CODEWITHSUNDEEP
javascriptarraysalgorithmsorting
Lion789
Lion789

Reputation: 4482

Remove duplicate objects from an array using javascript

I am trying to figure out an efficient way to remove objects that are duplicates from an array and looking for the most efficient answer. I looked around the internet everything seems to be using primitive data... or not scalable for large arrays. This is my current implementation which is can be improved and want to try to avoid labels.

 Test.prototype.unique = function (arr, artist, title, cb) {
        console.log(arr.length);
        var n, y, x, i, r;
        r = [];      
        o: for (i = 0, n = arr.length; i < n; i++) {

          for (x = 0, y = r.length; x < y; x++) {

                if (r[x].artist == arr[i].artist && r[x].title == arr[i].title) {
                    continue o;
                }
            }
            r.push(arr[i]);
        }

        cb(r);
    };

and the array looks something like this:

[{title: sky, artist: jon}, {title: rain, artist: Paul}, ....]

Order does not matter, but if sorting makes it more efficient then I am up for the challenge...

and for people who do not know o is a label and it is just saying jump back to the loop instead of pushing to the new array.

Pure javascript please no libs.

ANSWERS SO FAR:

The Performance Test for the answers below: http://jsperf.com/remove-duplicates-for-loops

Upvotes: 21

Views: 44813

Answers (9)

Abraham
Abraham

Reputation: 9875

For those who love ES6 and short stuff, here it's one solution:

const arr = [
  { title: "sky", artist: "Jon" },
  { title: "rain", artist: "Paul" },
  { title: "sky", artist: "Jon" }
];

Array.from(arr.reduce((a, o) => a.set(o.title, o), new Map()).values());


const arr = [
  { title: "sky", artist: "Jon" },
  { title: "rain", artist: "Paul" },
  { title: "sky", artist: "Jon" },
  { title: "rain", artist: "Jon" },
  { title: "cry", artist: "Jon" }
];

const unique = Array.from(arr.reduce((a, o) => a.set(o.title, o), new Map()).values());

console.log(`New array length: ${unique.length}`)

console.log(unique)

The above example only works for a unique title or id. Basically, it creates a new map for songs with duplicate titles.

Upvotes: 1

Raja Maragani
Raja Maragani

Reputation: 3

function remove_duplicates(objectsArray) {
    var arr = [], collection = []; 
    $.each(objectsArray, function (index, value) {
        if ($.inArray(value.id, arr) == -1) { 
            arr.push(value.id);
            collection.push(value);
        }
    });
    return collection;
}

Upvotes: 0

user3303554
user3303554

Reputation: 2335

Below is Henrique Feijo's answer with ample explanation and an example that you can cut and paste:

Goal: Convert an array of objects that contains duplicate objects (like this one)...

[
    {
        "id": 10620,
        "name": "Things to Print"
    },
    {
        "id": 10620,
        "name": "Things to Print"
    },
    {
        "id": 4334,
        "name": "Interesting"
    }
]

... Into an array of objects without duplicate objects (like this one):

[
    {
        "id": 10620,
        "name": "Things to Print"
    },
    {
        "id": 4334,
        "name": "Interesting"
    }
]

Explanation provided in the comments:

    var allContent = [{
      "id": 10620,
      "name": "Things to Print"
    }, {
      "id": 10620,
      "name": "Things to Print"
    }, {
      "id": 4334,
      "name": "Interesting"
    }]

     //Put Objects Into As Associative Array. Each key consists of a composite value generated by each set of values from the objects in allContent.
    var noDupeObj = {} //Create an associative array. It will not accept duplicate keys.
    for (i = 0, n = allContent.length; i < n; i++) {
      var item = allContent[i]; //Store each object as a variable. This helps with clarity in the next line.
      noDupeObj[item.id + "|" + item.name] = item; //This is the critical step.
      //Here, you create an object within the associative array that has a key composed of the two values from the original object. 
      // Use a delimiter to not have foo+bar handled like fo+obar
      //Since the associative array will not allow duplicate keys, and the keys are determined by the content, then all duplicate content are removed. 
      //The value assigned to each key is the original object which is along for the ride and used to reconstruct the list in the next step.
    }

     //Recontructs the list with only the unique objects left in the doDupeObj associative array
    var i = 0;
    var nonDuplicatedArray = [];
    for (var item in noDupeObj) {
      nonDuplicatedArray[i++] = noDupeObj[item]; //Populate the array with the values from the noDupeObj.
    }

    console.log(nonDuplicatedArray)

Upvotes: 1

Nico
Nico

Reputation: 4426

Here is a solution that works for me.

