solution doesn't work if you need to apply them to an XSD file in visual studio. So I turned it into a table valued function on the server.
SELECT OrderID, QTY, ItemID, SUM(QTY) OVER(PARTITION BY ItemID) AS ActualQTY
So if this solution doesn't work resort to answers below
You can do this:
SELECT
t2.OrderID,
t2.QTY,
t2.ItemID,
t1.ActualQTY
FROM
(
SELECT
ItemId,
SUM(QTY) AS ActualQTY
FROM tablename AS t1
GROUP BY ItemId
) AS t1
INNER JOIN tablename AS t2 ON t1.ItemId = t2.ItemID;
SQL Fiddle Demo
But if you want to update the column ActualQTY values, not just select, you can do this:
UPDATE t1
SET t1.ActualQTY = t2.ActualQTY
FROM tablename AS t1
INNER JOIN
(
SELECT
ItemId,
SUM(QTY) AS ActualQTY
FROM tablename AS t1
GROUP BY ItemId
) AS t2 ON t1.ItemID = t2.ItemID
SQL Fiddle Demo
SELECT b.OrderID,
b.QTY,
b.ItemID,
a.total_sum AS ActualQTY
FROM (SELECT ItemID,
Sum(QTY) AS total_sum
FROM tablename
GROUP BY ItemID) AS a
INNER JOIN tablename b
ON a.ItemID = b.ItemID
Select t.orderid, t.qty, t.itemid, a.actualq as actualqty
From yourtable t join (
Select sum(qty) as actualq, itemid
From yourtable
Group by itemid) a
On t.itemid = a.itemid
Related
I am trying to do a market-basket analysis for sales data. For this, I need two values, order_id and product_id.
I only want the order_id and the product_id when there are duplicate values of each not necessarily in the same row, but I don't want to get rid of those values, I want to get rid of the order_id and product_id if they only show in the database once. It should be ordered by order_id. The result should look something like this
from this
order_id
product_id
1
1
1
2
1
3
1
4
2
1
2
2
2
3
3
1
3
2
3
5
4
2
to this
order_id
product_id
1
1
1
2
1
3
2
1
2
2
2
3
3
1
3
2
SELECT order_id, product_id
FROM order_items
ORDER BY order_id
I am sure there is a far more elegant solution, but if I understand your question to want rows that have either an order_id or product_id that shows up more than once in the DB, then I believe this will work:
EDIT:
If you need them both to have multiple items in the list then I would just individually join them to a subquery as such:
SELECT t1.order_id, t1.product_id
FROM order_items t1
INNER JOIN (
SELECT ORDER_ID from order_items group by ORDER_ID having count(*) > 1) t2
ON t1.order_id = t2.order_id
INNER JOIN (
SELECT PRODUCT_ID from order_items group by PRODUCT_ID having count(*) > 1) t3
ON t1.product_id = t3.product_id
ORDER BY order_id
A couple of windowed aggregates would probably work best™ here:
WITH CTE AS(
SELECT order_id,
product_id,
COUNT(1) OVER (PARTITION BY order_id) AS orders,
COUNT(1) OVER (PARTITION BY product_id) AS products
FROM dbo.YourTable)
SELECT order_id,
product_id
FROM CTE
WHERE orders > 1
AND products > 1;
in keyword can make the query simpler, but I'm not sure about the performance...
select order_id, product_id from order_items
where order_id not in (
select order_id from order_items group by order_id having count(*) = 1
) and product_id not in (
select product_id from order_items group by product_id having count(*) = 1
);
I have a query the following works as expected
If((Select count(*) from table1 where product = 'carrot')< 5)
Begin
Select Top (5 - (Select count(*) from table1 where product = 'carrot'))
id, product From table2
WHere id NOT IN
(Select id from table1) AND product = 'carrot'
Order by newid()
END
What i want to do is Union or Union all say another product potatoes
If((Select count(*) from table1 where product = 'potato')< 5)
Begin
Select Top (5 - (Select count(*) from table1 where product = 'potato'))
id, product From table2
WHere id NOT IN
(Select id from table1) AND product = 'potato'
Order by newid()
END
I keep getting a syntax error, when i add UNION between IF or after END. Is this possible or another way is better....
What i am doing is trying to select a random sample of carrots, first i want to check if i have the 5 carrots in table1. if i do don't run sample.
If i do not have 5 total carrots run the sampler and return 5 carrots. I then filter out if they already exist in table 1 by the id. Then it subtracts the count from the new sample for a total of five.
It works well, now i want to run for other products eg lettuce, potatoes etc...
