How to filter an array from all elements of another array

Question

I'd like to understand the best way to filter an array from all elements of another one. I tried with the filter function, but it doesn't come to me how to give it the values i want to remove.
Something Like:

var array = [1,2,3,4];
var anotherOne = [2,4];
var filteredArray = array.filter(myCallback);
// filteredArray should now be [1,3]


function myCallBack(){
    return element ! filteredArray; 
    //which clearly can't work since we don't have the reference <,< 
}

in case the filter function is not usefull, how would you implement this ?
Edit: i checked the possible duplicate question, and it could be useful for those who understand javascript easily. The answer checked as good makes things easy.

Answer 1

You can use the this parameter of the filter() function to avoid to store your filter array in a global variable.

var filtered = [1, 2, 3, 4].filter(
    function(e) {
      return this.indexOf(e) < 0;
    },
    [2, 4]
);
console.log(filtered);

Answer 2

I would do as follows;

var arr1 = [1,2,3,4],
    arr2 = [2,4],
    res = arr1.filter(item => !arr2.includes(item));
console.log(res);

Answer 3

var array = [1,2,3,4];
var anotherOne = [2,4];
var filteredArray = array.filter(myCallBack);

function myCallBack(el){
  return anotherOne.indexOf(el) < 0;
}

In the callback, you check if each value of array is in anotherOne

https://jsfiddle.net/0tsyc1sx/

If you are using lodash.js, use _.difference

filteredArray = _.difference(array, anotherOne);

Demo

If you have an array of objects :

var array = [{id :1, name :"test1"},{id :2, name :"test2"},{id :3, name :"test3"},{id :4, name :"test4"}];

var anotherOne = [{id :2, name :"test2"}, {id :4, name :"test4"}];

var filteredArray  = array.filter(function(array_el){
   return anotherOne.filter(function(anotherOne_el){
      return anotherOne_el.id == array_el.id;
   }).length == 0
});

Demo array of objects

Demo diff array of objects with lodash

Answer 4

        /* Here's an example that uses (some) ES6 Javascript semantics to filter an object array by another object array. */

        // x = full dataset
        // y = filter dataset
        let x = [
            {"val": 1, "text": "a"},
            {"val": 2, "text": "b"},
            {"val": 3, "text": "c"},
            {"val": 4, "text": "d"},
            {"val": 5, "text": "e"}
            ],
            y = [
            {"val": 1, "text": "a"},
            {"val": 4, "text": "d"}               
            ];

        // Use map to get a simple array of "val" values. Ex: [1,4]
        let yFilter = y.map(itemY => { return itemY.val; });

        // Use filter and "not" includes to filter the full dataset by the filter dataset's val.
        let filteredX = x.filter(itemX => !yFilter.includes(itemX.val));

        // Print the result.
        console.log(filteredX);

Answer 5

The code below is the simplest way to filter an array with respect to another array. Both arrays can have objects inside them instead of values.

let array1 = [1, 3, 47, 1, 6, 7];
let array2 = [3, 6];
let filteredArray1 = array1.filter(el => array2.includes(el));
console.log(filteredArray1); 

Output: [3, 6]

Answer 6

All the above solutions "work", but are less than optimal for performance and are all approach the problem in the same way which is linearly searching all entries at each point using Array.prototype.indexOf or Array.prototype.includes. A far faster solution (far faster even than a binary search for most cases) would be to sort the arrays and skip ahead as you go along as seen below. However, one downside is that this requires all entries in the array to be numbers or strings. Also however, binary search may in some rare cases be faster than the progressive linear search. These cases arise from the fact that my progressive linear search has a complexity of O(2n₁+n₂) (only O(n₁+n₂) in the faster C/C++ version) (where n₁ is the searched array and n₂ is the filter array), whereas the binary search has a complexity of O(n₁ceil(log₂n₂)) (ceil = round up -- to the ceiling), and, lastly, the indexOf search has a highly variable complexity between O(n₁) and O(n₁n₂), averaging out to O(n₁ceil(n₂÷2)). Thus, indexOf will only be the fastest, on average, in the cases of (n₁,n₂) equaling {1,2}, {1,3}, or {x,1|x∈N}. However, this is still not a perfect representation of modern hardware. IndexOf is natively optimized to the fullest extent imaginable in most modern browsers, making it very subject to the laws of branch prediction. Thus, if we make the same assumption on indexOf as we do with progressive linear and binary search -- that the array is presorted -- then, according to the stats listed in the link, we can expect roughly a 6x speed up for IndexOf, shifting its complexity to between O(n₁÷6) and O(n₁n₂), averaging out to O(n₁ceil(n₂7÷12)). Finally, take note that the below solution will never work with objects because objects in JavaScript cannot be compared by pointers in JavaScript.

