MySQL 查询删除两个数据库 RDS（MYSQL DB）中常见的数据

Question

eg:

         **table 1**

 name     gender  email

 hemanth    m       test@gmail.com 
 rajesh     m       rajesh@gmail.com 
 kumar      m       kumar@gmail.com

      **table 2***

 name     gender    email

 hemanthsai m       testing@gmail.com 
 rajesh     m       rajesh@gmail.com 
 kumaras    f       kumaras@gmail.com

     **expected result in table 1***

 name     gender    email

 hemanthsai m       testing@gmail.com 
 rajesh     m       rajesh@gmail.com 
 kumaras    f       kumaras@gmail.com

this is just an example, in reality i have to update all the data with many columns at a time in mysql

What I've tried was

     INSERT INTO table1 (NAME,gender, email) VALUES(?, ?, ?), (?, ?, ?)
     ON DUPLICATE KEY UPDATE SET *;    ### this didn""t worked ###

I need something like:

     INSERT INTO table1 * ON DUPLICATE KEY UPDATE SET *; 

If this didn't work then I thought of delete common values on one table using another table and then insert the left values into the deleted table, so I need query to delete common values on one table based on another table..

thanks in advance,

Answer 1

Have you tried INSERT INTO SELECT ?

Example:

INSERT INTO `table1`
(`name`, `gender`, `email`)
SELECT t2.`name`, t2.`gender`, t2.`email`
FROM `table2` AS t2
ON DUPLICATE KEY UPDATE
`name`= t2.`name`,`gender` = t2.`gender`, `email` = t2.`email`

You also need to have a unique key which is common between table1 and table2. I assume that is the case.

EDIT

Taking your comment into account, what you are looking for might be REPLACE INTO .

REPLACE INTO `table1` 
SELECT *
FROM `table2`

Please note that this operation doesn't update the old values, but actually deletes them. This means that the PRIMARY_KEY might be replaced if it has for instance auto_increment() . Also note that this is a heavy operation as the table needs to be reindexed for every delete operation.

You can read more about it here: http://code.openark.org/blog/mysql/replace-into-think-twice

Personally I would use INSERT INTO even if there are many columns

Answer 2

Create another table:

CREATE TABLE table3 (
  name VARCHAR(50),
  gender VARCHAR(50),
  email VARCHAR(50));

Create a unique index on the new table:

CREATE UNIQUE INDEX index_name
ON table3 (name, gender, email);

Do INSERT IGNORE from both of table1 & table2:

INSERT IGNORE INTO table3 SELECT * FROM table1;

INSERT IGNORE INTO table3 SELECT * FROM table2;

Check your table3 , it should't have any duplicates anymore.

Demo

P/S: If you want to only check duplicate email - and allow same user having multiple emails, you can skip creating unique index and make email column in table3 as PRIMARY KEY instead.