I'm probably [super]overthinking this. I'm trying to analyze an array with values like [1,9], [4,6] [5,5], [6,4], [9,1] and duplicate digits (I'm having a super brain fart and can't even remember the term for numbers like this) remove (the last two) so that only [1,9], [4,6] [5,5] are printed.
I was thinking that turning this array into a string and using preg_match, but I'm pretty sure this wouldn't work even if I had the correct regex.
If you have an array of pairs like this:
$x = array(
array(1,9),
array(4,6),
array(5,5),
array(6,4),
array(9,1)
);
Here is one way to get the unique pairs:
foreach ($x as $pair) {
sort($pair);
$unique_pairs[implode(',', $pair)] = $pair;
}
This uses string representations of each sorted pair as keys in a new array, so the result will have distinct values by definition.
As far as the printing them out part of your question, once you have the unique values you can loop over them and print them out in whichever format you like, for example:
foreach ($unique_pairs as $pair) { vprintf("[%d,%d]<br>", $pair); }
It looks like elements are distributed symmetrically.
We can cut the array in two halves and get only the first half with array_slice():
$array = array(
array(1,9),
array(4,6),
array(5,5),
array(6,4),
array(9,1),
);
print_r(array_slice($array, 0, ceil(count($array) / 2)));
Result:
Array(
[0] => Array(
[0] => 1
[1] => 9
)
[1] => Array(
[0] => 4
[1] => 6
)
[2] => Array(
[0] => 5
[1] => 5
)
)
Demo at Codepad.
ceil() is used to round the number up to the next highest integer if there is an even number of items in the array. Example: if there is 3 items in the array, 5 / 2 will return 2.5, we want 3 items so we use ceil(2.5) which gives 3.
Example with 3 items:
$array = array(
array(1,9),
array(5,5),
array(9,1),
);
print_r(array_slice($array, 0, ceil(count($array) / 2)));
Result:
Array(
[0] => Array(
[0] => 1
[1] => 9
)
[1] => Array(
[0] => 5
[1] => 5
)
)
Example with 4 items:
$array = array(
array(1,9),
array(7,7),
array(7,7),
array(9,1),
);
print_r(array_slice($array, 0, ceil(count($array) / 2)));
Result:
Array(
[0] => Array(
[0] => 1
[1] => 9
)
[1] => Array(
[0] => 7
[1] => 7
)
)
If I'm correct in understanding what you are trying to do, you want to remove the final 2 elements from the array?
There is a function in PHP called array_pop that removes the final element from the array.
$array = array_pop($array);
So if you run this twice, you will remove the final 2 elements from the array.
This is how I'd do it (and I hope I am not overthinking this :))
$stringArray = array();
$stringArray[] = '1,9';
$stringArray[] = '4,6';
$stringArray[] = '5,5';
$stringArray[] = '6,4';
$stringArray[] = '9,1';
foreach($stringArray as &$numString) {
$numString = explode(',', $numString);
usort($numString, function($a, $b) {return $a - $b;});
$numString = implode(',', $numString);
}
$a = array_unique($a);
print_r($a);
You basically explode every element into a subarray, sort it and then implode it back. After calling the array_unique, you're left with unique values in the array.
The output would be
Array
(
[0] => 1,9
[1] => 4,6
[2] => 5,5
)
The result you suggest treats [a,b] as equivalent to [b,a] which makes the problem a lot more complex. The code below gives the result you asked for, but without really understanding what the problem is that you are trying to fix and whether [1,9] is equivalent to [9,1] in the solution:
$a=array(array(1,9),array(4,6),...
$dup=array();
for ($i=0; $i<count($a) -1; $i++) {
for ($j=$i+1; $j<count($a); $j++) {
if (($a[$i][0]==$a[$j[0] && $a[$i][1]==$a[$j[1])
|| ($a[$i][0]==$a[$j[1] && $a[$i][1]==$a[$j[0])) {
$dup[]=$j;
}
}
}
foreach ($dup as $i) {
unset($a[$i]);
}
So I'm actually going to assume your question to have a different meaning than everyone else did. I believe what you're asking is:
How do you filter out array items where a reverse of the item has already been used?
