Remove Duplicate Values from PHP Array with array_unique

This article explains how to use the PHP array_unique() function to selectively remove duplicate values from an array. You will learn the basic syntax of the function, how to use different sorting flags to control how duplicates are compared, and how to handle duplicates in both simple and multidimensional arrays.

Basic Usage of array_unique()

The array_unique() function takes an input array and returns a new array with duplicate values removed. By default, it keeps the first encountered key for each unique value and discards the subsequent duplicates.

$fruits = ["apple", "banana", "apple", "cherry", "banana"];
$unique_fruits = array_unique($fruits);

print_r($unique_fruits);

Output:

Array
(
    [0] => apple
    [1] => banana
    [3] => cherry
)

Note: The keys are preserved. If you want to re-index the array keys starting from 0, wrap the result in array_values().

$reindexed = array_values(array_unique($fruits));

Selective Removal Using Flags

The behavior of array_unique() can be customized using its second parameter, $flags. These flags determine how array elements are compared:

• SORT_STRING (Default): Compares items as strings.
• SORT_REGULAR: Compares items normally without changing types (e.g., integer 1 and string "1" will be treated as different).
• SORT_NUMERIC: Compares items numerically.
• SORT_LOCALE_STRING: Compares items as strings, based on the current locale.

Example: String vs. Regular Comparison

If you have an array with mixed data types, the flag you choose changes what is considered a duplicate.

$data = [1, "1", 2, "2"];

// Using SORT_STRING (default) treats 1 and "1" as duplicates
$string_compare = array_unique($data, SORT_STRING); 
// Output: [1, 2]

// Using SORT_REGULAR preserves the data types
$regular_compare = array_unique($data, SORT_REGULAR); 
// Output: [1, "1", 2, "2"]

Removing Duplicates from Multidimensional Arrays

The array_unique() function does not work directly on multidimensional arrays (arrays containing other arrays) because it cannot convert inner arrays to strings by default. To selectively remove duplicates based on a specific key in a multidimensional array, you can combine array_unique() with array_column().

For example, to remove duplicate users based on their email address:

$users = [
    ["id" => 1, "name" => "Alice", "email" => "alice@example.com"],
    ["id" => 2, "name" => "Bob", "email" => "bob@example.com"],
    ["id" => 3, "name" => "Alice Dup", "email" => "alice@example.com"],
];

// Extract the column you want to make unique
$emails = array_column($users, 'email');

// Find the unique values and get their original keys
$unique_keys = array_keys(array_unique($emails));

// Filter the original array using the unique keys
$unique_users = array_intersect_key($users, array_flip($unique_keys));

print_r(array_values($unique_users));

Output:

Array
(
    [0] => Array
        (
            [id] => 1
            [name] => Alice
            [email] => alice@example.com
        )
    [1] => Array
        (
            [id] => 2
            [name] => Bob
            [email] => bob@example.com
        )
)