Question

我有一个表，其中包含两个变量(a, b)和一个id的记录，就像这样：

a,b,id    
x,y,1    
x,y,2    
x,z,3    
t,z,4    
t,y,5
t,y,6
t,y,7

对于每个id，我想使用mysql检查每对（a，b）变量是否已经出现，并将其分配给变量。例如，上一个输入的输出应为：

a,b,id,repeated    
x,y,1,1 - this should be 1 not zero    
x,y,2,1    
x,z,3,0   
t,z,4,0    
t,y,5,0
t,y,6,1
t,y,7,1

我应该对一个较大的表（700000行）执行此操作，因此我宁愿不使用慢速方法。有什么想法吗？

我们也不能假设只有先前的ID具有重复的数据，因此查询必须考虑到ID彼此不相等。

Answer 1

您可以使用User-defined Session variables进行此操作。

DB Fiddle DEMO

SELECT 
  IF(@a_var = dt.a AND 
     @b_var = dt.b, 
     1, 
     0) AS repeated, 
  @a_var := dt.a AS a, 
  @b_var := dt.b AS b, 
  dt.id 
FROM 
(
  SELECT 
    a, 
    b, 
    id
  FROM your_table 
  ORDER BY a,b,id
) AS dt 
CROSS JOIN (SELECT @a_var := '', 
                   @b_var := '') AS user_init_vars

Answer 2

我们可以在此处使用相关子查询：

SELECT
    a,
    b,
    id,
    CASE WHEN EXISTS (SELECT 1 FROM yourTable t2
                      WHERE t1.a = t2.a AND t1.b = t2.b AND t2.id < t1.id)
         THEN 1 ELSE 0 END AS repeated
FROM yourTable t1
ORDER BY
    id;

Demo

Answer 3

尝试此查询：

select @aLag := '1', @bLag := '1';
select id, a, b, isRepeated from (
  select case when @aLag = a and @bLag = b then 1 else 0 end isRepeated,
         @ALag := a a,
         @bLag := b b,
         id
  from tbl
  order by a, b, id
) a order by id

Demo

Answer 4

这是一种不相关的方法...

WITH yourTable AS (
    SELECT 'x' AS a, 'y' AS b, 1 AS id UNION ALL
    SELECT 'x', 'y', 2 UNION ALL
    SELECT 'x', 'z', 3 UNION ALL
    SELECT 't', 'z', 4 UNION ALL
    SELECT 't', 'y', 5 UNION ALL
    SELECT 't', 'y', 6 UNION ALL
    SELECT 't', 'y', 7
)

SELECT DISTINCT x.*
              , y.a IS NOT NULL repeated
  FROM yourTable x
  LEFT
  JOIN yourTable y 
    ON y.a = x.a 
   AND y.b = x.b
   AND y.id < x.id;

https://dbfiddle.uk/?rdbms=mysql_8.0&fiddle=924d776f772007af9d46f36a5daaf8e2

检查重复记录的mysql

4 个答案:

DB Fiddle DEMO

Demo