Here's a working example. It does return the id either of the new record or the existing record. Unfortunately it also appears to change the auto-increment value with every update (even when no new records are inserted) so id's may not be continuous.

From the mysql docs :  In general, you should try to avoid using an ON DUPLICATE KEY UPDATE clause on tables with multiple unique indexes.

#!\C:\Strawberry\perl\bin\perl
use strict;
use warnings;
use DBI;

# CREATE TABLE test2 (
#  id int(11) NOT NULL AUTO_INCREMENT,
#  f1 varchar(45) DEFAULT NULL,
#  PRIMARY KEY (id),
#  UNIQUE KEY f1_UNIQUE (f1)
# ) ENGINE=InnoDB AUTO_INCREMENT=27 DEFAULT CHARSET=utf8;

my $table = 'test2';
my $col   = 'f1'; # unique key
my $entry = $ARGV[0] || 1;

#PERL DBI CONNECT
my $dbh = dbh();

my $sql = sprintf '
  INSERT INTO %s (%s) 
  VALUES (?) 
  ON DUPLICATE KEY 
  UPDATE id = LAST_INSERT_ID(id)',
  $dbh->quote_identifier($table),
  $dbh->quote_identifier($col) ;
  
my $sth = $dbh->prepare($sql);
my $rv = $sth->execute($entry);
print $rv.' '.$dbh->last_insert_id(undef,undef,$table,$col);
    
sub dbh{
  my $database = "test"; my $user = "user"; my $pw = "password";
  my $dsn = "dbi:mysql:$database:localhost:3306"; 
  my $dbh = DBI->connect($dsn, $user, $pw, 
     { RaiseError=>1, AutoCommit=>1 } );
  return $dbh;
}
[download]

What is the use of 'UPDATE id = LAST_INSERT_ID(id)'?

According to the MySQL docs LAST_INSERT_ID(id) will return the value of the argument, so you are actually replacing the existing value in the field 'id' by itself and then you are using that value as the next auto-incremented key value. That is likely to create even more errors as now the database doesn't know anymore where it was in the sequence it was using and will happily re-use an already existing number.

if expr is given as an argument to LAST_INSERT_ID(), the value of the argument is returned by the function and is remembered as the next value to be returned by LAST_INSERT_ID().

CountZero

A program should be light and agile, its subroutines connected like a string of pearls. The spirit and intent of the program should be retained throughout. There should be neither too little or too much, neither needless loops nor useless variables, neither lack of structure nor overwhelming rigidity." - The Tao of Programming, 4.1 - Geoffrey James

My blog: Imperial Deltronics

thanks! actually I didn't realize you can use placeholders with quote identifiers. For some reason I thought they had to be used separately.

So I didn't fully comprehend my own code as some of it was borrowed. If I understand the code correctly, id will be equal to the new auto-increment id, which also means I can call that id from last_insert_id()? Also if it wasn't a duplicate, the last_insert_id() would also have the new key? So basically, with this I can always rely on last_insert_id() to give me the key?

The last_insert_id is only useful with auto-incrementing keys (and even then I find it rarely needed), but in any case auto-incrementing keys should never give a duplicate key error.

that is a good point, and probably a good reason to use unique values as primary keys. I guess I am slightly confused on database design then. I am working with DNA sequences and they can be quite long strings. So I am using a crc32 to check for uniqueness. Would it still be a good idea to use that as a primary key? I merely assumed that was not an ideal key?