Question

我创建了2个表格如下

表ForgeRock：

id  status  amount    Name            date
1     0      4500     ram           04/02/2012
2     0      2000    shyam          05/09/2013
4     0      1500    ghanshyam      08/06/2015

表ForgeRock1：

id   status   amount    Name          date
3      0      4500     gopal       04/02/2012
2      0      8000   radheshyam    15/11/2013
4      1      1500    ghanshyam    08/06/2015

SQL查询

Select f1.id,
      case when f1.status = 1 then f1.status else 

f2.status end as Status,
     case when f1.status = 1 then f1.amount else 

f2.amount end as rank,
      f1.Name,
      case when f1.status = 1 then f1.date else 

f2.date end as date
      from ForgeRock f1
 inner join ForgeRock1 f2 on f1.id = f2.id and 

f1.Name=f2.Name 

     union all


 Select id,
      Status,
      amount,
      Name,
      date
      from ForgeRock
   where id not in (select distinct id from 

ForgeRock1)

union all

 Select id,
      Status,
      amount,
      Name,
      date
      from ForgeRock1
   where id not in (select distinct id from 

ForgeRock)

union all

Select f1.id,
      case when f1.status = 1 then f1.status else 

f2.status end as Status,
     case when f1.status = 1 then f1.amount else 

f2.amount end as rank,
      f1.Name,
      case when f1.status = 1 then f1.date else 

f2.date end as date
      from ForgeRock f1
 inner join ForgeRock1 f2 on f1.id = f2.id and 

(f1.Name!=f2.Name and f1.date!=f2.date) 

union all

Select f1.id,
      case when f1.status = 1 then f1.status else 

f2.status end as Status,
     case when f1.status = 1 then f1.amount else 

f2.amount end as rank,
      f1.Name,
      case when f1.status = 1 then f1.date else 

f2.date end as date
      from ForgeRock1 f1
 inner join ForgeRock f2 on f1.id = f2.id and 

(f1.Name!=f2.Name and f1.date!=f2.date)

结果：

id  Status  rank    Name          date
4     1     1500    ghanshyam   08/06/2015
1     0     4500    ram         04/02/2012
3     0     4500    gopal       04/02/2012
2     0     8000    shyam       15/11/2013
2     0     2000    radheshyam  05/09/2013

现在我想将重复记录ID更新为新ID （即上面的结果2是Duplicate，所以将一个记录id从2改为21）

Answer 1

在 SQL-Server 中，您可以尝试使用ROW_NUMBER()获取重复项，并将其替换为Common Table Expression，如下所示：

WITH cte1 AS (
Select f1.id,
      case when f1.status = 1 then f1.status else 

f2.status end as Status,
     case when f1.status = 1 then f1.amount else 

f2.amount end as rank,
      f1.Name,
      case when f1.status = 1 then f1.date else 

f2.date end as date
      from ForgeRock f1
 inner join ForgeRock1 f2 on f1.id = f2.id and 

f1.Name=f2.Name 

     union all


 Select id,
      Status,
      amount,
      Name,
      date
      from ForgeRock
   where id not in (select distinct id from 

ForgeRock1)

union all

 Select id,
      Status,
      amount,
      Name,
      date
      from ForgeRock1
   where id not in (select distinct id from 

ForgeRock)

union all

Select f1.id,
      case when f1.status = 1 then f1.status else 

f2.status end as Status,
     case when f1.status = 1 then f1.amount else 

f2.amount end as rank,
      f1.Name,
      case when f1.status = 1 then f1.date else 

f2.date end as date
      from ForgeRock f1
 inner join ForgeRock1 f2 on f1.id = f2.id and 

(f1.Name!=f2.Name and f1.date!=f2.date) 

union all

Select f1.id,
      case when f1.status = 1 then f1.status else 

f2.status end as Status,
     case when f1.status = 1 then f1.amount else 

f2.amount end as rank,
      f1.Name,
      case when f1.status = 1 then f1.date else 

f2.date end as date
      from ForgeRock1 f1
 inner join ForgeRock f2 on f1.id = f2.id and 

(f1.Name!=f2.Name and f1.date!=f2.date)
), cte2 AS (
SELECT id, [status], rank, name, [date], RN
FROM(
SELECT id, [status], rank, name, [date], ROW_NUMBER() OVER (PARTITION BY Id ORDER BY Id) RN
from (
Select f1.id,
      case when f1.status = 1 then f1.status else 

f2.status end as Status,
     case when f1.status = 1 then f1.amount else 

f2.amount end as rank,
      f1.Name,
      case when f1.status = 1 then f1.date else 

f2.date end as date
      from ForgeRock f1
 inner join ForgeRock1 f2 on f1.id = f2.id and 

f1.Name=f2.Name 

     union all


 Select id,
      Status,
      amount,
      Name,
      date
      from ForgeRock
   where id not in (select distinct id from 

ForgeRock1)

union all

 Select id,
      Status,
      amount,
      Name,
      date
      from ForgeRock1
   where id not in (select distinct id from 

ForgeRock)

union all

Select f1.id,
      case when f1.status = 1 then f1.status else 

f2.status end as Status,
     case when f1.status = 1 then f1.amount else 

f2.amount end as rank,
      f1.Name,
      case when f1.status = 1 then f1.date else 

f2.date end as date
      from ForgeRock f1
 inner join ForgeRock1 f2 on f1.id = f2.id and 

(f1.Name!=f2.Name and f1.date!=f2.date) 

union all

Select f1.id,
      case when f1.status = 1 then f1.status else 

f2.status end as Status,
     case when f1.status = 1 then f1.amount else 

f2.amount end as rank,
      f1.Name,
      case when f1.status = 1 then f1.date else 

f2.date end as date
      from ForgeRock1 f1
 inner join ForgeRock f2 on f1.id = f2.id and 

(f1.Name!=f2.Name and f1.date!=f2.date) )x
) x2
WHERE RN > 1
)
SELECT CASE WHEN cte1.Id = cte2.Id THEN REPLACE(cte1.Id, cte2.Id, CAST(cte1.Id AS NVARCHAR(20)) + '1') ELSE cte1.Id END AS Id, cte1.[Status], cte1.[rank], cte1.name, cte1.[date]
FROM cte1
LEFT JOIN cte2 ON cte1.[rank] = cte2.[rank] AND cte1.name = cte2.name

输出：

Id  Status  rank      name        date
4     1     1500    ghanshyam   2015-08-06
1     0     4500    ram         2012-04-02
3     0     4500    gopal       2012-04-02
2     0     8000    shyam       2013-09-11
21    0     2000    radheshyam  2013-05-09

Sql查询更新选择

1 个答案: