I would like to check if my array has any duplicates and return the duplicated values in an array. I want this to be as efficient as possible.
Example:
You can get the difference of the original array and a copy without duplicates using array_unique and array_diff_assoc:
array_diff_assoc($arr, array_unique($arr))
function returndup($array)
{
$results = array();
$duplicates = array();
foreach ($array as $item) {
if (in_array($item, $results)) {
$duplicates[] = $item;
}
$results[] = $item;
}
return $duplicates;
}
in addition to gumbo's answer:
function returndup($arr)
{
return array_diff_key($arr, array_unique($arr));
}
I did some tests and indeed @user187291's variant is the fastest. But, it turns out that @Gumbo's and @faebser's alternative are almost as fast, @faebser's being just slightly faster than @Gumbo's and sometimes even fastest of all.
Here's the code I used
$array = array(1, "hello", 1, "world", "hello");
$times = 1000000;
$start = microtime(true);
for ($i = 0; $i < $times; $i++) {
$dups = array();
foreach(array_count_values($array) as $val => $c)
if( $c > 1) $dups[] = $val;
}
$end = microtime(true);
echo 'variant 1 (user187291): ' . ($end - $start);
echo '<br><br><br>';
$start = microtime(true);
for ($i = 0; $i < $times; $i++)
$dups = array_unique(array_diff_assoc($array, array_unique($array)));
$end = microtime(true);
echo 'variant 2 (JAL): ' . ($end - $start);
echo '<br><br><br>';
$start = microtime(true);
for ($i = 0; $i < $times; $i++)
$dups = array_diff_assoc($array, array_unique($array));
$end = microtime(true);
echo 'variant 3 (Gumbo): ' . ($end - $start);
echo '<br><br><br>';
$start = microtime(true);
for ($i = 0; $i < $times; $i++)
$dups = array_diff_key($array, array_unique($array));
$end = microtime(true);
echo 'variant 4 (faebser): ' . ($end - $start);
echo '<br><br><br>';