function sortAnyArray(a,b) { return a>b ? 1 : (a===b ? 0 : -1); }
function sortIntArray(a,b) { return (a|0) - (b|0) |0; }
function fastFilter(array, handle) {
    var out=[], value=0;
    for (var i=0,  len=array.length|0; i < len; i=i+1|0)
        if (handle(value = array[i])) 
            out.push( value );
    return out;
}

const Math_clz32 = Math.clz32 || (function(log, LN2){
  return function(x) {
    return 31 - log(x >>> 0) / LN2 | 0; // the "| 0" acts like math.floor
  };
})(Math.log, Math.LN2);

/* USAGE:
  filterArrayByAnotherArray(
      [1,3,5],
      [2,3,4]
  ) yields [1, 5], and it can work with strings too
*/
function filterArrayByAnotherArray(searchArray, filterArray) {
    if (
        // NOTE: This does not check the whole array. But, if you know
        //        that there are only strings or numbers (not a mix of
        //        both) in the array, then this is a safe assumption.
        // Always use `==` with `typeof` because browsers can optimize
        //  the `==` into `===` (ONLY IN THIS CIRCUMSTANCE)
        typeof searchArray[0] == "number" &&
        typeof filterArray[0] == "number" &&
        (searchArray[0]|0) === searchArray[0] &&
        (filterArray[0]|0) === filterArray[0]
    ) {filterArray
        // if all entries in both arrays are integers
        searchArray.sort(sortIntArray);
        filterArray.sort(sortIntArray);
    } else {
        searchArray.sort(sortAnyArray);
        filterArray.sort(sortAnyArray);
    }
    var searchArrayLen = searchArray.length, filterArrayLen = filterArray.length;
    var progressiveLinearComplexity = ((searchArrayLen<<1) + filterArrayLen)>>>0
    var binarySearchComplexity= (searchArrayLen * (32-Math_clz32(filterArrayLen-1)))>>>0;
    // After computing the complexity, we can predict which algorithm will be the fastest
    var i = 0;
    if (progressiveLinearComplexity < binarySearchComplexity) {
        // Progressive Linear Search
        return fastFilter(searchArray, function(currentValue){
            while (filterArray[i] < currentValue) i=i+1|0;
            // +undefined = NaN, which is always false for <, avoiding an infinite loop
            return filterArray[i] !== currentValue;
        });
    } else {
        // Binary Search
        return fastFilter(
            searchArray,
            fastestBinarySearch(filterArray)
        );
    }
}

// see https://stackoverflow.com/a/44981570/5601591 for implementation
//  details about this binary search algorithm

function fastestBinarySearch(array){
  var initLen = (array.length|0) - 1 |0;
  
