Group by clause based on condition

Question

tblScore contains score for each problem
id     problemID      score
1          1             10
2          2             30
tblSubmission contains problem submissions for each user
id     user         problemID      accepted
1     UserA             1                 0
2     UserA             1                 0
3     UserA             1                 1
4     UserA             2                 1
5     UserB             1                 0
6     UserB             1                 1
7     UserB             2                 1
Explanation  :
For UserA :
 -  For problemID 1
      ->  submitted three times
      ->  rejected for first two submission and accepted on third submission.
 - For problemID 2
      -> submitted one time
      -> accepted on first submission
For UserB :
 -  For problemID 1
      ->  submitted two times
      ->  rejected for first submission and accepted on second submission.
 - For problemID 2
      -> submitted one time
      -> accepted on first submission
Now I would like to process the table and want to get the following result :
user              Score
UserA             36 (6 + 30)
UserB             38 (8 + 30)
Explanation :
- For each rejected submission, a -2 point penalty.
- UserA have submitted probelmID 1
  - > score of problemID 1 is 10.
  - > first two times rejectd
  - > third time accepted.
  -> score = 10 - 4 = 6
- UserA have submitted problemID 2
  - > score of problemID 2 is 30
  - > first time accepted. No penalty will be counted
  - > score = 30
so final score for UserA = 30 + 6 = 36
Similar for UserB.
I know all I am doing is explanations. I have tried a lot with gorup by and other stuffs. May be my lack of knowledge on SQL is the problem.
I would really appreciate if someone help me on this.
Thank You

Answer 1

Please post DDL,so that people do not have to guess what the keys,constraints,Declarative Referential Integrity,data types,etc. in your schema are. Learn how to follow ISO-11179 data element naming conventions and formatting rules. You have no idea,and
you even tibbled and used singular table names! Temporal data should use ISO-8601 formats. Code should be in Standard SQL as much as possible and not local dialect. 
This is minimal polite behavior on SQL forums. 
You have a magical,generic “id”,which does not exist in RDBMS. We would use a “<something>_id” instead. We need a key to make this a table. 
CREATE TABLE Scores
(problem_nbr INTEGER NOT NULL PRIMARY KEY,
 problem_score INTEGER NOT NULL
   CHECK (problem_score > 0));
INSERT INTO Scores
VALUES (1,10), (2,30);
Your next table is wrong, too. SQL is a declarative language and we do not use assembly language flags. 
CREATE TABLE Submissions 
(user_name CHAR(5) NOT NULL,
 problem_nbr INTEGER NOT NULL,
 submission_nbr INTEGER NOT NULL,
 PRIMARY KEY (user_name, problem_nbr, submission_seq),
 score INTEGER DEFAULT -2 NOT NULL);
INSERT INTO Submissions
VALUES
('UserA', 1, 1, -2), 
('UserA', 1, 2, -2), 
('UserA', 1, 3, 10), 
('UserA', 2, 1, 30), 
('UserB', 1, 1, -2), 
('UserB', 1, 2, 10), 
('UserB', 2, 1, 30);
You need a procedure to insert test result into the table. But do you stop when you get a passing grade? 
CREATE PROCEDURE Post_Test_Results
(@in_user_name CHAR(5),
 @in_problem_nbr INTEGER,
 @in_pass_fail CHAR(4))
AS
INSERT INTO Submissions
SELECT @in_user_name, @in_problem_nbr, 
       CASE WHEN @in_pass_fail = 'pass'
            THEN S.problem_score
            ELSE -2 END;
  FROM Scores AS S
 WHERE S.problem_nbr = @in_problem_nbr;
Put the answer in a VIEW, so itis always right and current:
CREATE VIEW Final_Scores
AS
SELECT user_name, problem_nbr, SUM(score) AS final_score 
  FROM Submissions
 GROUP BY user_name, problem_nbr;
--CELKO-- Books in Celko Series for Morgan-Kaufmann Publishing: Analytics and OLAP in SQL / Data and Databases: Concepts in Practice Data / Measurements and Standards in SQL SQL for Smarties / SQL Programming Style / SQL Puzzles and Answers / Thinking
in Sets / Trees and Hierarchies in SQL