Remove duplicate rows from 2d array, but preserve rows with non-empty value in a specific column

Question

I have a 2d array containing rows where a user might be represented more than once. I need to remove duplicate instances of user data, but I don't want to lose any meaningful data in the process. Rows with a non-empty value in the Flying Tour column should be prioritized over a row with an empty value in the same column.

Sample data:

$data = [
    [
        'Access ID' => 12345,
        'Registration Date' => '2018-02-27',
        'First Name' => 'Damian',
        'Last Name' => 'Martin',
        'Flying Tour' => ''
    ],
    [
        'Access ID' => 12345,
        'Registration Date' => '2018-02-27',
        'First Name' => 'Damian',
        'Last Name' => 'Martin',
        'Flying Tour' => 'Yes going'
    ],
    [
        'Access ID' => 789456,
        'Registration Date' => '2018-03-27',
        'First Name' => 'Ricky',
        'Last Name' => 'Smith',
        'Flying Tour' => ''
    ],
    [
        'Access ID' => 789456,
        'Registration Date' => '2018-03-27',
        'First Name' => 'Ricky',
        'Last Name' => 'Smith',
        'Flying Tour' => 'Two way going',
    ],
    [
        'Access ID' => 987654,
        'Registration Date' => '2018-04-27',
        'First Name' => 'Darron',
        'Last Name' => 'Butt',
        'Flying Tour' => ''
    ]
];

My code:

$results = [];
      foreach($data as $input){

      $isDuplicate = false;
      foreach($results as $result){
        if(
            strtolower($input['First Name'])===strtolower($result['First Name']) &&
            strtolower($input['Last Name'])===strtolower($result['Last Name'])      &&
            strtolower($input['Registration ID'])===strtolower($result['Registration ID']) &&
            strtolower(!empty($input['Flying Tour']))
        ){
            //a duplicate was found in results
            $isDuplicate = true;
            break;
        }
      }
      //if no duplicate was found
      if(!$isDuplicate) $results[]=$input;
}

Desired result:

Array
(
    [0] => Array
        (
            [Access ID] => 12345
            [Registration Date] => 2018-02-27
            [First Name] => Damian
            [Last Name] => Martin
            [Flying Tour] => Yes going
        )

    [1] => Array
        (
            [Access ID] => 789456
            [Registration Date] => 2018-03-27
            [First Name] => Ricky
            [Last Name] => Smith
            [Flying Tour] => Two way going
        )

    [2] => Array
        (
            [Access ID] => 987654
            [Registration Date] => 2018-04-27
            [First Name] => Darron
            [Last Name] => Butt
            [Flying Tour] => 
        )

)

Some changes are made please see

_Flying Tour_ for Darron Butt ought to be empty. Ricky Smith instead should have _Flying Tour_ as _Two way going_. — Oluwafemi Sule, Mar 26 '18 at 09:28
What you ask for, and what you show us you want, is contradictory. — StackSlave, Mar 26 '18 at 09:42

Jeffrey · Answer 1 · 2018-03-26T09:39:20.817

The point here is that for the first run, $results is still empty. Thus, the whole logic that can set $isDuplicate = true never runs for the first instance.

This means that for the first record, $isDuplicate = false and thus it will always be added.

Furthermore, strtolower(!empty($input['Flying Tour'])) is a bit weird... and should give an error in my opinion (even if it doesnt...)

You should move the !empty($input['Flying Tour']) outside the for loop, since it does not need the current values in the new array for its check. This will solve the problem.

if(!$isDuplicate && !empty($input['Flying Tour'])) $results[]=$input;

If you want to learn and improve your code, have a look at array_filter it is made for this purpose: filtering array results.

--edit

foreach($results as $id => $result){
    if(
        strtolower($input['First Name'])===strtolower($result['First Name']) &&
        strtolower($input['Last Name'])===strtolower($result['Last Name'])      &&
        strtolower($input['Registration ID'])===strtolower($result['Registration ID']))
    ){
        // Check if the $results value for Flying Tour is empty
        // if yes: update its value $results[ $id ]['Flying Tour'] = $input['Flying Tour'], but still mark as duplicate(!)