  const compGoto = Math_clz32(initLen) & 31;
  return function(sValue) {
    var len = initLen |0;
    switch (compGoto) {
      case 0:
        if (len & 0x80000000) {
          const nCB = len & 0x80000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 1:
        if (len & 0x40000000) {
          const nCB = len & 0xc0000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 2:
        if (len & 0x20000000) {
          const nCB = len & 0xe0000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 3:
        if (len & 0x10000000) {
          const nCB = len & 0xf0000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 4:
        if (len & 0x8000000) {
          const nCB = len & 0xf8000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 5:
        if (len & 0x4000000) {
          const nCB = len & 0xfc000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 6:
        if (len & 0x2000000) {
          const nCB = len & 0xfe000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 7:
        if (len & 0x1000000) {
          const nCB = len & 0xff000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 8:
        if (len & 0x800000) {
          const nCB = len & 0xff800000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 9:
        if (len & 0x400000) {
          const nCB = len & 0xffc00000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 10:
        if (len & 0x200000) {
          const nCB = len & 0xffe00000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 11:
        if (len & 0x100000) {
          const nCB = len & 0xfff00000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 12:
        if (len & 0x80000) {
          const nCB = len & 0xfff80000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 13:
        if (len & 0x40000) {
          const nCB = len & 0xfffc0000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 14:
        if (len & 0x20000) {
          const nCB = len & 0xfffe0000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 15:
        if (len & 0x10000) {
          const nCB = len & 0xffff0000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 16:
        if (len & 0x8000) {
          const nCB = len & 0xffff8000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 17:
        if (len & 0x4000) {
          const nCB = len & 0xffffc000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 18:
        if (len & 0x2000) {
          const nCB = len & 0xffffe000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 19:
        if (len & 0x1000) {
          const nCB = len & 0xfffff000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 20:
        if (len & 0x800) {
          const nCB = len & 0xfffff800;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 21:
        if (len & 0x400) {
          const nCB = len & 0xfffffc00;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 22:
        if (len & 0x200) {
          const nCB = len & 0xfffffe00;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 23:
        if (len & 0x100) {
          const nCB = len & 0xffffff00;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 24:
        if (len & 0x80) {
          const nCB = len & 0xffffff80;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 25:
        if (len & 0x40) {
          const nCB = len & 0xffffffc0;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 26:
        if (len & 0x20) {
          const nCB = len & 0xffffffe0;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 27:
        if (len & 0x10) {
          const nCB = len & 0xfffffff0;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 28:
        if (len & 0x8) {
          const nCB = len & 0xfffffff8;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 29:
        if (len & 0x4) {
          const nCB = len & 0xfffffffc;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 30:
        if (len & 0x2) {
          const nCB = len & 0xfffffffe;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 31:
        if (len & 0x1) {
          const nCB = len & 0xffffffff;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
    }
    // MODIFICATION: Instead of returning the index, this binary search
    //                instead returns whether something was found or not.
    if (array[len|0] !== sValue) {
       return true; // preserve the value at this index
    } else {
       return false; // eliminate the value at this index
    }
  };
}

^{Please see my other post here for more details on the binary search algorithm used}

If you are squeamish about file size (which I respect), then you can sacrifice a little performance in order to greatly reduce the file size and increase maintainability.

function sortAnyArray(a,b) { return a>b ? 1 : (a===b ? 0 : -1); }
function sortIntArray(a,b) { return (a|0) - (b|0) |0; }
function fastFilter(array, handle) {
    var out=[], value=0;
    for (var i=0,  len=array.length|0; i < len; i=i+1|0)
        if (handle(value = array[i])) 
            out.push( value );
    return out;
}

/* USAGE:
  filterArrayByAnotherArray(
      [1,3,5],
      [2,3,4]
  ) yields [1, 5], and it can work with strings too
*/
function filterArrayByAnotherArray(searchArray, filterArray) {
    if (
        // NOTE: This does not check the whole array. But, if you know
        //        that there are only strings or numbers (not a mix of
        //        both) in the array, then this is a safe assumption.
        typeof searchArray[0] == "number" &&
        typeof filterArray[0] == "number" &&
        (searchArray[0]|0) === searchArray[0] &&
        (filterArray[0]|0) === filterArray[0]
    ) {
        // if all entries in both arrays are integers
        searchArray.sort(sortIntArray);
        filterArray.sort(sortIntArray);
    } else {
        searchArray.sort(sortAnyArray);
        filterArray.sort(sortAnyArray);
    }
    // Progressive Linear Search
    var i = 0;
    return fastFilter(searchArray, function(currentValue){
        while (filterArray[i] < currentValue) i=i+1|0;
        // +undefined = NaN, which is always false for <, avoiding an infinite loop
        return filterArray[i] !== currentValue;
    });
}

