How do I prevent MySQL from auto-incrementing the Primary Key while using ON DUPLICATE KEY UPDATE when the duplicate is a different unique column?

Question

Consider the following table:

+-------------+------------------+------+-----+---------+----------------+
| Field       | Type             | Null | Key | Default | Extra          |
+-------------+------------------+------+-----+---------+----------------+
| vendor_id   | int(10) unsigned | NO   | PRI | NULL    | auto_increment |
| vendor_name | varchar(100)     | NO   | UNI | NULL    |                |
| count       | int(10) unsigned | NO   |     | 1       |                |
+-------------+------------------+------+-----+---------+----------------+

I have the following MySQL query:

INSERT INTO `table` 
   (`vendor_name`) 
VALUES 
   ('foobar') ON DUPLICATE KEY UPDATE `count` = `count` + 1

The intent of this query is to insert a new vendor name to the table and in case the vendor name already exists, the column count should be incremented by 1. This works however the primary key of the current column will also be auto-incremented. How can I prevent MySQL from auto-incrementing the primary key in these cases? Is there a way to do this with one query?

Thank you.

Answer 1

This works however the primary key of the current column will also be auto-incremented. How can I prevent MySQL from auto-incrementing the primary key in these cases?

By using an UPDATE statement when the value already exists:

IF EXISTS(SELECT NULL
            FROM TABLE
           WHERE vendor_name = $vendor_name) THEN

    UPDATE TABLE
       SET count = count + 1
     WHERE vendor_name = $vendor_name

ELSE

    INSERT INTO TABLE
       (vendor_name)
    VALUES
       ($vendor_name

END IF

I tried the alternative to ON DUPLICATE KEY UPDATE, REPLACE INTO:

REPLACE INTO vendors SET vendor_name = 'foobar', COUNT = COUNT + 1

It updates the count, and the vendor_id so it's worse...

The database & data doesn't care if the numbers aren't sequential, only that the values are unique. If you can live with that, I'd use the ON DUPLICATE UPDATE syntax though I admit the behaviour is weird (understandable considering using an INSERT statement).

Answer 2

I think this might do it. But it's very much against the principles of Daoism - you're really going against the grain.

There is probably a better solution.

INSERT INTO `table` 
   (`vendor_name`) 
VALUES 
   ('foobar') ON DUPLICATE KEY UPDATE `count` = `count` + 1, `vendor_id`=`vendor_id`-1