sql-- 找到重複數據並刪除、有重複數據不插入或更新的處理方法

表結構:

需求:找到相同的內容並刪除

方法1:

經過分組找出篩選出count大於1的數據

select r_content, r_userid, count(*) from reviews GROUP BY r_content, r_userid having count(*) > 1

 group by能夠利用聚合函數(count、 sum、 avg)進行分組

having 對分組的數據進行下一步篩選

 

 經過inner join找出相同數據

select *
from reviews a
         INNER JOIN
     (select r_content, r_userid, count(*) from reviews GROUP BY r_content, r_userid having count(*) > 1) b
     on a.r_content = b.r_content and a.r_userid = b.r_userid;

加行號

select a.*,
       if(@tmp = CONCAT(r_content, r_userid), @rownum := @rownum + 1, @rownum := 1) as rownum,
       @tmp := CONCAT(r_content, r_userid)
from (
         select a.*
         from reviews a
                  INNER JOIN
              (select r_content, r_userid, count(*) from reviews GROUP BY r_content, r_userid having count(*) > 1) b
              on a.r_content = b.r_content and a.r_userid = b.r_userid) a,
     (select @rownum := 0, @tmp := '') b

取出行號大於的刪除之

delete from reviews where r_id in(
select r_id from (
select a.*,
       if(@tmp = CONCAT(r_content, r_userid), @rownum := @rownum + 1, @rownum := 1) as rownum,
       @tmp := CONCAT(r_content, r_userid)
from (
         select a.*
         from reviews a
                  INNER JOIN
              (select r_content, r_userid, count(*) from reviews GROUP BY r_content, r_userid having count(*) > 1) b
              on a.r_content = b.r_content and a.r_userid = b.r_userid) a,
     (select @rownum := 0, @tmp := '') b) a where rownum>1);

方法2:

經過group_concat來組合id,並刪除

select group_concat(ids separator '|') as ids from(
select group_concat(r_id) as ids ,r_content,r_userid,count(*) from  reviews  group by r_content,r_userid having count(*)>1) a

 

有重複數據不插入或更新的處理方法

表設計:

需求:若是有相同內容則不插入,記錄插入相同數據的條數

insert  into news (news_title, news_abstract,news_code) values ('這是一條新聞','新聞內容是XXX',md5(concat('這是一條新聞','新聞內容是XXX'))) on duplicate key update dumpnum=dumpnum+1;