从子查询中选择“最好的结果”？

Question

SELECT * FROM table1 
WHERE col1 = ANY (
    SELECT DISTINCT col2 FROM table2 
    WHERE first = 'blah' 
    OR second = 'blahblah' 
    OR third = 'blahblahblah'); 

该子查询检查三列（first,second,third）。在这个例子中，我使用ANY，而是我希望它返回这样的：

If found `first` return `first` 
else if found `second` return `second` 
else if found `third` return `third` 

知道我在说什么？我只能优先选择1排（最好）first,second,third

Answer 1

我想你想要这样的：

SELECT * FROM table1 
WHERE col1 = (
    SELECT col2 FROM table2 
    WHERE first = 'blah' 
    OR second = 'blahblah' 
    OR third = 'blahblahblah' 
    ORDER BY 
    first = 'blah' DESC, 
    second = 'blahblah' DESC, 
    third = 'blahblahblah' DESC 
    LIMIT 1 
); 

Answer 2

这是做你想做的吗？

SELECT t.*, 'first' 
FROM table1 t 
WHERE col1 IN (SELECT col2 FROM table2 WHERE first = 'blah') 
UNION ALL 
SELECT t.*, 'second' 
FROM table1 t 
WHERE col1 IN (SELECT col2 FROM table2 WHERE first <> 'blah' AND second = 'blahblah') 
UNION ALL 
SELECT t.*, 'third' 
FROM table1 t 
WHERE col1 IN (SELECT col2 FROM table2 WHERE first <> 'blah' AND second <> 'blahblah' AND third = 'blahblahblah'); 

对于每个匹配（基于col1）它返回三个条件中的第一个。

Answer 3

试试这个，

SELECT *, COALESCE(first, second, third) as best FROM tab