<?php
// The example set you gave
$numberSets = [[1, 9], [4, 6], [5, 5], [6, 4], [9, 1]];
// Initialize an empty array to keep track of what we've seen
$keys = [];
// We use array filter to get rid of items we don't want
// (Notice that we use & on $keys, so that we can update the variable in the global scope)
$numberSets = array_filter($numberSets, function($set) use(&$keys) {
// Reverse the array
$set = array_reverse($set);
// Create a string of the items
$key = implode('', $set);
// Get the reverse of the numbers
$reversedKey = strrev($key);
// If the palindrome of our string was used, return false to filter
if (isset($keys[$reversedKey])) {
return false;
}
// Set the key so it's not used again
// Since $keys is being passed by reference it is updated in global scope
$keys[$key] = true;
// Return true to NOT filter this item, since it or it's reverse were not matched
return true;
});
var_dump($numberSets);
Related
I'm trying to convert mib-style strings into PHP array indices. The trick is that I have to do this for a variable number of strings. As an example:
$strings = ['1.1.1' => 1, '1.1.2' => 2, '1.2.1' => 1];
# Given the above, generate the below:
$array = [ 1 => [ 1 => [1 => 1, 2 => 2] ], 2 => [1 => 1] ] ] ] ]
I can't think of a way to do it that isn't just a brute-force, inefficient method. Any helpful function/approach/advice is welcome.
You could take a recursive approach since the problem/result you provide seems to be of a recursive nature. (You can achieve the same result with a loop, applying the same logic as the recursive function ofcourse)
So under the assumption that there are no conflicting string inputs/edge cases what so ever, the following could be one approach:
Loop over all the strings and their values, break it up and create a nested structure by passing the result array by its reference.
function createNested($pieces, $currentIndex, &$previous, $value)
{
$index = $pieces[$currentIndex];
// Our base case: when we reached the final/deepest level of nesting.
// Hence when the we reached the final index.
if ($currentIndex == count($pieces) - 1) {
// Can now safely assign the value to index.
$previous[$index] = $value;
} else {
// Have to make sure we do not override the key/index.
if (!key_exists($index, $previous)) {
$previous[$index] = [];
}
// If the key already existed we can just make a new recursive call (note one level deeper as we pass the array that $previous[$index] points to.
createNested($pieces, $currentIndex + 1, $previous[$index], $value);
}
}
$strings = ['1.1.1' => 1, '1.1.2' => 2, '1.2.1' => 1];
$result = [];
foreach ($strings as $string => $value) {
// Break up the string by .
createNested(explode('.', $string), 0, $result, $value);
}
echo '<pre>';
print_r($result);
echo '</pre>';
Will output:
Array
(
[1] => Array
(
[1] => Array
(
[1] => 1
[2] => 2
)
[2] => Array
(
[1] => 1
)
)
)
I have a multi-dimensional array and from where i want to choose 11 different words. Each word from different array index.
Here is the array link: My multi-dimensional array
array (
'w' =>
array (
0 => 'walls',
1 => 'well',
2 => 'why',
),
'e' =>
array (
0 => 'end',
),
'a' =>
array (
0 => 'advantage',
1 => 'afford',
2 => 'affronting',
3 => 'again',
4 => 'agreeable',
5 => 'ask',
6 => 'at',
),
'c' =>
array (
0 => 'children',
1 => 'civil',
2 => 'continual',
)
);
My Desire Output:
From w => well
From e => end
From a => again
and so on.
Output like: array(well, end, again, ...) as array.
