How to 'insert if not exists' in MySQL?

2009-09-01 mysql sql primary-key sql-insert

I started by googling, and found this article which talks about mutex tables.

I have a table with ~14 million records. If I want to add more data in the same format, is there a way to ensure the record I want to insert does not already exist without using a pair of queries (ie, one query to check and one to insert is the result set is empty)?

Does a unique constraint on a field guarantee the insert will fail if it's already there?

It seems that with merely a constraint, when I issue the insert via php, the script croaks.

Answers

Any simple constraint should do the job, if an exception is acceptable. Examples :

  • primary key if not surrogate
  • unique constraint on a column
  • multi-column unique constraint

Sorry is this seems deceptively simple. I know it looks bad confronted to the link you share with us. ;-(

But I neverleless give this answer, because it seem to fill your need. (If not, it may trigger your updating your requirements, which would be "a Good Thing"(TM) also).

Edited: If an insert would break the database unique constraint, an exception is throw at the database level, relayed by the driver. It will certainly stop your script, with a failure. It must be possible in PHP to adress that case ...

use INSERT IGNORE INTO table

see http://bogdan.org.ua/2007/10/18/mysql-insert-if-not-exists-syntax.html

there's also INSERT … ON DUPLICATE KEY UPDATE syntax, you can find explanations on dev.mysql.com


Post from bogdan.org.ua according to Google's webcache:

18th October 2007

To start: as of the latest MySQL, syntax presented in the title is not possible. But there are several very easy ways to accomplish what is expected using existing functionality.

There are 3 possible solutions: using INSERT IGNORE, REPLACE, or INSERT … ON DUPLICATE KEY UPDATE.

Imagine we have a table:

CREATE TABLE `transcripts` (
`ensembl_transcript_id` varchar(20) NOT NULL,
`transcript_chrom_start` int(10) unsigned NOT NULL,
`transcript_chrom_end` int(10) unsigned NOT NULL,
PRIMARY KEY (`ensembl_transcript_id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1;

Now imagine that we have an automatic pipeline importing transcripts meta-data from Ensembl, and that due to various reasons the pipeline might be broken at any step of execution. Thus, we need to ensure two things:

  1. repeated executions of the pipeline will not destroy our database

  2. repeated executions will not die due to ‘duplicate primary key’ errors.

Method 1: using REPLACE

It’s very simple:

REPLACE INTO `transcripts`
SET `ensembl_transcript_id` = 'ENSORGT00000000001',
`transcript_chrom_start` = 12345,
`transcript_chrom_end` = 12678;

If the record exists, it will be overwritten; if it does not yet exist, it will be created. However, using this method isn’t efficient for our case: we do not need to overwrite existing records, it’s fine just to skip them.

Method 2: using INSERT IGNORE Also very simple:

INSERT IGNORE INTO `transcripts`
SET `ensembl_transcript_id` = 'ENSORGT00000000001',
`transcript_chrom_start` = 12345,
`transcript_chrom_end` = 12678;

Here, if the ‘ensembl_transcript_id’ is already present in the database, it will be silently skipped (ignored). (To be more precise, here’s a quote from MySQL reference manual: “If you use the IGNORE keyword, errors that occur while executing the INSERT statement are treated as warnings instead. For example, without IGNORE, a row that duplicates an existing UNIQUE index or PRIMARY KEY value in the table causes a duplicate-key error and the statement is aborted.”.) If the record doesn’t yet exist, it will be created.

This second method has several potential weaknesses, including non-abortion of the query in case any other problem occurs (see the manual). Thus it should be used if previously tested without the IGNORE keyword.

Method 3: using INSERT … ON DUPLICATE KEY UPDATE:

Third option is to use INSERT … ON DUPLICATE KEY UPDATE syntax, and in the UPDATE part just do nothing do some meaningless (empty) operation, like calculating 0+0 (Geoffray suggests doing the id=id assignment for the MySQL optimization engine to ignore this operation). Advantage of this method is that it only ignores duplicate key events, and still aborts on other errors.

As a final notice: this post was inspired by Xaprb. I’d also advise to consult his other post on writing flexible SQL queries.

on duplicate key update, or insert ignore can be viable solutions with MySQL.


Example of on duplicate key update update based on mysql.com

INSERT INTO table (a,b,c) VALUES (1,2,3)
  ON DUPLICATE KEY UPDATE c=c+1;

UPDATE table SET c=c+1 WHERE a=1;

Example of insert ignore based on mysql.com

INSERT [LOW_PRIORITY | DELAYED | HIGH_PRIORITY] [IGNORE]
    [INTO] tbl_name [(col_name,...)]
    {VALUES | VALUE} ({expr | DEFAULT},...),(...),...
    [ ON DUPLICATE KEY UPDATE
      col_name=expr
        [, col_name=expr] ... ]

Or:

INSERT [LOW_PRIORITY | DELAYED | HIGH_PRIORITY] [IGNORE]
    [INTO] tbl_name
    SET col_name={expr | DEFAULT}, ...
    [ ON DUPLICATE KEY UPDATE
      col_name=expr
        [, col_name=expr] ... ]

Or:

INSERT [LOW_PRIORITY | HIGH_PRIORITY] [IGNORE]
    [INTO] tbl_name [(col_name,...)]
    SELECT ...
    [ ON DUPLICATE KEY UPDATE
      col_name=expr
        [, col_name=expr] ... ]

Solution:

INSERT INTO `table` (`value1`, `value2`) 
SELECT 'stuff for value1', 'stuff for value2' FROM DUAL 
WHERE NOT EXISTS (SELECT * FROM `table` 
      WHERE `value1`='stuff for value1' AND `value2`='stuff for value2' LIMIT 1) 

Explanation:

The innermost query

SELECT * FROM `table` 
      WHERE `value1`='stuff for value1' AND `value2`='stuff for value2' LIMIT 1

used as the WHERE NOT EXISTS-condition detects if there already exists a row with the data to be inserted. After one row of this kind is found, the query may stop, hence the LIMIT 1 (micro-optimization, may be omitted).