To prove the difference in speed, let us examine some JSPerfs. For filtering an array of 16 elements, binary search is roughly 17% faster than indexOf while filterArrayByAnotherArray is roughly 93% faster than indexOf. For filtering an array of 256 elements, binary search is roughly 291% faster than indexOf while filterArrayByAnotherArray is roughly 353% faster than indexOf. For filtering an array of 4096 elements, binary search is roughly 2655% faster than indexOf while filterArrayByAnotherArray is roughly 4627% faster than indexOf.

Reverse-filtering (like an AND gate)

The previous section provided code to take array A and array B, and remove all elements from A that exist in B:

filterArrayByAnotherArray(
    [1,3,5],
    [2,3,4]
);
// yields [1, 5]

This next section will provide code for reverse-filtering, where we remove all elements from A that DO NOT exist in B. This process is functionally equivalent to only retaining the elements common to both A and B, like an AND gate:

reverseFilterArrayByAnotherArray(
    [1,3,5],
    [2,3,4]
);
// yields [3]

Here is the code for reverse filtering:

function sortAnyArray(a,b) { return a>b ? 1 : (a===b ? 0 : -1); }
function sortIntArray(a,b) { return (a|0) - (b|0) |0; }
function fastFilter(array, handle) {
    var out=[], value=0;
    for (var i=0,  len=array.length|0; i < len; i=i+1|0)
        if (handle(value = array[i])) 
            out.push( value );
    return out;
}

const Math_clz32 = Math.clz32 || (function(log, LN2){
  return function(x) {
    return 31 - log(x >>> 0) / LN2 | 0; // the "| 0" acts like math.floor
  };
})(Math.log, Math.LN2);

/* USAGE:
  reverseFilterArrayByAnotherArray(
      [1,3,5],
      [2,3,4]
  ) yields [3], and it can work with strings too
*/
function reverseFilterArrayByAnotherArray(searchArray, filterArray) {
    if (
        // NOTE: This does not check the whole array. But, if you know
        //        that there are only strings or numbers (not a mix of
        //        both) in the array, then this is a safe assumption.
        // Always use `==` with `typeof` because browsers can optimize
        //  the `==` into `===` (ONLY IN THIS CIRCUMSTANCE)
        typeof searchArray[0] == "number" &&
        typeof filterArray[0] == "number" &&
        (searchArray[0]|0) === searchArray[0] &&
        (filterArray[0]|0) === filterArray[0]
    ) {
        // if all entries in both arrays are integers
        searchArray.sort(sortIntArray);
        filterArray.sort(sortIntArray);
    } else {
        searchArray.sort(sortAnyArray);
        filterArray.sort(sortAnyArray);
    }
    var searchArrayLen = searchArray.length, filterArrayLen = filterArray.length;
    var progressiveLinearComplexity = ((searchArrayLen<<1) + filterArrayLen)>>>0
    var binarySearchComplexity= (searchArrayLen * (32-Math_clz32(filterArrayLen-1)))>>>0;
    // After computing the complexity, we can predict which algorithm will be the fastest
    var i = 0;
    if (progressiveLinearComplexity < binarySearchComplexity) {
        // Progressive Linear Search
        return fastFilter(searchArray, function(currentValue){
            while (filterArray[i] < currentValue) i=i+1|0;
            // +undefined = NaN, which is always false for <, avoiding an infinite loop
            // For reverse filterning, I changed !== to ===
            return filterArray[i] === currentValue;
        });
    } else {
        // Binary Search
        return fastFilter(
            searchArray,
            inverseFastestBinarySearch(filterArray)
        );
    }
}