Use the following code:
$f = array_keys($result); // grouping the indices, namely, the characters
$a = "";
for($c=0;$c<count($f);$c++){
$a .= $f[$c];
} // grouping the indices stored in array $f to a string, $a
$words = array();
for($c=0;$c<11;$c++){
$random = $a[rand(0,strlen($a)-1)];
$k = $result[$random];
// $k stores the array of the character index, stored in $result
$random2 = rand(0,count($k)-1);
$words[$c] = $k[$random2];
// choose a word from a given character array
$a = preg_replace("/".$random."/","",$a);
// remove the character from $a to prevent picking words which start with the same character
}
print_r($words);
I've tested and it was proved working
https://3v4l.org/qi1VP
You can achieve this usin array_rand() function :
PHP
$words = [];
$limit = 3; //Replace this with your limit, 11
$count = 0;
shuffle($array);
foreach($array as $key => $value) {
$words[] = $value[array_rand($value)];
$count++;
if ($limit == $count) {
break;
}
}
EvalIn
Check Online, and let me know.
using shuffle and array_slice you can get what you want.
A shuffle function makes your array random, and array slice slice 11 sub array from it.
Array slice takes 3 argument, first one is the array, second one is the offset from where you want to start and last one how much you need to cut.
$words = array();
shuffle($result);
$res = array_slice($result, 0, 11);
foreach($res as $key => $value){
shuffle($value);
$words[] = $value[0];
}
print_r($words);
I am using the Flot jQuery plugin to create a graph on how many visitors there have been per platform. I would like to create a 4th line with total visitors, calculated by previously retrieved data.
I need to combine several multi-dimensional Indexed arrays, but not simply merging them recursively. I.E:
$arr1 = [[2016/05/04,2],[2016/05/03,4],[2016/05/02,6]];
$arr2 = [[2016/05/04,1],[2016/05/03,3],[2016/05/02,2]];
$arr3 = [[2016/05/04,6],[2016/05/03,7],[2016/05/02,8]];
The output should be:
$arrTotal = [[2016/05/04,9],[2016/05/03,14],[2016/05/02,16]];
How do I accomplish this in a (fairly) simple way?
First of all, you cannot declare your dates the way you did:
$arr1 = [[2016/05/04,2],[2016/05/03,4],[2016/05/02,6]];
Because it's going to take 2016, divide it by 5 then divide it by 4. You need to put them into quotes.
$arr1 = [['2016/05/04',2],['2016/05/03',4],['2016/05/02',6]];
But to create an associative array, you should do it this way:
$arr1 = array('2016/05/04' => 2, '2016/05/03' => 4, '2016/05/02' => 6);
$arr2 = array('2016/05/04' => 1, '2016/05/03' => 3, '2016/05/02' => 2);
$arr3 = array('2016/05/04' => 6, '2016/05/03' => 7, '2016/05/02' => 8);
Now all you want to do, is loop through each array and sum them up.
$merge = array();
function mergeArray(Array &$merge, Array $array){
// Loop through each key and value
foreach($array as $key => $value)
// Make sure the value is numeric
if(is_numeric($value)){
if(!isset($merge[$key]))
$merge[$key] = $value;
else
$merge[$key] += $value;
}
}
mergeArray($merge, $arr1);
mergeArray($merge, $arr2);
mergeArray($merge, $arr3);
And now if you dump the $merge:
array(3) {
["2016/05/04"]=>
int(9)
["2016/05/03"]=>
int(14)
["2016/05/02"]=>
int(16)
}
Build a method that will sum the values by respecting the keys of existing values.
$arr1 = array('2016/05/04'=>2,'2016/05/03'=>4,'2016/05/02'=>6);
$arr2 = array('2016/05/04'=>1,'2016/05/03'=>3,'2016/05/02'=>2);
$arr3 = array('2016/05/04'=>2,'2016/05/03'=>7,'2016/05/02'=>8);
function array_sum(&$new_arr,$arr) {
foreach ($arr as $date_key => $num_value) {
// initialize date in new array with 0, if not done previously
if (! isset($new_arr[$date_key])) { $new_arr[$date_key] = 0; }
// add number for indexed element of array
$new_arr[$date_key] += $num_value;
}
}
$new_arr = array();
array_sum($new_array,$arr1);
array_sum($new_array,$arr2);
array_sum($new_array,$arr3);
You are trying to sum up every second value from each nested array relatively to their position in the parent array.There's a short and simple solution using array_map, array_sum and array_column functions:
$groupped = array_map(null, $arr1,$arr2,$arr3);
$result = array_map(function($v){
return [$v[0][0], array_sum(array_column($v, 1))];
}, $groupped);
print_r($result);
The output:
Array
(
[0] => Array
(
[0] => 2016/05/04
[1] => 9
)
[1] => Array
(
[0] => 2016/05/03
[1] => 14
)
[2] => Array
(
[0] => 2016/05/02
[1] => 16
)
)
I want to filter a array by a number and update its status in the first array.
