Question

I have test_scores table with following fields:

Table schema:

  id (number)
  score1  (number)
  score2  (number)
  score3  (number)
  score4  (number)

Sample data:

id score1 score2 score3 score4
1  10     05      30    50
2  05     15      10    00
3  25     10      05    15

Expected result set:

id col_name col_value
1  score4   50
2  score2   15
3  score1   25

What is a good SQL for this?(I am using MySQL.)

Original requirement didn’t include the col names in the rowset. I was able to get the result using following SQL:

SELECT   A.id, MAX(A.score) AS max_score
FROM     (
            SELECT id, score1 as score FROM test_scores UNION
            SELECT id, score2 as score FROM test_scores UNION
            SELECT id, score3 as score FROM test_scores UNION
            SELECT id, score4 as score FROM test_scores
          ) AS A
GROUP BY A.id

But with the introduction of col_name in the result-set I am stumped..

Answer 1

I solved this problem using the following SQL:

SELECT id, GREATEST(score1, score3, score3, score4) AS col_value,
       CASE GREATEST(score1, score3, score3, score4) 
         WHEN score1 THEN 'score1'
         WHEN score2 THEN 'score2'
         WHEN score3 THEN 'score3'
         WHEN score4 THEN 'score4'
       END AS col_name
FROM test_scores

Please let me know if there is a better solution.