But i want an UNION or UNION All. hope makes sense.
I'd be interested to see whether this way works-
Select Top (5 - (Select count(*) from table1 where product = 'carrots')< 5)
id
, product
From table2
WHere id NOT IN (Select id from table2)
AND (Select count(*) from table1 where product = 'carrots')< 5)
UNION ALL
Select Top (5 - (Select count(*) from table1 where product = 'potatoes')< 5)
id
, product
From table2
WHere id NOT IN (Select id from table2)
AND (Select count(*) from table1 where product = 'potatoes')< 5)
Your style is interesting, feels procedural rather than set-based.
You can try it this way
If(((Select count(*) from table1 where product = 'carrot'< 5) and (Select count(*) from table1 where product ='potato' <5))
)
Begin
Select Top (5 - (Select count(*) from table1 where product = 'carrot')) id, product
From table2
WHere id NOT IN (Select id from table1) AND product = 'carrot' Order by newid()
Union all
Select Top (5 - (Select count(*) from table1 where product = 'potato')) id, product From table2
WHere id NOT IN (Select id from table1) AND product = 'potato' Order by newid()
END
IF statements in SQL do not behave as sub-queries or row-sets in SQL, as you've found out. They are for branching the flow of control only.
Here is a more set based approach you could take:
SELECT ProdSamples.*
FROM
(
SELECT Table2.*, ROW_NUMBER() OVER (PARTITION BY table2.Product ORDER BY NEWID()) RowNum
FROM Table2
LEFT JOIN Table1
ON Table1.id = Table2.id
WHERE Table1.id IS NULL
) ProdSamples
JOIN
(
SELECT Product, COUNT(*) ProdCount
FROM Table1
GROUP BY Product
) ProdCounts
ON ProdSamples.Product = ProdCounts.Product
AND ProdSamples.RowNum <= (5 - ProdCounts.ProdCount)
The first sub-query ProdSamples returns all the products from Table2 that do not have an id in Table1. The RowNum field ranks them in random order partitioned by Product.
The second sub-query ProdCounts is the count of records for each product in Table1. Then it joins these sub-queries together and only returns the records from ProdSamples where the RowNum is lower or equal to the number of samples you want to return.
Having a transaction table with the following rows:
Id UserId PlatformId TransactionTypeId
-------------------------------------------------
0 1 3 1
1 1 1 2
2 2 3 2
3 3 2 1
4 2 3 1
How do I write a stored procedure that can aggregate the rows into a new table with the following format?
Id UserId Platforms TransactionTypeId
-------------------------------------------------
0 1 {"p3":1,"p1":1} {"t1":1,"t2":1}
1 2 {"p3":2} {"t2":1,"t1":1}
3 3 {"p2":1} {"t1":1}
So the rows are gouped by User, count each platform/transactionType and store as key/value json string.
Ref: My previous related question
You could use GROUP BY and FOR JSON:
SELECT MIN(ID) AS ID, UserId, MIN(sub.x) AS Platforms, MIN(sub2.x) AS Transactions
FROM tab t
OUTER APPLY (SELECT CONCAT('p', platformId) AS platform, cnt
FROM (SELECT PlatformId, COUNT(*) AS cnt
FROM tab t2 WHERE t2.UserId = t.UserId
GROUP BY PlatformId) s
FOR JSON AUTO) sub(x)
OUTER APPLY (SELECT CONCAT('t', TransactiontypeId) AS Transactions, cnt
FROM (SELECT TransactiontypeId, COUNT(*) AS cnt
FROM tab t2 WHERE t2.UserId = t.UserId
GROUP BY TransactiontypeId) s
FOR JSON AUTO) sub2(x)
GROUP BY UserId;
DBFiddle Demo
Result is a bit different(array of key-value) but please treat it as starting point.
Your sample JSON is not really a json, but since you want it that way:
SELECT u.UserId, plt.pValue, ttyp.ttValue
FROM Users AS [u]
CROSS APPLY (
SELECT '{'+STUFF( (SELECT ',"'+pn.pName+'":'+LTRIM(STR(pn.pCount))
FROM (SELECT p.Name AS pName, COUNT(*) AS pCount
FROM transactions t
left JOIN Platforms p ON p.PlatformID = t.PlatformId
WHERE t.UserId = u.UserId
GROUP BY p.PlatformId, p.Name
) pn
FOR XML PATH('')),1,1,'')+'}'
) plt(pValue)
CROSS APPLY (
SELECT '{'+STUFF( (SELECT ',"'+tty.ttName+'":'+LTRIM(STR(tty.ttCount))
FROM (SELECT tt.Name AS ttName, COUNT(*) AS ttCount
FROM transactions t
left JOIN dbo.TransactionType tt ON tt.TransactionTypeId = t.TransactionTypeID
WHERE t.UserId = u.UserId
GROUP BY tt.TransactionTypeId, tt.Name
) tty
FOR XML PATH('')),1,1,'')+'}'
) ttyp(ttValue)
WHERE EXISTS (SELECT * FROM transactions t WHERE u.UserId = t.UserId)
ORDER BY UserId;
DBFiddle Sample
I want to group the key with count greater than 3, and the query will return the rest of the records also. I don't want to use Union All, is there any other way to do it?
