(PHP 4, PHP 5)

mysql_insert_idВозвращает идентификатор, сгенерированный при последнем INSERT-запросе


int mysql_insert_id ([ resource $link_identifier = NULL ] )

Возвращает идентификатор, сгенерированный колонкой с AUTO_INCREMENT последним запросом (обычно INSERT).

Список параметров


Соединение MySQL. Если идентификатор соединения не был указан, используется последнее соединение, открытое mysql_connect(). Если такое соединение не было найдено, функция попытается создать таковое, как если бы mysql_connect() была вызвана без параметров. Если соединение не было найдено и не смогло быть создано, генерируется ошибка уровня E_WARNING.

Возвращаемые значения

Идентификатор, сгенерированный колонкой с AUTO_INCREMENT последним запросом в случае успеха , 0, если последний запрос не генерирует значение AUTO_INCREMENT value, и FALSE, если соединение MySQL не было установлено.


Пример #1 Пример использования mysql_insert_id()

if (!
$link) {
'Ошибка соединения: ' mysql_error());

mysql_query("INSERT INTO mytable (product) values ('kossu')");
printf("Идентификатор последней вставленной записи %d\n"mysql_insert_id());



mysql_insert_id() конвертирует возвращаемый функцией MySQL C API тип значения функции mysql_insert_id() в тип long (называемый int в PHP). Если ваша колонка AUTO_INCREMENT имеет тип BIGINT (64 бита), то значение, возвращаемое функцией в результате преобразования может быть искажено. Используйте вместо данной функции внутреннюю MySQL-функцию LAST_INSERT_ID() в SQL-запросе. Подробнее о максимальных значениях целых чисел смотрите в разделе документации, посвященном целым числам.


Так как mysql_insert_id() работает с последним выполненным запросом, вызывайте mysql_insert_id() сразу же после запроса, генерирующего новое значение.


Значение в SQL функции MySQL LAST_INSERT_ID() всегда содержит последний сгенерированный ID и не обнуляется между запросами.

Смотрите также

  • mysql_query() - Посылает запрос MySQL
  • mysql_info() - Возвращает информацию о последнем запросе


You can't do an INSERT DELAYED and expect to get anything but zero, for it runs in a separate thread, and mysql_insert_id() is tied to the current thread.
1999-12-09 19:14:15
If you want to use the ID that was generated for one table and insert it into a second table, you can use SQL statements like this: 

INSERT INTO foo (auto,text)
    VALUES(NULL,'text');              # generate ID by inserting NULL
INSERT INTO foo2 (id,text)
    VALUES(LAST_INSERT_ID(),'text');  # use ID in second table

...found here:

It works even without inserting the NULL value for some reason ;)
The following is great for monitoring:
    $new_id = mysql_insert_id();
    print "New id: $new_id\n";

Hope it helps you all, cheers.
2001-08-04 02:17:14
Just a reminder, mysql_insert_id() should be called after 'mysql_affected_rows()', but BEFORE 'mysql_query("COMMIT")'.
2002-12-16 20:30:47
This might be obvious, but it tripped me up - be careful when using last_insert_id() with persistent connections - running last_insert_id() after a failed update/insert/etc will return the last insert id of the last successful update/insert made by that CONNECTION rather than 0 for the number of rows updated by the previous non-working query, and who knows what the last query run on that connection was.
2003-04-24 17:36:43
To get the NEXT insert id use the mysql query SHOW TABLE STATUS LIKE 'tablename' and get the field auto_increment...
2003-11-13 17:29:26
any zerofills on your id get chopped off on this function because it returns an int.
2004-02-11 18:36:37
In response to treylane at example dot com.

It is very very very important that you put in an "or die" or some other form of error handling.

Some scripts can fail invisibly and insert invalid data throughout your whole database because of mysql_insert_id
inserting the last successful insertid rather than reporting that the last attempt failed.

example of an or die statement.