I have two array $arr1,$arr2
$arr1 = array(
0=>array('number'=>100,name=>'john'),
1=>array('number'=>200,name=>'johnny')
);
$arr2= array(
0=>array('number'=>300,name=>'r'),
1=>array('number'=>100,name=>'b'),
2=>array('number'=>200,name=>'c')
);
Final output should be an array like this
$arr1 = array(
0=>array('number'=>100,name=>'b'),
1=>array('number'=>200,name=>'c')
);
Any ideas to start off please ?
For specialized array modifications like this, the method of choice is array walk. It allows you to apply a custom function to each element in a given array.
Now, because of your data format, you will have to do a loop. Wrikken is asking if you can retrieve or transform the data to provide faster access. The algorithm below is O(n^2): it will require as many cycles as there are elements in the first array times the number of elements in the second array, or exactly count($arr1) * count($arr2).
function updateNameFromArray($element, $key, $arr2) {
foreach($arr2 as $value) {
if($value['number'] == $element['number']) {
$element['name'] == $value['name'];
break;
}
}
}
array_walk($arr1, "updateNameFromArray", $arr2);
Now, what Wrikken is suggesting is that if your arrays can be changed to be keyed on the 'number' property instead, then the search/replace operation is much easier. So if this were your data instead:
$arr1 = array(
100=>array('number'=>100,name=>'john'),
200=>array('number'=>200,name=>'johnny')
);
// notice the keys are 100 and 200 instead of 0,1
$arr2= array(
300=>array('number'=>300,name=>'r'),
100=>array('number'=>100,name=>'b'),
200=>array('number'=>200,name=>'c')
);
// notice the keys are 300, 100 and 200 instead of 0,1, 2
Then you could do this in O(n) time, with only looping over the first array.
foreach($arr1 as $key => $value) {
if(isset($arr2[$key])) {
$value['number'] = $arr2[$key]['number'];
}
}
Try this. It's not that clean but i think it would work.
<?php
$arr1 = array(0=>array('number'=>100,'name'=>'john'),1=>array('number'=>200,'name'=>'johnny'));
$arr2= array(0=>array('number'=>300,'name'=>'r'),1=>array('number'=>100,'name'=>'b'),2=>array('number'=>200,'name'=>'c'));
foreach( $arr1 as $key=>$data1 )
{
foreach( $arr2 as $key2=>$data2 )
{
if( $data1['number'] == $data2['number'] )
{
$arr1[$key]['name'] = $arr2[$key2]['name'];
}
}
}
print_r( $arr1 );
?>
the output would be :
Array
(
[0] => Array
(
[number] => 100
[name] => b
)
[1] => Array
(
[number] => 200
[name] => c
)
)
There isn't really a simple way for this to be accomplished with generic PHP functions, so, You might need to create mapping arrays.
The way I would approach this, is creating a loop that goes through the first array, and maps the number value as a key to the index of it's place in $arr1 giving you:
$tmp1 = array();
foreach ($arr1 as $key => $number_name) {
$tmp1[$number_name['number']] = $key;
}
This should give you an array that looks like
$tmp1 [
100 => 0,
200 => 1
];
Then I would loop through the second array, get the number value, if that existed as a key in $tmp1, get the associated value (being the key for $arr1), and use that to update the name in $arr1.
