Oracle：僅從最小行數中選擇數值（id）

Question

SELECT 
    ass.assessmentAmount -- want to fetch assessmentAmount of min(ass.assessmentId) 
    ass.assessmentId 
FROM 
    --bunch of joins 
WHERE 
ass.assessmentId = (SELECT min(ass2.assessmentId) FROM Assessment ass2 
    --same bunch of joins 

它看起來很混亂，因爲我有6個連接條件，我不想重複兩次。有沒有另外一種方法呢？

Answer 1

使用MIN(ass.assessmentId) OVER()解析函數：

SELECT * 
FROM (
    SELECT ass.assessmentAmount, 
     ass.assessmentId, 
     MIN(ass.assessmentId) OVER() AS min_assessmentId 
    FROM --bunch of joins 
) 
WHERE assessmentId = min_assessmentId; 

您還可以使用RANK()：

SELECT * 
FROM (
    SELECT ass.assessmentAmount, 
     ass.assessmentId, 
     RANK() OVER (ORDER BY ass.assessmentId) AS rnk 
    FROM --bunch of joins 
) 
WHERE rnk = 1; 

如果assessmentId是UNIQUE，只能然後至少有一行你可以用ROW_NUMBER代替RANK;

SELECT * 
FROM (
    SELECT ass.assessmentAmount, 
     ass.assessmentId 
    FROM --bunch of joins 
    ORDER BY ass.assessmentId ASC 
) 
WHERE ROWNUM = 1; 

Answer 2

使用CTE與ROW_NUMBER

with CTE as 
(
select assessmentId, 
     assessmentAmount , 
     row_number() over (order by assessmentid asc) as rn 
from --bunch of joins 
) 
select * 
from CTE 
where rn = 1