$result = mysql_query($sql)
or die("Invalid query: " . mysql_error());
$EventID = mysql_insert_id();
2004-02-26 05:40:08
If you use this function after doing an INSERT ... SELECT to insert multiple rows at once, you get the autonumber ID of the *first* row added by the INSERT.

e.g. if there are 4 records in table 'init' that have column 'type' = 2
I want to add these 4 records to table 'game'
Table game has an autonumber column 'game_id' that is currently at 32.

If I do this query:

INSERT INTO game (type, players, rounds)
SELECT type, players, rounds FROM init
WHERE type = 2

Then mysql_insert_id() will return 33, not 36.
2004-06-24 14:47:41
I believe the "resource link" being referred to is not what is returned from mysql_query() but the $link returned from mysql_connect(). mysql_insert_id() will just use the most recent connection if there is no explicit $link being used.

So the above example in the manual page itself should behave the same with mysql_insert_id($link) at the end instead of the mysql_insert_id() they used. If you had multiple connections, the $link might come in handy.

Also in reading the mysql manual itself, there is some enlightening information on the fact that this does appear to be totally safe to use because it is on a per-connection basis.

Here's the relevant quote from the manual on LAST_INSERT_ID() which is located here:

"The last ID that was generated is maintained in the server on a per-connection basis. This means the value the function returns to a given client is the most recent AUTO_INCREMENT value generated by that client. The value cannot be affected by other clients, even if they generate AUTO_INCREMENT values of their own. This behavior ensures that you can retrieve your own ID without concern for the activity of other clients, and without the need for locks or transactions."

Sounds safe to me. I couldn't imagine this would be done any other way *but* on a per-connection basis, otherwise chaos would ensue. The only way to test it would be to perform a multi-thread type test. Perhaps someone is up for it and wants to post their results somewhere? :)
2004-10-01 02:04:00
In reply to: sly at noiretblanc dot org:

Make sure that auto_increment has an capital A as the first letter, otherwise it WON'T work! So you have to spell it as Auto_increment... And then it works fine.
2004-11-10 19:38:12
It's possible to do the same with M$ Server.

function odbc_insert_id()
  $query = "SELECT @@IDENTITY AS ID;";
  $result = odbc_exec($this->m_rConnectionID, $query);
  $row = odbc_fetch_object($result);
  return $row->ID;
2005-02-15 14:37:05
A bit more on return values:
mysql_insert_id() returns 0 if you haven't actually manipulated anything yet.

Also, it returns 0 even if the DB connection is lost[0] between inserting and calling mysql_insert_id() - so you can always count on getting an integer.

[0] By 'lost' I mean e.g. a DB crash. Actually closing the open link and then trying to communicate with the DB will of course result in an error.
2005-02-20 08:57:46
Beware, mysql_insert_id() only returns the value of the last syntaxically correct insert statement.

If your code has a problem and the insert is not understood by the server then the value of the last working insert command is returned.

Put something else in place such as "select count( id ) from table" before and after the mysql_insert_id() call to ensure that a row was inserted.
2005-05-08 13:25:13
Just a quick note. mysql_insert_id() does work with REPLACE.
2005-07-21 17:36:45
My apologies for the error below (that was modified out of a class) - as you cannot define a constant as an array.

replace the line:








$mysql_id is now an array in which the first element $mysql_id[0] holds the last inserted id.

Apologies if anyone struggled over that one... esp. the noobs.
2005-09-01 16:50:28
Take care of setting an empty value for the AUTO_INCREMENT Field else you never get a value except zero returned from mysq_insert_id() ....

Ciao Ephraim
2005-12-02 12:48:28
As mentioned by frumler at the-beach dot no_spam dot net

the LAST_INSERT_ID works like a charm when inserting values into tables.

I'm not sure why anyone would need mysql_insert_id() when LAST_INSERT_ID is readily available.