The intermediate query

SELECT 'stuff for value1', 'stuff for value2' FROM DUAL

represents the values to be inserted. DUAL refers to a special one row, one column table present by default in all Oracle databases (see https://en.wikipedia.org/wiki/DUAL_table). On a MySQL-Server version 5.7.26 I got a valid query when omitting FROM DUAL, but older versions (like 5.5.60) seem to require the FROM information. By using WHERE NOT EXISTS the intermediate query returns an empty result set if the innermost query found matching data.

The outer query

INSERT INTO `table` (`value1`, `value2`) 

inserts the data, if any is returned by the intermediate query.

Here is a PHP function that will insert a row only if all the specified columns values don't already exist in the table.

  • If one of the columns differ, the row will be added.

  • If the table is empty, the row will be added.

  • If a row exists where all the specified columns have the specified values, the row won't be added.

    function insert_unique($table, $vars)
    {
      if (count($vars)) {
        $table = mysql_real_escape_string($table);
        $vars = array_map('mysql_real_escape_string', $vars);
    
        $req = "INSERT INTO `$table` (`". join('`, `', array_keys($vars)) ."`) ";
        $req .= "SELECT '". join("', '", $vars) ."' FROM DUAL ";
        $req .= "WHERE NOT EXISTS (SELECT 1 FROM `$table` WHERE ";
    
        foreach ($vars AS $col => $val)
          $req .= "`$col`='$val' AND ";
    
        $req = substr($req, 0, -5) . ") LIMIT 1";
    
        $res = mysql_query($req) OR die();
        return mysql_insert_id();
      }
    
      return False;
    }
    

Example usage :

<?php
insert_unique('mytable', array(
  'mycolumn1' => 'myvalue1',
  'mycolumn2' => 'myvalue2',
  'mycolumn3' => 'myvalue3'
  )
);
?>
REPLACE INTO `transcripts`
SET `ensembl_transcript_id` = 'ENSORGT00000000001',
`transcript_chrom_start` = 12345,
`transcript_chrom_end` = 12678;

If the record exists, it will be overwritten; if it does not yet exist, it will be created.

Try the following:

IF (SELECT COUNT(*) FROM beta WHERE name = 'John' > 0)
  UPDATE alfa SET c1=(SELECT id FROM beta WHERE name = 'John')
ELSE
BEGIN
  INSERT INTO beta (name) VALUES ('John')
  INSERT INTO alfa (c1) VALUES (LAST_INSERT_ID())
END

There are several answers that cover how to solve this if you have a UNIQUE index that you can check against with ON DUPLICATE KEY or INSERT IGNORE. That is not always the case, and as UNIQUE has a length constraint (1000 bytes) you might not be able to change that. For example, I had to work with metadata in WordPress (wp_postmeta).

I finally solved it with two queries:

UPDATE wp_postmeta SET meta_value = ? WHERE meta_key = ? AND post_id = ?;
INSERT INTO wp_postmeta (post_id, meta_key, meta_value) SELECT DISTINCT ?, ?, ? FROM wp_postmeta WHERE NOT EXISTS(SELECT * FROM wp_postmeta WHERE meta_key = ? AND post_id = ?);

Query 1 is a regular UPDATE query with no effect when the dataset in question is not there. Query 2 is an INSERT which depends on a NOT EXISTS, i.e. the INSERT is only executed when the dataset doesn't exist.

Something worth noting is that INSERT IGNORE will still increment the primary key whether the statement was a success or not just like a normal INSERT would.

This will cause gaps in your primary keys that might make a programmer mentally unstable. Or if your application is poorly designed and depends on perfect incremental primary keys, it might become a headache.

Look into innodb_autoinc_lock_mode = 0 (server setting, and comes with a slight performance hit), or use a SELECT first to make sure your query will not fail (which also comes with a performance hit and extra code).

Update or insert without known primary key

If you already have a unique or primary key, the other answers with either INSERT INTO ... ON DUPLICATE KEY UPDATE ... or REPLACE INTO ... should work fine (note that replace into deletes if exists and then inserts - thus does not partially update existing values).

But if you have the values for some_column_id and some_type, the combination of which are known to be unique. And you want to update some_value if exists, or insert if not exists. And you want to do it in just one query (to avoid using a transaction). This might be a solution:

INSERT INTO my_table (id, some_column_id, some_type, some_value)
SELECT t.id, t.some_column_id, t.some_type, t.some_value
FROM (
    SELECT id, some_column_id, some_type, some_value
    FROM my_table
    WHERE some_column_id = ? AND some_type = ?
    UNION ALL
    SELECT s.id, s.some_column_id, s.some_type, s.some_value
    FROM (SELECT NULL AS id, ? AS some_column_id, ? AS some_type, ? AS some_value) AS s
) AS t
LIMIT 1
ON DUPLICATE KEY UPDATE
some_value = ?

Basically, the query executes this way (less complicated than it may look):

  • Select an existing row via the WHERE clause match.
  • Union that result with a potential new row (table s), where the column values are explicitly given (s.id is NULL, so it will generate a new auto-increment identifier).
  • If an existing row is found, then the potential new row from table s is discarded (due to LIMIT 1 on table t), and it will always trigger an ON DUPLICATE KEY which will UPDATE the some_value column.
  • If an existing row is not found, then the potential new row is inserted (as given by table s).

Note: Every table in a relational database should have at least a primary auto-increment id column. If you don't have this, add it, even when you don't need it at first sight. It is definitely needed for this "trick".

Related