使用重复记录更新sql中的查询

Question

我有一个表(数百万条记录)与(dataid，url)的唯一索引。表格如下所示：

id  dataid url
1   230    https://www.example.com/123
3   230    http://example.com/123

我无法运行查询

UPDATE table_name SET url = REPLACE(url, 'http://', 'https://www.')

因为存在重复和违反唯一键约束的情况。在这种情况下，我想删除'id‘值最大的记录。我该怎么做呢？

Answer 1

delete 
    from table a 
    join table b on a.dataid = b.dataid 
where 
    a.dataid = 230 and a.id > b.id;

试试看

Answer 2

这将找出应该删除的行

select max(id), REPLACE(url, 'http://', 'https://www.') as url from table
group by REPLACE(url, 'http://', 'https://www.') 
having count(*)>1

这将删除它们

delete t1 from table as t1 inner join
(
select max(id), REPLACE(url, 'http://', 'https://www.') as url from table
group by REPLACE(url, 'http://', 'https://www.') 
having count(*)>1
) as t2 on t1.id=t2.id

现在更新您的数据

UPDATE table_name SET url = REPLACE(url, 'http://', 'https://www.')

Answer 3

delete tst
where id in (select max(id)
             from tst
             group by dataid, REPLACE(url, 'http://', 'https://www.')
             having count(*) = 2);

UPDATE tst SET url = REPLACE(url, 'http://', 'https://www.');