Say you have a table called "transaction" and a table called "accounts".  Obviously each account must be created using a transaction, so every time a record is created in the accounts table, a record must be created in the transaction table containing the same account_id(auto_increment) that was just created by mysql.

Here's a simple way to do this:

="INSERT INTO accounts (account_id,type) VALUES (NULL,'saving')";
$transaction_query="INSERT INTO transactions(transaction_id,account_id) VALUES (NULL,LAST_INSERT_ID)";


2005-12-06 23:53:28
If you insert a data row by using the ON DUPLICATE KEY UPDATE clause in an INSERT-statement, the mysql_insert_id() function will return not the same results as if you directly use LAST_INSERT_ID() in MySQL.

See the following example:

// insert a datarow, primary key is auto_increment
   // value is a unique key
$query "INSERT INTO test (value) VALUES ('test')";
mysql_query$query );

mysql_query"SELECT LAST_INSERT_ID()" ),
'<br>mysql_insert_id: ',


This will print:

mysql_insert_id: 1

In this case the function returns the same as the MySQL-Statement.
But see the insert on an existing key:

"INSERT INTO test (value) 
                  VALUES ('test')
                  ON DUPLICATE KEY UPDATE value = 'test2'"
mysql_query$query );

mysql_query"SELECT LAST_INSERT_ID()" ),
'<br>mysql_insert_id: ',


This will print:

mysql_insert_id: 1

By using the ON DUPLICATE KEY UPDATE clause, only the old datarow will be modified, if the INSERT statement causes a duplicate entry, but the LAST_INSERT_ID() function returns the next auto_increment value for the primary key, which is by the way not set as the next auto_increment value in the database.

The mysql_insert_id() function returns the primary key of the old (and changed) data row. For me this is the right operation method, because the LAST_INSERT_ID() function returns a value which is not referenced to a data row at all.

Greets from Munich.

2005-12-14 03:31:55
Be careful when using "insert ignore". If the unique index already exists, the record will not be added, but it WILL give you the id of the next auto_increment value even though it didn't create it.

"insert ignore into sometable set num=10";
mysql_query($sql) or die();

// same record, database is unique on 'num'
$sql "insert ignore into sometable set num=10";
mysql_query($sql) or die();

would give:

2006-01-31 16:55:28
When used in transactions, mysql_insert_id() MUST be called before committing. Otherwise, it will return unpredictable results.
2006-09-19 12:01:29
It's not true that mysql_insert_id() only returns the ID generated for an AUTO_INCREMENT column by the previous INSERT query.

We can use LAST_INSERT_ID() statement that will return value for mysql_insert_id().


$sql = "UPDATE `mytable` SET `mytable_id`= LAST_INSERT_ID(`mytable_id` + 1) WHERE ...;

$last_id = mysql_insert_id();

LAST_INSERT_ID() statement will affect mysql_insert_id().

2006-10-29 03:57:11
and the primary reason to use neither of those last two solutions is both could result in race conditions when there is more than a single user with access to the database. when you use MAX or ORDER BY DESC LIMIT 1 you will retrieve the maximum value from the table, at that moment. That doesn't mean another user doesn't do an insertion in the primary and secondary table BEFORE you do your insertion in the secondary table. You have thus inserted a row with a correct id into the second table. Always use last_insert_id()
2007-07-18 11:35:16
Forget about using MAX to get the last inserted id. Race conditions like other users inserting between your SELECT MAX(.. and your INSERT may render your id unusable.

The WAY to get the id is by using mysql_insert_id() or the mysql SQL function LAST_INSERT_ID(). 

Take care, if using mysql_insert_id() you should provide the resource returned by the mysql_connect, not the resultset returned by mysql_query.
2007-07-31 17:06:39
If mysql_insert_id() returns 0 or null, check your auto increment field is not being set by your sql query, also if you have multiple db connections like I did, the solution is to create a seperate db connection for this query.
2007-08-13 10:53:54
It seems that, in case of a very first INSERT, mysql_insert_id() returns 0. My guess is that MySQL does not trigger auto increment if the table is empty. Once an entry is present, it works as expected.
2007-09-20 16:13:18
Just wanted to re-iterate previous comment on receiving NULL or 0 return statement from calling mysql_insert_id() after insert statement.

