这个解是非递归的。注意，元素的顺序会有些混合。

function flatten($array) {
    $return = array();
    while(count($array)) {
        $value = array_shift($array);
        if(is_array($value))
            foreach($value as $sub)
                $array[] = $sub;
        else
            $return[] = $value;
    }
    return $return;
}

/**
 * For merging values of a multidimensional array into one 
 *
 * $array = [
 *     0 => [
 *         0 => 'a1',
 *         1 => 'b1',
 *         2 => 'c1',
 *         3 => 'd1'
 *     ],
 *     1 => [
 *         0 => 'a2',
 *         1 => 'b2',
 *         2 => 'c2',
 *     ]
 * ];
 *
 * becomes : 
 *
 * $array = [
 *     0 => 'a1',
 *     1 => 'b1',
 *     2 => 'c1',
 *     3 => 'd1',
 *     4 => 'a2',
 *     5 => 'b2',
 *     6 => 'c2',
 *     
 * ]
 */
array_reduce
(
    $multiArray
    , function ($lastItem, $currentItem) {
        $lastItem = $lastItem ?: array();
        return array_merge($lastItem, array_values($currentItem));
    }
);

依据片段

<?php
//recursive solution

//test array
$nested_array = [[1,2,[3]],4,[5],[[[6,[7=>[7,8,9,10]]]]]];

/*-----------------------------------------
function call and return result to an array
------------------------------------------*/
$index_count = 1;
$flatered_array = array();
$flatered_array = flat_array($nested_array, $index_count);

/*-----------------------------------------
Print Result
-----------------------------------------*/
echo "<pre>";
print_r($flatered_array);


/*-----------------------------------------
function to flaten an array 
-----------------------------------------*/
function flat_array($nested_array, & $index_count, & $flatered_array) {

  foreach($nested_array AS $key=>$val) {
      if(is_array($val)) {
        flat_array($val, $index_count, $flatered_array);
      }
      else {
        $flatered_array[$index_count] = $val;
        ++$index_count;
      }      
  }

return $flatered_array;
}
?>

非递归，非基于引用的实现，这可能比基于递归的实现更容易理解。 可以管理任意深度多维数组，不能平化关联数组。 它的工作原理是每个周期将数组压平一层，直到它完全有效。

function array_flatten(): array{
    $result = func_get_args();
    // check all elements of $list are not arrays
    $_is_flat = function (array $list): bool {
        foreach ($list as $val) {
            if (is_array($val)) {
                return false;
            }
        }
        return true;
    };
    do {
        $tmp = [];
        foreach ($result as $val) {
            if (is_array($val)) {
                if (!array_is_list($val)) {
                    throw new \Exception(sprintf("array_flatten can't handle associative arrays: %s", json_encode($val)));
                }
                $tmp = array_merge($tmp, $val);
            } else {
                $tmp[] = $val;
            }
        }
        $result = $tmp;
    } while (!$_is_flat($result));
    return $result;
}

以下是它处理的情况:

assertEquals(array_flatten(1, 2), $expected = [1, 2], 'array_flatten 1a');
assertEquals(array_flatten([1], [2]), $expected = [1, 2], 'array_flatten 1b');
assertEquals(array_flatten([1], [[2], 3]), $expected = [1, 2, 3], 'array_flatten 1c');
assertEquals(array_flatten(1, [2, 3], [4, 5]), $expected = [1, 2, 3, 4, 5], 'array_flatten 2');
assertEquals(array_flatten(2, 3, [4, 5], [6, 7], 8), $expected = [2, 3, 4, 5, 6, 7, 8], 'array_flatten 3');
assertEquals(array_flatten([2, 3, [4, 5], [6, 7], 8]), $expected = [2, 3, 4, 5, 6, 7, 8], 'array_flatten 4');
assertEquals(array_flatten([2, [3, [4, [5]], [6, [7]], 8]]), $expected = [2, 3, 4, 5, 6, 7, 8], 'array_flatten complex');

二维数组的解

请试试这个:

$array  = your array

$result = call_user_func_array('array_merge', $array);

echo "<pre>";
print_r($result);

编辑:8月21日-13日

下面是适用于多维数组的解决方案:

function array_flatten($array) {
    $return = array();
    foreach ($array as $key => $value) {
        if (is_array($value)){
            $return = array_merge($return, array_flatten($value));
        } else {
            $return[$key] = $value;
        }
    }

    return $return;
}

$array  = Your array

$result = array_flatten($array);

echo "<pre>";
print_r($result);

裁判:http://php.net/manual/en/function.call-user-func-array.php

我只是想指出这是一个折叠，所以可以使用array_reduce

array_reduce($my_array, 'array_merge', array());

编辑:注意，这可以组成平任意数量的水平。我们可以通过以下几种方式做到这一点:

// Reduces one level
$concat   = function($x) { return array_reduce($x, 'array_merge', array()); };

// We can compose $concat with itself $n times, then apply it to $x
// This can overflow the stack for large $n
$compose  = function($f, $g) {
    return function($x) use ($f, $g) { return $f($g($x)); };
};
$identity = function($x) { return $x; };
$flattenA = function($n) use ($compose, $identity, $concat) {
    return  function($x) use ($compose, $identity, $concat, $n) {
        return ($n === 0)? $x
                         : call_user_func(array_reduce(array_fill(0, $n, $concat),
                                                       $compose,
                                                       $identity),
                                          $x);
    };
};

// We can iteratively apply $concat to $x, $n times
$uncurriedFlip     = function($f) {
    return  function($a, $b) use ($f) {
        return $f($b, $a);
    };
};
$iterate  = function($f) use ($uncurriedFlip) {
    return  function($n) use ($uncurriedFlip, $f) {
    return  function($x) use ($uncurriedFlip, $f, $n) {
        return ($n === 0)? $x
                         : array_reduce(array_fill(0, $n, $f),
                                        $uncurriedFlip('call_user_func'),
                                        $x);
    }; };
};
$flattenB = $iterate($concat);

// Example usage:
$apply    = function($f, $x) {
    return $f($x);
};
$curriedFlip = function($f) {
    return  function($a) use ($f) {
    return  function($b) use ($f, $a) {
        return $f($b, $a);
    }; };
};

var_dump(
    array_map(
        call_user_func($curriedFlip($apply),
                       array(array(array('A', 'B', 'C'),
                                   array('D')),
                             array(array(),
                                   array('E')))),
        array($flattenA(2), $flattenB(2))));

当然，我们也可以使用循环，但这个问题要求的是一个组合子函数，类似于array_map或array_values。