Subquery Group By clause

Question

In the AdventureWorks2012 database, I have to write a query that shows all columns from the Sales.SalesOrderHeader table and the average LineTotal from the Sales.SalesOrderDetail table

Attempt 1

SELECT * 
FROM Sales.SalesOrderHeader
    (SELECT AVG (LineTotal) 
    FROM Sales.SalesOrderDetail
    WHERE LineTotal <> 0)
GROUP BY LineTotal

I get the following error:

Msg 156, Level 15, State 1, Line 3
Incorrect syntax near the keyword 'SELECT'.
Msg 102, Level 15, State 1, Line 5
Incorrect syntax near ')'.

Attempt 2

SELECT *
FROM Sales.SalesOrderHeader h
    JOIN (
    SELECT AVG(LineTotal)
    FROM Sales.SalesOrderDetail d
    GROUP BY LineTotal) AS AvgLineTotal
ON d.SalesOrderID = h.SalesOrderID

I get the following error:

Msg 8155, Level 16, State 2, Line 7
No column name was specified for column 1 of 'AvgLineTotal'.
Msg 4104, Level 16, State 1, Line 7
The multi-part identifier "d.SalesOrderID" could not be bound.

Subqueries are very confusing for me. What am I doing wrong? Thanks.

Answer 1

Well, you're mixing your aliases and some other things.

Second version should look like that

SELECT h.*, d.avgLineTotal
FROM Sales.SalesOrderHeader h
    JOIN (
    SELECT SalesOrderID, --you need to get this to make a join on it
    AVG(LineTotal)as avgLineTotal --as stated by error, you have to alias this (error 1)
    FROM Sales.SalesOrderDetail
    GROUP BY SalesOrderID) d --this will be used as subquery alias (error 2)
ON d.SalesOrderID = h.SalesOrderID

another solution would be

select h.field1, h.field2,  -- etc. all h fields
coalesce(AVG(sod.LineTotal), 0)
from Sales.SalesOrderHeader h
LEFT JOIN Sales.SalesOrderDetail d on d.SalesOrderID = h.SalesOrderID
GROUP BY h.field1, h.field2 --etc. all h fields