[php] How to group subarrays by a column value?

I have the following array

Array
(
    [0] => Array
        (
            [id] => 96
            [shipping_no] => 212755-1
            [part_no] => reterty
            [description] => tyrfyt
            [packaging_type] => PC
        )

    [1] => Array
        (
            [id] => 96
            [shipping_no] => 212755-1
            [part_no] => dftgtryh
            [description] => dfhgfyh
            [packaging_type] => PC
        )

    [2] => Array
        (
            [id] => 97
            [shipping_no] => 212755-2
            [part_no] => ZeoDark
            [description] => s%c%s%c%s
            [packaging_type] => PC
        )

)

How can I group the array by id? Is there any native php functions are available to do this?

While this approach works, I want to do this using a foreach, since with the above I will get duplicate items, which I'm trying to avoid?

On the above example id have 2 items, so its need to be inside of the id

Expanding on @baba's answer, which I like, but creates a more complex three level deep multi-dimensional (array(array(array))):

$group = array();
 foreach ( $array as $value ) {
   $group[$value['id']][] = $value; 
 }

// output only data from id 96
foreach ($group as $key=>$value) { //outer loop
 foreach ($value as $k=>$v){ //inner loop
  if($key==96){ //if outer loop is equal to 96 (could be variable)
   for ($i=0;$i<count($k);$i++){ //iterate over the inner loop
        printf($key.' has a part no. of '.$v['part_no'].' and shipping no. of '.$v['shipping_no'].'<br>');
   }
 }
}
 }

Will output:

96 has a part no. of reterty and shipping number of 212755-1

96 has a part no. of dftgtryh and shipping number of 212755-1

I just threw this together, inspired by .NET LINQ

<?php

// callable type hint may be "closure" type hint instead, depending on php version
function array_group_by(array $arr, callable $key_selector) {
  $result = array();
  foreach ($arr as $i) {
    $key = call_user_func($key_selector, $i);
    $result[$key][] = $i;
  }  
  return $result;
}

 $data = array(
        array(1, "Andy", "PHP"),
        array(1, "Andy", "C#"),
        array(2, "Josh", "C#"),
        array(2, "Josh", "ASP"),
        array(1, "Andy", "SQL"),
        array(3, "Steve", "SQL"),
    );

$grouped = array_group_by($data, function($i){  return $i[0]; });

var_dump($grouped);

?>

And voila you get

array(3) {
  [1]=>
  array(3) {
    [0]=>
    array(3) {
      [0]=>
      int(1)
      [1]=>
      string(4) "Andy"
      [2]=>
      string(3) "PHP"
    }
    [1]=>
    array(3) {
      [0]=>
      int(1)
      [1]=>
      string(4) "Andy"
      [2]=>
      string(2) "C#"
    }
    [2]=>
    array(3) {
      [0]=>
      int(1)
      [1]=>
      string(4) "Andy"
      [2]=>
      string(3) "SQL"
    }
  }
  [2]=>
  array(2) {
    [0]=>
    array(3) {
      [0]=>
      int(2)
      [1]=>
      string(4) "Josh"
      [2]=>
      string(2) "C#"
    }
    [1]=>
    array(3) {
      [0]=>
      int(2)
      [1]=>
      string(4) "Josh"
      [2]=>
      string(3) "ASP"
    }
  }
  [3]=>
  array(1) {
    [0]=>
    array(3) {
      [0]=>
      int(3)
      [1]=>
      string(5) "Steve"
      [2]=>
      string(3) "SQL"
    }
  }
}

In a more functional programming style, you could use array_reduce

$groupedById = array_reduce($data, function (array $accumulator, array $element) {
  $accumulator[$element['id']][] = $element;

  return $accumulator;
}, []);

It's trivial to do with LINQ, which is implemented in PHP in several libraries, including YaLinqo*. It allows performing SQL-like queries on arrays and objects. The groubBy function is designed specifically for grouping, you just need to specify the field you want to group by:

$grouped_array = from($array)->groupBy('$v["id"]')->toArray();

Where '$v["id"]' is a shorthand for function ($v) { return $v["id"]; } which this library supports.

The result will be exactly like in the accepted answer, just with less code.

