array_diff() based on one column between two arrays with associative rows [duplicate]

Question

I have two multidimensional arrays which are indexed arrays of associative rows.

$array1 = array(
    array('name' => 'foo', 'id' => 12),
    array('name' => 'bar', 'id' => 34),
    array('name' => 'boo', 'id' => 56),
);

$array2 = array(
    array('name' => 'bar', 'id' => 34),
    array('name' => 'boo', 'id' => 56),
    array('name' => 'bar', 'id' => 78),
);

It is possible that rows might have different id values but the same name value -- such as bar in my sample input data.

I need to compare the arrays based on id values only.

Expected Output: (because ids 34 and 56 are found in $array2)

array(
    array('name' => 'foo', 'id' => 12)
)

I tried $one_not_two = array_diff($array1['id'], $array2['id']); which does not return anything.

I also tried $one_not_two = array_diff($array1['id'], $array2['id']);
which returned an error "argument is not an array."

Originally, I got around it by extracting the ids into a one-dimensional array, then just comparing those. Now a new feature in our application requires me to compare the rows while maintaining the multidimensional structure. Any advice?

Our servers are currently running php 5.3 if that makes any difference.

Community · Accepted Answer · 2017-05-23 12:17:26Z

4

Because the arrays are multidimensional you have to extract the ids like this:

$ids1 = array();
foreach($array1 as $elem1)
    $ids1[] = $elem1['id'];

$ids2 = array();
foreach($array2 as $elem2)
    $ids2[] = $elem2['id'];

$one_not_two = array_diff($ids1,$ids2);

For your specific question, check out array_diff() with multidimensional arrays

edited May 23, 2017 at 12:17

CommunityBot

11 silver badge

answered Nov 13, 2013 at 21:18

Joren

3,37327 silver badges49 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

Isaac Over a year ago

Thanks but I mentioned in my question that I tried something similar already. That would be an easy solution but something in the rest of the script requires $one_not_two to contain the mapping name => id.

Joren Over a year ago

Then you should look at this question: stackoverflow.com/questions/11821680/…

Isaac Over a year ago

Added your comment to the answer

Jonathan Kuhn · Accepted Answer · 2013-11-13 21:30:50Z

1

You could use array_udiff to create a custom diff function:

$diff = array_udiff($array1,
                    $array2,
                    function ($a, $b){
                        if($a['id'] == $b['id']){
                            return 0;
                        } else {
                            return ($a['id'] < $b['id'] ? -1 : 1);
                        }
                    }
                );

http://codepad.viper-7.com/2u5EWg

answered Nov 13, 2013 at 21:30

Jonathan Kuhn

15.3k3 gold badges34 silver badges43 bronze badges

Comments

Albert Kozłowski · Accepted Answer · 2013-11-13 21:34:00Z

0

If id is unique you can do that:

$one_not_two = array();

foreach ($array1 as $val) {
    $one_not_two[$val['id']] = $val;
}

foreach ($array1 as $val) {
    if (!isset($one_not_two[$val['id']])) {
        $one_not_two[$val['id']] = $val;
}

answered Nov 13, 2013 at 21:34

Albert Kozłowski

4762 silver badges8 bronze badges

Comments

Isaac · Accepted Answer · 2013-11-15 17:09:43Z

0

In the end I solved it by changing Array1 to an associative array of name => id

answered Nov 15, 2013 at 17:09

Isaac

6251 gold badge13 silver badges30 bronze badges

Comments

AFwcxx · Accepted Answer · 2018-07-31 02:31:06Z

0

Having the same problem as you.

Solved it with: $result = array_diff_assoc($array2, $array1);

Reference: PHP: array_diff_assoc

answered Jul 31, 2018 at 2:31

AFwcxx

4811 gold badge4 silver badges15 bronze badges

Collectives™ on Stack Overflow

array_diff() based on one column between two arrays with associative rows [duplicate]

5 Answers 5

3 Comments

Comments

Comments

Comments

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

5 Answers 5

3 Comments

Comments

Comments

Comments

Comments

Linked

Related