If you have multiple mysql connections (i.e. mysql_connect() or mysqli_connect()) on the page you will need to specify the the connection you are using when calling this function!

2007-10-11 23:04:15
Using 'SELECT MAX(id)+1...' will not return the next auto_increment id. This function is totaly unreliable by two reasons.
1. In race conditions there is no guarantee that other user will not insert new record while your function have done its work. This will render your "generated" last_id obsolete. It is rare case but it happens.
2. Most of all, if the last record(s) in the table is deleted the max id will no longer match the auto_increment value, because auto_increment never repeats numbers, it increases whenever an insert statement is completed and does not decrease if you erase the last record!!! 

e.g. If you have this table with the last record deleted:
id name
1. car
2. plane
3. truck - [erased]

auto_increment is 4
but MAX(id) is 2!!!
2008-02-14 09:11:10
Why on earth are you all arguing about the best way to get the next auto_increment value? The whole point is that it increments automatically. The name should be a give away.

Don't calculate the next id, just leave the field blank and let the database issue the id itself.
2008-02-20 10:45:23
There's nothing inherently wrong with using auto-increment fields. There's also nothing wrong with the main competetive idea, which is for the database to supply a primitive sequence of non-repeating identifiers, typically integers. This is rather like which side of the road you drive on.

The bigger problem is when people don't understand what they are doing with database access. It's like driving a car without really knowing the rules of the road. Such people wind up making bad decisions without realizing it, and then, eventually, something breaks.

Databases are complex beasts, and worth taking the time to really understand. Learn about the implications and limitations of different approaches to solving problems. Then, you will be prepared to pick a solution based on what has to work.
2008-02-23 19:33:57
I don't get all the fuss around this.

I read:
"The value of mysql_insert_id() is affected only by statements issued within the current client connection. It is not affected by statements issued by other clients."


I can't really see what's inaccurate about that.

"In the case of a multiple-row INSERT statement, mysql_insert_id() returns the first automatically generated AUTO_INCREMENT value; if no such value is generated, it returns the last last explicit value inserted into the AUTO_INCREMENT column."

I must be missing something here but why would you insert multiple rows and then only handle the last one with some favoured behaviour? You could just as well insert them one at a time and then handle each row separately with the latest id.

I can't see what's wrong with that.

However I can see what's wrong with simply using max(my_table.id_column) because of the concurrent access issues this would imply.
2008-04-08 10:27:48
"Why on earth are you all arguing about the best way to get the next auto_increment value? The whole point is that it increments automatically. The name should be a give away."

Because you need the same last id value in another table. So you need a way to identify it, in order to use it in the next query. This way its easier than using a new query to get the id like ORDER BY id DESC LIMIT 1
2008-05-02 06:22:03
Other methods seem to have problems with missing records in auto increment sometimes you will have records 1 2 5 6 most functions would return the value of 5 for next auto increment when indeed it would be 7. This is the only way I found to make this work so I can use my customer number and the record number to provide a truly unique customer number that is also useful. 

$next_increment = 0;
$qShowStatus = "SHOW TABLE STATUS LIKE 'your_table'";
$qShowStatusResult = mysql_query($qShowStatus) or die(mysql_error());
$row = mysql_fetch_assoc($qShowStatusResult);
$next_increment = $row['Auto_increment'];
echo $next_increment;

then you can do something like this 
echo  $next_increment ."-". rand();

My first post: I hope this is useful to someone
2008-06-19 16:13:47
Be careful, because this operates on the last performed query, it includes UPDATEs and SELECTs as 'queries'. For example, this is what I set up.

