如何使用条件和 PARTITION BY 制作 ROW_NUMBER() function？

Question

I'm looking for help on a SQL query.

Let's say I have a dataset that includes multiple levels of groups. Here's an example: we have students, that took some class, and then had some tests in each class. Let's say I have to use this dataset. I want to do some sort of query (I'm thinking ROW_NUMBER) that increments when there's a new Class ID, but not when there's a new Test ID, PARTITION BY Student ID.

Sample dataset:

StudentID   ClassID   TestID
----------------------------
1            1          1
1            2          1
1            3          1
2            1          1
2            1          2
3            2          1

Desired output:

StudentID   ClassID   TestID   ClassNumber
------------------------------------------
1            1          1      1
1            2          1      2
1            3          1      3
2            1          1      1
2            1          2      1
3            2          1      1

I want to do things like see if the student did better in their 2nd class versus their 1st class, regardless of what the class was.

This is the code I'm trying:

SELECT 
    StudentID, ClassID, TestID, 
    ROW_NUMBER() OVER (PARTITION BY StudentID ORDER BY StudentID ASC) AS "ClassNumber"
FROM 
    mytable

Now I know this is WRONG, but I don't know how to get it to do what I want it to do?

Answer 1

From your sample data, it seems like you can use partition by studentID, testID order by ClassID :

select 
    t.*,
    row_number() over(partition by studentID, testID order by ClassID) ClassNumber
from mytable t

Note: with that few sample data it is hard to tell wich ranking fucntion would best suit your use case ( row_number() , rank() , dense_rank() ). Since you used row_number() , I sticked to it.

Demo on DB Fiddle

StudentID | ClassID | TestID | ClassNumber
--------: | ------: | -----: | :----------
        1 |       1 |      1 | 1          
        1 |       2 |      1 | 2          
        1 |       3 |      1 | 3          
        2 |       1 |      1 | 1          
        2 |       1 |      2 | 1          
        3 |       2 |      1 | 1          

Answer 2

You may want dense_rank() :

dense_rank() over (partition by studentID order by ClassID) as ClassNumber

If you had:

4            1          1
4            1          2
4            2          2

This would assign:

4            1          1         1
4            1          2         1
4            2          2         2

which is how I understand your question.

Answer 3

I think this is what you are looking for:

Select StudentID, 
ClassID, 
TestID, 
ROW_NUMBER() Over (Partition By StudentID, TestID Order By StudentID) As ClassNumber 
From myTable