Sql Query! Can someone explain this? - sql

Question

Q1 : List employees who have the biggest salary in their departments
Q2 : List employees that don't have a boss in the same department
Q1 Answer)
Select e.name, Max_Sal.Sal
From Employees e
JOIN (Select d.DepartmentID,max(e.Salary) As Sal
From Employees e
JOIN Departments d ON e.DepartmentId=d.DepartmentId
Group by d.departmentID
) As Max_Sal
ON e.DepartmentId=Max_Sal.DepartmentID
and e.Salary=Max_Sal.Sal;
In Q2, I don't understand the question, [don't have a boss in the same department]. How to know boss is in which department?

Answer 1

Q2 : List employees that don't have a boss in the same department
There is a slight trick to this question.
1 - Join
SELECT e.*
FROM Employees e
LEFT OUTER JOIN Employees b on e.BossId = b.EmployeeId
Notice the following:
We've used the Employees table twice, but I've given them separate names. e is every employee; and b is that employee's boss.
We use a LEFT OUTER JOIN to account for the possibility that an employee does not have a boss (b will be NULL) If we had used an INNER JOIN, the employees with no boss would have been omitted from the result.
Now, we have a table of every employee, connected to their boss. But we still need to filter the list as requested.
2 - Filter
There are three possible states that an employee can be in:
They have no boss.
They have a boss, who works in the same department.
They have a boss, who works in a different department.
It's important to realize that the requested output needs to contain both the employees in situation 3 and the employees in situation 1. We should only omit the employees in situation 2.
Using pseudo-SQL, this means your query should look like this:
SELECT *
FROM the_data
WHERE
situation_1_applies
OR
situation_3_applies
The individual filters are not that hard to write:
SELECT e.*
FROM Employees e
LEFT OUTER JOIN Employees b on e.BossId = b.EmployeeId
WHERE
-- situation_1_applies
-- When b is NULL, all of b's columns are NULL
b.DepartmentId is NULL
OR
-- situation_3_applies
b.DepartmentId <> e.DepartmentId

Answer 2

Don't have a boss in the same department, so you can think bossID in Employees table is null. So you can write an SQL query for a group all of the people have the same department who have the bossID is null :), So it not real because employees have to have a boss (not null), If you show me full field of table, maybe I will show you exactly my answer.
In the other hand, you can think, They have the same bossID, so your mission will write an SQL query for a group all of the people have the same bossID but they have a different department.
I have an example MySQL query (for same bossID but different department):
SELECT * FROM Employees as e
INNER JOIN department d ON d.departmentId = e.deparmentId
WHERE e.DepartmentID in (
SELECT DepartmentID
FROM Employees
GROUP BY DepartmentID
HAVING COUNT(*) = 1 /* it's fixed */
) AND e.BossID = 2 /* any value of bossID you want, you can join it with Boss table */
===== Update my answer =====
I have an idea for the question.
Anyone must have a boss, and boss also is an employee in table SQL.
' employeeName ' department name ' boss name
--------------------------------------------------------
' King ' Director '
' Victor ' Marketing ' King
' Angela ' Product ' King
' Ken ' Marketing ' Victor
' LaLa ' Marketing ' Victor
' Jery ' Business ' Ken
And because don't have a boss in the same department -> their department can't have their boss. => the result as the following table:
' employeeName ' department name ' boss name
--------------------------------------------------------
' King ' Director '
' Victor ' Marketing ' King
' Angela ' Product ' King
' Jery ' Business ' Ken
If you have a right way to understand this question, you can post here for everyone can understand. Thanks