        //a duplicate was found in results
        $isDuplicate = true;
        break;
    }
  }

if(!$isDuplicate && !empty($input['Flying Tour'])) $results[]=$input; was good but what happend it return only those value that having Flying Tour — Saurav, Mar 26 '18 at 09:17
ah, now I get it. What you want is to check on the 3 values, if they are the same as the one in $results already (you got that part). Then, you also want to check if the value in `$results` currently has no value for `Flying Tour`. If that is the case, you want to update that record with the new value in `$input`. Though, still mark that record as duplicate, else it will show up twice... — Jeffrey, Mar 26 '18 at 09:36

Alive to die - Anant · Accepted Answer · 2023-01-23T05:15:38.443

Use foreach() along with array keys to check for duplicates:

$results = [];
foreach ($data as $input) {
    if (!isset($results[$input['Access ID'] . '_' . $input['First Name'] . '_' . $input['Last Name']])) {
            $results[$input['Access ID'] . '_' . $input['First Name'] . '_' . $input['Last Name']] = $input;
    } else {
        if ($results[$input['Access ID'] . '_' . $input['First Name'] . '_' . $input['Last Name']]['Flying Tour'] == '') {
            $results[$input['Access ID'] . '_' . $input['First Name'] . '_' . $input['Last Name']] = $input;
        }
    }
}

$results = array_values($results);
//array_multisort( array_column($results, "First Name"), SORT_ASC, $results );
echo "<pre/>";
print_r($results);

Sample Output: https://3v4l.org/2KcSN

Oluwafemi Sule · Answer 3 · 2018-03-26T09:39:16.733

Compose a key from values that should be unique.

If the composed key from each loop array doesn't exists in $results include it in the results.

Also if composed key exists and Flying Tour value is empty replace with new array.

$results = [];

foreach ($data as $input) {
    $key = implode([
        $input['First Name'], 
        $input['Last Name'], 
        $input['Access ID']
    ], '-');

    if (array_key_exists($key, $results) 
        && !empty($results[$key]['Flying Tour'])) continue;

    $results[$key] = $input;
}

var_dump(array_values($results));

Karlo Kokkak · Answer 4 · 2018-03-26T09:32:23.153

0

Uses foreach to iterate over $data. Storing duplicates info via $duplicates variable, and using it to remove duplicate values.

Alternative:

$duplicates = array();
$results = $data;

foreach( $results as $k=>$v ) {
    if( isset( $duplicates[$v['Access ID']] ) ) {
        unset( $results[$k] );
    }
    else $duplicates[$v['Access ID']] = TRUE;
}

print_r( $results );

edited Mar 26 '18 at 09:32

answered Mar 26 '18 at 09:19

Karlo Kokkak

3,674
4
18
33

score 0 · Answer 5 · edited Jan 23 '23 at 01:58

echo '<pre>';
$data = array_reverse($data);
$combined = array_map(function($row){

    $row = array_map('trim', $row);
    unset($row['Flying Tour']);
    //print_r($row);exit;
    return array_reduce($row, function($v1,$v2){
        return $v1 . "|" . $v2;
    });
}, $data);

$uniqueElements  = array_unique($combined);

$uniqueIndexes = array_keys($uniqueElements);
$newData = [];

foreach ($data as $key => $value) {
    if(in_array($key, $uniqueIndexes)){
        $newData[] = $value;
    }
}

print_r($newData);

score 0 · Answer 6 · edited Jan 23 '23 at 01:57

$arrayvalues = array();   // Set the array before foreach 
foreach( $res as $key=>$value ) {
    if( isset( $arrayvalues[$value['Access ID']] ) ) {   // Here checking the Array value and $value are same
        unset( $res[$key] );            // unset the array key value here
    }
    else $arrayvalues[$value['Access ID']] = $value['Access ID'];      // If it is not it will display here and returns true.
}


$i=0;
$testvalues = array();   // Set the array before foreach 
foreach ($res as $key => $value) {
    print_r($value);
    $arr[$i] = $value;
    $testvalues[$i]=$arr[$i];

    // unset($arr[$key]);
    $i++;
}

echo "<pre>";
print_r($testvalues) ;   // Get the respective out put here
echo "</pre>";

score 0 · Answer 7 · answered Jan 23 '23 at 01:47

To group by unique users, you only need to reference their unique access number. Use that value as grouping first-level keys in the result array. As you iterate, if there is no saved row for the access id, then save the row. If the row exists, but the Flying Tour value is empty, then save the current row in its place. When finished iterating, you can remove the associative first-level keys by calling array_values().

Code: (Demo)

$result = [];
foreach ($data as $row) {
    if (empty($result[$row['Access ID']]['Flying Tour'])) {
        $result[$row['Access ID']] = $row;
    }
}
var_export(array_values($result));

Remove duplicate rows from 2d array, but preserve rows with non-empty value in a specific column

7 Answers7