Question

我有一张数据表：

id   one    two    three   four     five    six
------------------------------------------------
1    12     32     2       5        34       13
2    43     12     3       33       22       17
3    11     31     3       15       13       13
4    43     12     52      73       29       19
5    3      2      2       3        9        9
6    4      1      3       7        2        19
-------------------------------------------------

所以我知道如何计算，分组和按一栏排序，如下所示：

select one, count(one) from table_numbers group by one order by count(one) desc

以上查询将给我们：

one     count(one) 
-----------------
43      2
3       1
4       1
11      1
12      1
-----------------

那么如何在一个查询中获取所有列的上述数据呢？

像这样：

One  Count(one) Two Count(two)  Three   Count(three)    Four    Count(four)
--------------------------------------------------------------------------- 
43   2          12  2           3       3               73       1           
3    1          1   1           2       2               3        1           
4    1          2   1           52      1               5        1           
11   1          31  1           null    null            7        1          
12   1          32  1           null    null            15       1          
null null      null null        null    null            33       1          
---------------------------------------------------------------------------

现在有没有办法在一个单独的SQL查询中执行此操作？可能正在使用联接或内联视图或其他任何内容？或者这可以在单个查询中进行吗？

[update]我想计算每列的重复值，并按降序排序。

[更新]如果您想使用表格数据：

CREATE TABLE IF NOT EXISTS `table_numbers` (
  `id` int(11) NOT NULL AUTO_INCREMENT,  `one` int(2) NOT NULL,
  `two` int(2) NOT NULL,  `three` int(2) NOT NULL,
  `four` int(2) NOT NULL,  `five` int(2) NOT NULL,
  `six` int(2) NOT NULL,  PRIMARY KEY (`id`),
  KEY `col_one` (`one`),  KEY `col_two` (`two`),
  KEY `col_three` (`three`),  KEY `col_four` (`four`),
  KEY `col_five` (`five`),  KEY `col_six` (`six`)
) ;

INSERT INTO `table_numbers` (`id`, `one`, `two`, `three`, `four`, `five`, `six`)
VALUES
(1, 12, 32, 2, 5, 34, 13),(2, 43, 12, 3, 33, 22, 17),(3, 11, 31, 3, 15, 13, 13),
(4, 43, 12, 52, 73, 29, 19),(5, 3, 2, 2, 3, 9, 9),(6, 4, 1, 3, 7, 2, 19);

提前致谢！

杰伊： - ）

Answer 1

在一个查询中获取列数的简便方法：

(select 'one' as col, one as item, count(one) as count from table_numbers group by one) UNION
(select 'two', two, count(two) from table_numbers group by two) UNION
(select 'three', three, count(three) from table_numbers group by three) UNION
(select 'four', four, count(four) from table_numbers group by four) UNION
(select 'five', five, count(five) from table_numbers group by five) UNION
(select 'six', six, count(six) from table_numbers group by six) UNION
ORDER BY col, count DESC

它要复杂得多如果你想要列中的摘要（3列的例子）：

SELECT tone.item as One, tone.count as `Count(one)`, ttwo.item as Two, ttwo.count as `Count(two)`, tthree.item as Three, tthree.count as `Count(three)` 
FROM 
  (SELECT @rownumtmp:=@rownumtmp+1 as rownum 
   FROM 
    (SELECT DISTINCT col, count 
     FROM (
      (select 'one' as col, one as item, count(one) as count from table_numbers group by one) UNION
      (select 'two', two, count(two) from table_numbers group by two) UNION
      (select 'three', three, count(three) from table_numbers group by three)) tmp) tmp2, 
    (SELECT @rownumtmp:=0) r) tmp2 LEFT OUTER JOIN 
  (SELECT @rownum1:=@rownum1+1 as rownum, item, count FROM (SELECT one as item, count(one) as count from  table_numbers group by one ORDER BY count DESC, one) d, (SELECT @rownum1:=0) r)  tone ON tmp2.rownum=tone.rownum LEFT OUTER JOIN
  (SELECT @rownum2:=@rownum2+1 as rownum, item, count FROM (SELECT two as item, count(two) as count from  table_numbers group by two ORDER BY count DESC, two) d, (SELECT @rownum2:=0) r)  ttwo ON tmp2.rownum=ttwo.rownum LEFT OUTER JOIN
  (SELECT @rownum3:=@rownum3+1 as rownum, item, count FROM (SELECT three as item, count(three) as count from  table_numbers group by three ORDER BY count DESC, three) d, (SELECT @rownum3:=0) r)  tthree ON tmp2.rownum=tthree.rownum 
WHERE tone.item IS NOT NULL OR ttwo.item IS NOT NULL OR tthree.item IS NOT NULL

上述查询的结果如下所示：

One  Count(one) Two Count(two)  Three   Count(three)
-----------------------------------------------------   
43   2          12  2           3       3            
3    1          1   1           2       2       
4    1          2   1           52      1        
11   1          31  1           null    null            
12   1          32  1           null    null                
------------------------------------------------------

Answer 2

你碰巧是这个意思吗？：

select one, two, three, four, five, six, count(1) 
from table_numbers group by one, two, three, four, five, six
order by count(1) desc

你想找到重复的吗？

Answer 3

我怀疑在单一查询中这样做的可能性!! 查询在执行时只遍历一次..在您的情况下，它需要计数不止一列。

Answer 4

你的意思是？

SELECT
    one AS val, COUNT(*) AS c
FROM ( 
    SELECT one FROM table_numbers UNION ALL
    SELECT two FROM table_numbers UNION ALL
    SELECT three FROM table_numbers UNION ALL
    SELECT four FROM table_numbers UNION ALL
    SELECT five FROM table_numbers UNION ALL
    SELECT six FROM table_numbers 
) x GROUP BY one ORDER BY c DESC;

Answer 5

我看到可以做到的唯一方法是...将表中的数据导出为一个长整数字符串。即一个平面文件，并将数据作为一个字段导回到新表中。编写sql语句将副本外推到新字段并进行总计数。执行相同操作，直到字段中导入的数据为空。希望有所帮助。

MySQL：count，group by和order by：整个表数据？

5 个答案: