Primary key repeat in MySQL database 3 Solution

The following article describes the actual processing scheme for repeated primary keys in the MySQL database. We all know that when the data is inserted into the data table, when the unique key (unique key and primary key) is repeated, a duplicate key error occurs.

There are three solutions to this problem: the following data structure is an example

 
 

  
  
MySQL> use test;  
  
  
MySQL> create table `user` (`userid` int(11) DEFAULT NULL, `username` varchar(255) NOT NULL DEFAULT '');  
 
 

Add the userid column primary key

 
 

  
  
MySQL> alter table `user` add primary key `userid` (`userid`); 
 
 

Insert data

 
 

  
  
MySQL> insert into `user` values (1, 'eric'), (2, 'jesus');  
 
 

Now I want to insert or edit a record with userid 1, but I don't know if it already exists.

Actual solution 1 when the primary key in the MySQL database is repeated. Delete the key before inserting it.

 
 

  
  
MySQL> delete from user where userid = 1;  
  
  
MySQL> insert into user values (1, 'xxxxx') ;  
 
 

The actual solution 2 when the primary key in the MySQL database is repeated. Use replace

 
 

  
  
MySQL> replace into user values (1, 'newvalue');  
 
 

In this case, the logic is like this. MySQL first checks whether the record exists. If yes, it first deletes the record and then inserts the record independently. so you can see that affected rows is two after this statement is executed (the premise is that there is only one record of userid 1 in your data table)

Actual solution 3 when the primary key in the MySQL database is repeated

 
 

  
  
insert into ... on duplicate key update  
  
  
MySQL> insert into user1 values (1, 'newvalueagain') on duplicate key update user1.username = VALUES(username);  
 
 

The affected rows of this statement is also 2.

Of course, another way to handle this problem is to use php directly,

Select first. If no result is found, insert; otherwise, update.

You can also update it first and find that affected rows is 0, then insert.

But obviously neither of these methods directly handed over the work to MySQL for high processing efficiency