// Loop through $arr2
foreach ($arr2 as $number_name) {
// Get the number value
$number = $number_name['number'];
// Find the $arr1 index
if (isset($tmp1[$number])) {
$arr1_key = $tmp1[$number];
// Set the $arr1 name value
$arr1[$arr1_key]['name'] = $number_name['name'];
}
}
<?php
//Set the arrays
$arr1 = array(
array('number'=>100,'name'=>'john'),
array('number'=>200,'name'=>'johnny')
);
$arr2= array(
array('number'=>300,'name'=>'r'),
array('number'=>100,'name'=>'b'),
array('number'=>200,'name'=>'c')
);
// use a nested for loop to iterate and compare both arrays
for ($i=0;$i<count($arr1);$i++):
for ($j=0;$j<count($arr2);$j++):
if ($arr2[$j]['number']==$arr1[$i]['number'])
$arr1[$i]['name']=$arr2[$j]['name'];
endfor;
endfor;
print_r($arr1);
OUTPUT:
Array (
[0] => Array ( [number] => 100 [name] => b )
[1] => Array ( [number] => 200 [name] => c )
)
That being said, you should probably reconsider the very way your data is structured. Do you really need a multi-dimensional array or can you use a simple associative array, like so:
// set the arrays
$arr1 = array(
'john'=>100,
'johnny'=>200
);
$arr2 = array(
'r'=>300,
'b'=>100,
'c'=>200
);
// find values in arr2 common to both arrays
$arr3 = array_intersect($arr2, $arr1);
// change the key of arr1 to match the corresponding key in arr2
foreach ($arr3 as $key=>$value) {
$old_key = array_search($value, $arr1);
$arr1[$key]=$arr1[$old_key];
unset($arr1[$old_key]);
}
print_r($arr1);
OUTPUT:
Array (
[b] => 100
[c] => 200
)
I have an array which may contain numeric or associative keys, or both:
$x = array('a', 'b', 'c', 'foo' => 'bar', 'd', 'e');
print_r($x);
/*(
[0] => a
[1] => b
[2] => c
[foo] => bar
[3] => d
[4] => e
)*/
I want to be able to remove an item from the array, renumbering the non-associative keys to keep them sequential:
$x = remove($x, "c");
print_r($x);
/* desired output:
(
[0] => a
[1] => b
[foo] => bar
[2] => d
[3] => e
)*/
Finding the right element to remove is no issue, it's the keys that are the problem. unset doesn't renumber the keys, and array_splice works on an offset, rather than a key (ie: take $x from the first example, array_splice($x, 3, 1) would remove the "bar" element rather than the "d" element).
This should re-index the array while preserving string keys:
$x = array_merge($x);
You can fixet with next ELEGANT solution:
For example:
<?php
$array = array (
1 => 'A',
2 => 'B',
3 => 'C'
);
unset($array[2]);
/* $array is now:
Array (
1 => 'A',
3 => 'C'
);
As you can see, the index '2' is missing from the array.
*/
// SOLUTION:
$array = array_values($array);
/* $array is now:
Array (
0 => 'A',
1 => 'C'
);
As you can see, the index begins from zero.
*/
?>
I've come up with this - though I'm not sure if it's the best:
// given: $arr is the array
// $item is the item to remove
$key = array_search($item, $arr); // the key we need to remove
$arrKeys = array_keys($arr);
$keyPos = array_search($key, $arrKeys); // the offset of the item in the array
unset($arr[$key]);
array_splice($arrKeys, $keyPos, 1);
for ($i = $keyPos; $i < count($arrKeys); ++$i) {
if (is_int($arrKeys[$i])) --$arrKeys[$i]; // shift numeric keys back one
}
$arr = array_combine($arrKeys, $arr); // recombine the keys and values.
There's a few things I've left out, just for the sake of brevity. For example, you'd check if the array is associative, and also if the key you're removing is a string or not before using the above code.
Try array_diff() it may not order the new array correctly though
if not the following should work
You will need to iterate over it in the remove function.
function remove($x,$r){
$c = 0;
$a = array();
foreach ($x as $k=>$v){
if ($v != $r) {
if (is_int($k)) {
$a[$c] = $v;
$c++;
}
else {
$a[$k] = $v;
}
}
}
return $a;
}
DC
I don't think there is an elegant solution to this problem, you probably need to loop to the array and reorder the keys by yourself.