_{* developed by me}

I think this works better in PHP 5.5+

$IdVar = array_column($data, 'id');

$arr = array();

foreach($old_arr as $key => $item)
{
   $arr[$item['id']][$key] = $item;
}

ksort($arr, SORT_NUMERIC);

You can try the following:

$group = array();

foreach ( $array as $value ) {
    $group[$value['id']][] = $value;
}

var_dump($group);

Output:

array
  96 => 
    array
      0 => 
        array
          'id' => int 96
          'shipping_no' => string '212755-1' (length=8)
          'part_no' => string 'reterty' (length=7)
          'description' => string 'tyrfyt' (length=6)
          'packaging_type' => string 'PC' (length=2)
      1 => 
        array
          'id' => int 96
          'shipping_no' => string '212755-1' (length=8)
          'part_no' => string 'dftgtryh' (length=8)
          'description' => string 'dfhgfyh' (length=7)
          'packaging_type' => string 'PC' (length=2)
  97 => 
    array
      0 => 
        array
          'id' => int 97
          'shipping_no' => string '212755-2' (length=8)
          'part_no' => string 'ZeoDark' (length=7)
          'description' => string 's%c%s%c%s' (length=9)
          'packaging_type' => string 'PC' (length=2)

function groupeByPHP($array,$indexUnique,$assoGroup,$keepInOne){
$retour = array();
$id = $array[0][$indexUnique];
foreach ($keepInOne as $keep){
    $retour[$id][$keep] = $array[0][$keep];
}
foreach ($assoGroup as $cle=>$arrayKey){
    $arrayGrouped = array();
        foreach ($array as $data){
            if($data[$indexUnique] != $id){
                $id = $data[$indexUnique];
                foreach ($keepInOne as $keep){
                    $retour[$id][$keep] = $data[$keep];
                }
            }
            foreach ($arrayKey as $val){
                $arrayGrouped[$val] = $data[$val];
            }
            $retour[$id][$cle][] = $arrayGrouped;
            $retour[$id][$cle] = array_unique($retour[$id][$cle],SORT_REGULAR);
        }
}
return  $retour;
}

Try this function

groupeByPHP($yourArray,'id',array('desc'=>array('part_no','packaging_type')),array('id','shipping_no')) 

This should group an associative array Ejm Group By Country

function getGroupedArray($array, $keyFieldsToGroup) {   
    $newArray = array();

    foreach ($array as $record) 
        $newArray = getRecursiveArray($record, $keyFieldsToGroup, $newArray);

    return $newArray;
}
function getRecursiveArray($itemArray, $keys, $newArray) {
    if (count($keys) > 1) 
        $newArray[$itemArray[$keys[0]]] = getRecursiveArray($itemArray,    array_splice($keys, 1), $newArray[$itemArray[$keys[0]]]);
    else
        $newArray[$itemArray[$keys[0]]][] = $itemArray;

    return $newArray;
}

$countries = array(array('Country'=>'USA', 'State'=>'California'),
                   array('Country'=>'USA', 'State'=>'Alabama'),
                   array('Country'=>'BRA', 'State'=>'Sao Paulo'));

$grouped = getGroupedArray($countries, array('Country'));

This array_group_by function achieves what you are looking for:

$grouped = array_group_by($arr, 'id');

It even supports multi-level groupings:

$grouped = array_group_by($arr, 'id', 'part_no');

function array_group_by($arr, array $keys) {

if (!is_array($arr)) {
    trigger_error('array_group_by(): The first argument should be an array', E_USER_ERROR);
}
if (count($keys)==0) {
    trigger_error('array_group_by(): The Second argument Array can not be empty', E_USER_ERROR);
}

// Load the new array, splitting by the target key
$grouped = [];
foreach ($arr as $value) {
    $grouped[$value[$keys[0]]][] = $value;
}

// Recursively build a nested grouping if more parameters are supplied
// Each grouped array value is grouped according to the next sequential key
if (count($keys) > 1) {
        foreach ($grouped as $key => $value) {
       $parms = array_merge([$value], [array_slice($keys, 1,count($keys))]);
       $grouped[$key] = call_user_func_array('array_group_by', $parms);

    }
}
return $grouped;

}

1. GROUP BY one key

This function works as GROUP BY for array, but with one important limitation: Only one grouping "column" ($identifier) is possible.

function arrayUniqueByIdentifier(array $array, string $identifier)
{
    $ids = array_column($array, $identifier);
    $ids = array_unique($ids);
    $array = array_filter($array,
        function ($key, $value) use($ids) {
            return in_array($value, array_keys($ids));
        }, ARRAY_FILTER_USE_BOTH);
    return $array;
}

2. Detecting the unique rows for a table (twodimensional array)

This function is for filtering "rows". If we say, a twodimensional array is a table, then its each element is a row. So, we can remove the duplicated rows with this function. Two rows (elements of the first dimension) are equal, if all their columns (elements of the second dimension) are equal. To the comparsion of "column" values applies: If a value is of a simple type, the value itself will be use on comparing; otherwise its type (array, object, resource, unknown type) will be used.