// see https://stackoverflow.com/a/44981570/5601591 for implementation
//  details about this binary search algorithim

function inverseFastestBinarySearch(array){
  var initLen = (array.length|0) - 1 |0;
  
  const compGoto = Math_clz32(initLen) & 31;
  return function(sValue) {
    var len = initLen |0;
    switch (compGoto) {
      case 0:
        if (len & 0x80000000) {
          const nCB = len & 0x80000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 1:
        if (len & 0x40000000) {
          const nCB = len & 0xc0000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 2:
        if (len & 0x20000000) {
          const nCB = len & 0xe0000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 3:
        if (len & 0x10000000) {
          const nCB = len & 0xf0000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 4:
        if (len & 0x8000000) {
          const nCB = len & 0xf8000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 5:
        if (len & 0x4000000) {
          const nCB = len & 0xfc000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 6:
        if (len & 0x2000000) {
          const nCB = len & 0xfe000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 7:
        if (len & 0x1000000) {
          const nCB = len & 0xff000000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 8:
        if (len & 0x800000) {
          const nCB = len & 0xff800000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 9:
        if (len & 0x400000) {
          const nCB = len & 0xffc00000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 10:
        if (len & 0x200000) {
          const nCB = len & 0xffe00000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 11:
        if (len & 0x100000) {
          const nCB = len & 0xfff00000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 12:
        if (len & 0x80000) {
          const nCB = len & 0xfff80000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 13:
        if (len & 0x40000) {
          const nCB = len & 0xfffc0000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 14:
        if (len & 0x20000) {
          const nCB = len & 0xfffe0000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 15:
        if (len & 0x10000) {
          const nCB = len & 0xffff0000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 16:
        if (len & 0x8000) {
          const nCB = len & 0xffff8000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 17:
        if (len & 0x4000) {
          const nCB = len & 0xffffc000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 18:
        if (len & 0x2000) {
          const nCB = len & 0xffffe000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 19:
        if (len & 0x1000) {
          const nCB = len & 0xfffff000;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 20:
        if (len & 0x800) {
          const nCB = len & 0xfffff800;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 21:
        if (len & 0x400) {
          const nCB = len & 0xfffffc00;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 22:
        if (len & 0x200) {
          const nCB = len & 0xfffffe00;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 23:
        if (len & 0x100) {
          const nCB = len & 0xffffff00;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 24:
        if (len & 0x80) {
          const nCB = len & 0xffffff80;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 25:
        if (len & 0x40) {
          const nCB = len & 0xffffffc0;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 26:
        if (len & 0x20) {
          const nCB = len & 0xffffffe0;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 27:
        if (len & 0x10) {
          const nCB = len & 0xfffffff0;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 28:
        if (len & 0x8) {
          const nCB = len & 0xfffffff8;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 29:
        if (len & 0x4) {
          const nCB = len & 0xfffffffc;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 30:
        if (len & 0x2) {
          const nCB = len & 0xfffffffe;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
      case 31:
        if (len & 0x1) {
          const nCB = len & 0xffffffff;
          len ^= (len ^ (nCB-1)) & ((array[nCB] <= sValue |0) - 1 >>>0);
        }
    }
    // MODIFICATION: Instead of returning the index, this binary search
    //                instead returns whether something was found or not.
    // For reverse filterning, I swapped true with false and vice-versa
    if (array[len|0] !== sValue) {
       return false; // preserve the value at this index
    } else {
       return true; // eliminate the value at this index
    }
  };
}

For the slower smaller version of the reverse filtering code, see below.

function sortAnyArray(a,b) { return a>b ? 1 : (a===b ? 0 : -1); }
function sortIntArray(a,b) { return (a|0) - (b|0) |0; }
function fastFilter(array, handle) {
    var out=[], value=0;
    for (var i=0,  len=array.length|0; i < len; i=i+1|0)
        if (handle(value = array[i])) 
            out.push( value );
    return out;
}