ID
1
1
1
2
3
3
4
4
4
4
Return
1
1
1
2
3
3
4
You can use ranking- and aggregate functions:
WITH CTE AS
(
SELECT ID,
CNT = COUNT(*) OVER (PARTITION BY ID),
RN = ROW_NUMBER() OVER (PARTITION BY ID ORDER BY ID)
FROM dbo.TableName
)
SELECT ID
FROM CTE
WHERE CNT <= 3 OR RN = 1
Demo
I'd do it like this
SELECT
GroupedData.ID
FROM
(SELECT ID, CNT = COUNT(*)
FROM dbo.TableName
GROUP BY ID) GroupedData AS g
LEFT JOIN dbo.TableName AS t
ON t.id = g.id and g.CNT<=3
This also allows you to add further columns which report details for the group or individual record as appropriate
SELECT
g.ID,
ISNULL(t.RecordName,'Grouped Records') as RecordName,
ISNULL(t.NumericField,g.NumericField) as NumericField
FROM
(
SELECT ID, CNT = COUNT(*), SUM(NumericField) as NumericField
FROM dbo.TableName
GROUP BY ID
) GroupedData AS g
LEFT JOIN dbo.TableName AS t
ON t.id = g.id and g.CNT<=3
I have 2 tables in SQL Server 2008 and I want to get the details from those 2 tables using join.
T-1 : vwHardwareConsolidate
|ID|||Qty|Type|Task_Id|
T-2 :
|MasterID|Task_Id|Act_Qty|
I want to get id, task_name, sum(qty), task_id from T1 and Masterid, Act_Qty from T2
I have tried this query
select
ID as MasterID, Task_id, Task_Name as Items,
SUM(Qty) as Req_Qty, 0 as Act_Qty
from
vwHardwareConsolidate
where
type = 'Reqrd' and ID = '21'
Group by
Task_Name,id,Task_id
union
(select
m.MasterID, m.Task_Id, vw.Task_Name as Items, 0 as Req_Qty, m.Act_Qty
from
vwHardwareConsolidate vw
Right join
(select
MasterID, m.Task_Id, 0 as Req_Qty, sum(Act_qty) as Act_Qty
from
tbl_MaterialDistribution_Detail m
where
MasterID = '21'
group by
m.Task_Id, MasterID) as m on m.Task_Id = vw.Task_id)
vwHardwareConsolidate
ID Site_name Qty Task_Name Type
1 CITY 1 A16Port_Switch Reqrd
1 CITY 1 Digital_Camera Reqrd
1 CITY 1 Electronic_Pen Reqrd
tbl_MaterialDistribution_Detail:
MasterID|TaskId|Act_qty
7 31 1
2 32 1
12 39 3
Please try this
select t1.ID, t1.Task_Name,Sum(t1.Qty) as Qty,t1.Task_Id,t2.MasterID,t2.Act_Qty
from vwHardwareConsolidate as t1
left outer join table2 as t2
on t2.Task_ID=t1.Task_ID
Group By t1.ID, t1.Site_name, t1.Task_Name,t1.Qty,t1.Type,t1.Task_Id,t2.MasterID,t2.Act_Qty
May this will help you.
SQL Fiddle Demo
You can use
SELECT T1.Id, T1.SiteName,T1.TaskName,T.Type,T2.Act_Qty,T2.MasterID,T2.Task_Id,SUM(Qty)AS
Qty FROM T1 INNER JOIN T2 ON T1.Task_Id=T2=Task_Id GROUP BY T1.Id, T1.SiteName,
T1.TaskName,T.Type,T2.Act_Qty,T2.MasterID,T2.Task_Id
something this way,
select a.id,a.Task_Name,b.qty,c.Act_Qty from vwHardwareConsolidate a inner join
tbl_MaterialDistribution_Detail c on a.id=c.task_id
inner join
(slect id,sum(qty)qty from vwHardwareConsolidate group by id)b
on a.id=b.id