The strategy is simple: Make from the original array a shallow array, where the elements are imploded "columns" of the original array; then apply array_unique(...) on it; and as last use the detected IDs for filtering of the original array.

function arrayUniqueByRow(array $table = [], string $implodeSeparator)
{
    $elementStrings = [];
    foreach ($table as $row) {
        // To avoid notices like "Array to string conversion".
        $elementPreparedForImplode = array_map(
            function ($field) {
                $valueType = gettype($field);
                $simpleTypes = ['boolean', 'integer', 'double', 'float', 'string', 'NULL'];
                $field = in_array($valueType, $simpleTypes) ? $field : $valueType;
                return $field;
            }, $row
        );
        $elementStrings[] = implode($implodeSeparator, $elementPreparedForImplode);
    }
    $elementStringsUnique = array_unique($elementStrings);
    $table = array_intersect_key($table, $elementStringsUnique);
    return $table;
}

It's also possible to improve the comparing, detecting the "column" value's class, if its type is object.

The $implodeSeparator should be more or less complex, z.B. spl_object_hash($this).

3. Detecting the rows with unique identifier columns for a table (twodimensional array)

This solution relies on the 2nd one. Now the complete "row" doesn't need to be unique. Two "rows" (elements of the first dimension) are equal now, if all relevant "fields" (elements of the second dimension) of the one "row" are equal to the according "fields" (elements with the same key).

The "relevant" "fields" are the "fields" (elements of the second dimension), which have key, that equals to one of the elements of the passed "identifiers".

function arrayUniqueByMultipleIdentifiers(array $table, array $identifiers, string $implodeSeparator = null)
{
    $arrayForMakingUniqueByRow = $removeArrayColumns($table, $identifiers, true);
    $arrayUniqueByRow = $arrayUniqueByRow($arrayForMakingUniqueByRow, $implodeSeparator);
    $arrayUniqueByMultipleIdentifiers = array_intersect_key($table, $arrayUniqueByRow);
    return $arrayUniqueByMultipleIdentifiers;
}

function removeArrayColumns(array $table, array $columnNames, bool $isWhitelist = false)
{
    foreach ($table as $rowKey => $row) {
        if (is_array($row)) {
            if ($isWhitelist) {
                foreach ($row as $fieldName => $fieldValue) {
                    if (!in_array($fieldName, $columnNames)) {
                        unset($table[$rowKey][$fieldName]);
                    }
                }
            } else {
                foreach ($row as $fieldName => $fieldValue) {
                    if (in_array($fieldName, $columnNames)) {
                        unset($table[$rowKey][$fieldName]);
                    }
                }
            }
        }
    }
    return $table;
}

Recursive function grouping 2-dimensional array by keys from first to last

Input:

$arr = array(
    '0' => array(
        'key0' => 'value0',
        'key1' => 'value1',
        'key2' => 'value02',
    ),
    '2' => array(
        'key0' => 'value0',
        'key1' => 'value1',
        'key2' => 'value12',
    ),
    '3' => array(
        'key0' => 'value0',
        'key1' => 'value3',
        'key2' => 'value22',
    ),
);
$keys = array('key0', 'key1', 'key2');

Output:

$arr = array(
    'value0' => array(
        'value1 => array(
            'value02' => null,
            'value12' => null,
        ),
        'value3' => 'value22',
    ),
);

Code:

function array_group_by_keys(&$arr, $keys) {

    if (count($arr) < 2){
        $arr = array_shift($arr[0]);
        return;
    }

    foreach ($arr as $k => $item) {
        $fvalue = array_shift($item);
        $arr[$fvalue][] = $item;
        unset($arr[$k]);
    }

    array_shift($keys);
    foreach ($arr as &$sub_arr) {
        array_group_by_keys($sub_arr, $keys);
    }
}

$arr = Data Araay;

$fldName = Group By Colum Name;

function array_group_by( $arr, $fldName) {
    $groups = array();
    foreach ($arr as $rec) {
        $groups[$rec[$fldName]] = $rec;
    }
    return $groups;
}

function object_group_by( $obj, $fldName) {
    $groups = array();
    foreach ($obj as $rec) {
        $groups[$rec->$fldName] = $rec;
    }
    return $groups;
}

Check indexed function from Nspl:

use function \nspl\a\indexed;
$grouped = indexed($data, 'id');

Consume and cache the column value that you want to group by, then push the remaining data as a new subarray of the group you have created in the the result.

function array_group(array $data, $by_column)
{
    $result = [];
    foreach ($data as $item) {
        $column = $item[$by_column];
        unset($item[$by_column]);
        $result[$column][] = $item;
    }
    return $result;
}

for($i = 0 ; $i < count($arr)  ; $i++ )
{
    $tmpArr[$arr[$i]['id']] = $arr[$i]['id'];
}
$vmpArr = array_keys($tmpArr);
print_r($vmpArr);