/* USAGE:
  reverseFilterArrayByAnotherArray(
      [1,3,5],
      [2,3,4]
  ) yields [3], and it can work with strings too
*/
function reverseFilterArrayByAnotherArray(searchArray, filterArray) {
    if (
        // NOTE: This does not check the whole array. But, if you know
        //        that there are only strings or numbers (not a mix of
        //        both) in the array, then this is a safe assumption.
        typeof searchArray[0] == "number" &&
        typeof filterArray[0] == "number" &&
        (searchArray[0]|0) === searchArray[0] &&
        (filterArray[0]|0) === filterArray[0]
    ) {
        // if all entries in both arrays are integers
        searchArray.sort(sortIntArray);
        filterArray.sort(sortIntArray);
    } else {
        searchArray.sort(sortAnyArray);
        filterArray.sort(sortAnyArray);
    }
    // Progressive Linear Search
    var i = 0;
    return fastFilter(searchArray, function(currentValue){
        while (filterArray[i] < currentValue) i=i+1|0;
        // +undefined = NaN, which is always false for <, avoiding an infinite loop
        // For reverse filter, I changed !== to ===
        return filterArray[i] === currentValue;
    });
}

Answer 7

If you need to compare an array of objects, this works in all cases:

let arr = [{ id: 1, title: "title1" },{ id: 2, title: "title2" }]
let brr = [{ id: 2, title: "title2" },{ id: 3, title: "title3" }]

const res = arr.filter(f => brr.some(item => item.id === f.id));
console.log(res);

Answer 8

There are many answers for your question, but I don't see anyone using lambda expresion:

var array = [1,2,3,4];
var anotherOne = [2,4];
var filteredArray = array.filter(x => anotherOne.indexOf(x) < 0);

Answer 9

The OA can also be implemented in ES6 as follows

ES6:

 const filtered = [1, 2, 3, 4].filter(e => {
    return this.indexOf(e) < 0;
  },[2, 4]);

Answer 10

Below is an example

let firstArray=[1,2,3,4,5];
let secondArray=[2,3];  
let filteredArray = firstArray.filter((a) => secondArray.indexOf(a)<0);
console.log(filteredArray); //above line gives [1,4,5]

Answer 11

with object filter result

[{id:1},{id:2},{id:3},{id:4}].filter(v=>!([{id:2},{id:4}].some(e=>e.id === v.id)))

Answer 12

The best description to filter function is https://developer.mozilla.org/pl/docs/Web/JavaScript/Referencje/Obiekty/Array/filter

You should simply condition function:

function conditionFun(element, index, array) {
   return element >= 10;
}
filtered = [12, 5, 8, 130, 44].filter(conditionFun);

And you can't access the variable value before it is assigned

Answer 13

You can setup the filter function to iterate over the "filter array".

var arr = [1, 2, 3 ,4 ,5, 6, 7];
var filter = [4, 5, 6];

var filtered = arr.filter(
  function(val) {
    for (var i = 0; i < filter.length; i++) {
      if (val == filter[i]) {
        return false;
      }
    }
    return true;
  }
); 

Answer 14

You can use the filter and then for the filter function use a reduction of the filtering array which checks and returns true when it finds a match then invert on return (!). The filter function is called once per element in the array. You are not doing a comparison of any of the elements in the function in your post.

var a1 = [1, 2, 3, 4],
  a2 = [2, 3];

var filtered = a1.filter(function(x) {
  return !a2.reduce(function(y, z) {
    return x == y || x == z || y == true;
  })
});

document.write(filtered);

Answer 15

var arr1= [1,2,3,4];
var arr2=[2,4]

function fil(value){
return value !=arr2[0] &&  value != arr2[1]
}

document.getElementById("p").innerHTML= arr1.filter(fil)

<!DOCTYPE html> 
<html> 
<head> 
</head>
<body>
<p id="p"></p>

Answer 16

function arr(arr1,arr2){
  
  function filt(value){
    return arr2.indexOf(value) === -1;
    }
  
  return arr1.filter(filt)
  }

document.getElementById("p").innerHTML = arr([1,2,3,4],[2,4])