INSERT post into database
UPDATE child forums with insert ID (insert ID is correct)
Insert ID = 0 because of last query
Send the user to their post - but fail because the insert ID is zero. 

So store it in a variable like $insert_id instead of querying it every time.
2008-07-06 16:47:35
If i can't get a good increment.
I use this function

function get_new_id($table){
$select 'select max(`id`) +1as `id` from `'.$table.'`  where `id` != <some big id>';
$query mysql_query($select);
$obj mysql_fetch_object($query);
2008-07-22 05:25:26
Here's an elegant way to INSERT using UPDATE syntax.

function insert_update($table,$fields,$id=NULL)
$sql="INSERT INTO $table (id) VALUES(NULL);UPDATE $table SET $fields WHERE id = LAST_INSERT_ID()";
$sql="UPDATE $table SET $fields WHERE id = $id";


$fields="`username` = 'Ultimater',`userlevel` = 'member'";
//insert a record
//update a record
2009-05-18 08:53:21
Apparently the value returned by mysql_insert_id() may be correct for BIGINT auto_increment keys below the value of INT, but it may wrap to negative when the BIGINT auto_increment passes the largest signed value of INT.  A timebomb for very large tables...

PHP 5.2.10, MySQL 5.0.81, assume the connection and selection...

if (!
$res mysql_query($sql)) die(mysql_error());

$sql "ALTER TABLE noise AUTO_INCREMENT = 2147483646";
if (!
$res mysql_query($sql)) die(mysql_error());

$kount 0;
while (
$kount 3)
$sql "INSERT INTO noise () VALUES ()";
   if (!
$res mysql_query($sql)) die(mysql_error());
$nid mysql_insert_id($dbcx);

// int(2147483646)
// int(2147483647)
// int(-2147483648)
2009-09-18 14:48:37
I thought this would be relevant to all the people using mysqli and looking for the ID after INSERT command :

function insert_join($catid$disc_id) {
// insert a new item into the database
$conn db_connect();
// insert new item
$demande "insert into categories_disc values ('', '".$catid."', '".$disc_id."')";
$resultat $conn->query($demande);
   if (!
$resultat) {
   } else {
$conn->insert_id// function will now return the ID instead of true. 


Then, on the other side, let us call this function as follows :

$cat_id !== false) {
"<p>Category stuff was added to the database as follows : <br>";
"<hr>ID de la category : ".$cat_id."</p><hr>";

2009-09-25 09:16:51
How to get ID of the last updated row in MySQL?

down vote
I've found an answer to this problem :)

by Pomyk

SET @update_id := 0;
UPDATE some_table SET row = 'value', id = (SELECT @update_id := id)
WHERE some_other_row = 'blah' LIMIT 1; 
SELECT @update_id;
EDIT by aefxx

This technique can be further expanded to retrieve the ID of every row affected by an update statement:

SET @uids := null;
UPDATE footable
   SET foo = 'bar'
 WHERE fooid > 5
   AND ( SELECT @uids := CONCAT_WS(',', fooid, @uids) );
SELECT @uids;
This will return a string with all the IDs concatenated by a colon.

(questions: 1388025  form stackoverflow)
2013-07-12 20:09:28
Get current AUTO_INCREMENT value for any table


$table=$db->prepare("SHOW TABLE STATUS FROM -DatabaseName-");
$sonuc = $table->get_result();
while ($satir=$sonuc->fetch_assoc()){
      if (isset($satir["Name"])== "-TableName-"){
echo $ai["-TableName-"];
2015-05-05 16:41:01
You can create a extra collum like with name "key" and generate this "key" using date() function:

$key = date("Y-m-d H:i:s:u");

Then do the INSERT

"INSERT into 'table' ('col_1', 'col_2', 'key') values ('value 1', 'value 2', '$key')";

And now, you can SELECT the last ID

"SELECT ID from 'table' where key = '$key'";
2015-11-25 13:06:16

    Поддержать сайт на родительском проекте КГБ