Question

I’m looking for SQL query for a problem that I have.

I have 2 columns – the first is “employee full name” (containing FULL name) and the second is “Hours”.

Employee Full Name | Hours | grading
-------------------------------------
john ran             122
john medi            177
john mat             138
jack rom             66
jack ton             88

I would like to update the database with the “grading” column in the following way:

grouping employees by their FIRST name, and then grading them (in ascending order) by working hours (highest working hours for employee with a specific first name) gets grade of 1, second 2, etc…).

The result should be like this:

Employee Full Name | Hours | grading
------------------------------------
john ran             122      3
john medi            177      1
john mat             138      2
jack rom             66       2
jack ton             88       1

There are no 2 employees with the same full name

Answer 1

You can use a CTE to perform this update using row_number()

;with cte as
(
  select *,
    row_number() 
      over(partition by substring([EMPLOYEE FULL NAME], 1, charindex(' ', [EMPLOYEE FULL NAME])) 
           order by hours desc) rn
  from yourtable
)
update cte
set grading = rn;

See SQL Fiddle with Demo

Ideally, you will want to normalize your database to have the first and last name in separate columns.