<p id="p"></p>

Answer 17

A more flexible filtering array from another array which contain object properties

function filterFn(array, diffArray, prop, propDiff) {
    diffArray = !propDiff ? diffArray : diffArray.map(d => d[propDiff])
    this.fn = f => diffArray.indexOf(f) === -1
    if (prop) {
         return array.map(r => r[prop]).filter(this.fn)
    } else {
         return array.filter(this.fn)
    }
}

//You can use it like this;

var arr = [];

for (var i = 0; i < 10; i++) {
    var obj = {}
    obj.index = i
    obj.value = Math.pow(2, i)
    arr.push(obj)
}

var arr2 = [1, 2, 3, 4, 5]

var sec = [{t:2}, {t:99}, {t:256}, {t:4096}]

var log = console.log.bind(console)

var filtered = filterFn(arr, sec, 'value', 't')

var filtered2 = filterFn(arr2, sec, null, 't')

log(filtered, filtered2)

Answer 18

You can write a generic filterByIndex() function and make use of type inference in TS to save the hassle with the callback function:

let's say you have your array [1,2,3,4] that you want to filter() with the indices specified in the [2,4] array.

var filtered = [1,2,3,4,].filter(byIndex(element => element, [2,4]))

the byIndex function expects the element function and an array and looks like this:

byIndex = (getter: (e:number) => number, arr: number[]) => (x: number) => {
    var i = getter(x);
    return arr.indexOf(i); 
}

result is then

filtered = [1,3]

Answer 19

The following examples use new Set() to create a filtered array that has only unique elements:

Array with primitive data types: string, number, boolean, null, undefined, symbol:

const a = [1, 2, 3, 4];
const b = [3, 4, 5];
const c = Array.from(new Set(a.concat(b)));

Array with objects as items:

const a = [{id:1}, {id: 2}, {id: 3}, {id: 4}];
const b = [{id: 3}, {id: 4}, {id: 5}];
const stringifyObject = o => JSON.stringify(o);
const parseString = s => JSON.parse(s);
const c = Array.from(new Set(a.concat(b).map(stringifyObject)), parseString);

Answer 20

The solution of Jack Giffin is great but doesn't work for arrays with numbers bigger than 2^32. Below is a refactored, fast version to filter an array based on Jack's solution but it works for 64-bit arrays.

const Math_clz32 = Math.clz32 || ((log, LN2) => x => 31 - log(x >>> 0) / LN2 | 0)(Math.log, Math.LN2);

const filterArrayByAnotherArray = (searchArray, filterArray) => {

    searchArray.sort((a,b) => a > b);
    filterArray.sort((a,b) => a > b);

    let searchArrayLen = searchArray.length, filterArrayLen = filterArray.length;
    let progressiveLinearComplexity = ((searchArrayLen<<1) + filterArrayLen)>>>0
    let binarySearchComplexity = (searchArrayLen * (32-Math_clz32(filterArrayLen-1)))>>>0;

    let i = 0;

    if (progressiveLinearComplexity < binarySearchComplexity) {
      return searchArray.filter(currentValue => {
        while (filterArray[i] < currentValue) i=i+1|0;
        return filterArray[i] !== currentValue;
      });
    }
    else return searchArray.filter(e => binarySearch(filterArray, e) === null);
}

const binarySearch = (sortedArray, elToFind) => {
  let lowIndex = 0;
  let highIndex = sortedArray.length - 1;
  while (lowIndex <= highIndex) {
    let midIndex = Math.floor((lowIndex + highIndex) / 2);
    if (sortedArray[midIndex] == elToFind) return midIndex; 
    else if (sortedArray[midIndex] < elToFind) lowIndex = midIndex + 1;
    else highIndex = midIndex - 1;
  } return null;
}

Answer 21

Here is how you can do it when the items in the arrays are objects.

The idea is to find the array of only the keys in an inner array using the map function

Then check if the array of those keys contain a specific element key in the outer array.

const existsInBothArrays = array1.filter((element1) =>
    array2.map((element2) => element2._searchKey).includes(element1._searchKey),
  );