Helper functions:

// sorts an array of objects according to one field
// call like this: sortObjArray(myArray, "name" );
// it will modify the input array
sortObjArray = function(arr, field) {
    arr.sort(
        function compare(a,b) {
            if (a[field] < b[field])
                return -1;
            if (a[field] > b[field])
                return 1;
            return 0;
        }
    );
}

// call like this: uniqueDishes = removeDuplicatesFromObjArray(dishes, "dishName");
// it will NOT modify the input array
// input array MUST be sorted by the same field (asc or desc doesn't matter)
removeDuplicatesFromObjArray = function(arr, field) {
    var u = [];
    arr.reduce(function (a, b) {
        if (a[field] !== b[field]) u.push(b);
        return b;
    }, []);
    return u;
}

and then simply call:

        sortObjArray(dishes, "name");
        dishes = removeDuplicatesFromObjArray(dishes, "name");

Upvotes: 3

Henrique Feijo
Henrique Feijo

Reputation: 673

I see, the problem there is that the complexity is squared. There is one trick to do it, it's simply by using "Associative arrays".

You can get the array, loop over it, and add the value of the array as a key to the associative array. Since it doesn't allow duplicated keys, you will automatically get rid of the duplicates.

Since you are looking for title and artist when comparing, you can actually try to use something like:

var arrResult = {};
for (i = 0, n = arr.length; i < n; i++) {
    var item = arr[i];
    arrResult[ item.title + " - " + item.artist ] = item;
}

Then you just loop the arrResult again, and recreate the array.

var i = 0;
var nonDuplicatedArray = [];    
for(var item in arrResult) {
    nonDuplicatedArray[i++] = arrResult[item];
}

Updated to include Paul's comment. Thanks!

Upvotes: 30

ShiningSandy
ShiningSandy

Reputation: 11

I use this function. its not doing any sorting, but produces result. Cant say about performance as never measure it.

var unique = function(a){
    var seen = [], result = [];
    for(var len = a.length, i = len-1; i >= 0; i--){
        if(!seen[a[i]]){
            seen[a[i]] = true;
            result.push(a[i]);
        }
    }
    return result;
}

var ar = [1,2,3,1,1,1,1,1,"", "","","", "a", "b"]; console.log(unique(ar));// this will produce [1,2,3,"", "a", "b"] all unique elements.

Upvotes: 1

Mohammed Safeer
Mohammed Safeer

Reputation: 21545

If you are using underscore js, it is easy to remove duplicate object. http://underscorejs.org/#uniq

Upvotes: 0

sabzapple
sabzapple

Reputation: 1

Below code compares object with JSON as String format and removes duplicates and works fine with simple arrays.

    Array.prototype.unique=function(a){
     return function(){
        return this.filter(a)
     }
   }(
   function(a,b,c){
     var tmp=[]; 
     c.forEach(function(el){
        tmp.push(JSON.stringify(el))
    }); 
    return tmp.indexOf(JSON.stringify(a),b+1)<0
  })

Upvotes: 0

Scott Mermelstein
Scott Mermelstein

Reputation: 15397

Basic sort-then-unique implementation, fiddle HERE:

function unique(arr) {
    var comparer = function compareObject(a, b) {
        if (a.title == b.title) {
            if (a.artist < b.artist) {
                return -1;
            } else if (a.artist > b.artist) {
                return 1;
            } else {
                return 0;
            }
        } else {
            if (a.title < b.title) {
                return -1;
            } else {
                return 1;
            }
        }
    }

    arr.sort(comparer);
    console.log("Sorted: " + JSON.stringify(arr));
    for (var i = 0; i < arr.length - 1; ++i) {
        if (comparer(arr[i], arr[i+1]) === 0) {
            arr.splice(i, 1);
            console.log("Splicing: " + JSON.stringify(arr));
        }
    }
    return arr;
}

It may or may not be the most efficient, and should be entirely scalable. I've added some console.logs so you can see it as it works.

EDIT

In the interest of saving on the space the function used, I did that for loop at the end, but it seems likely that didn't properly find only unique results (depsite it passing my simple jsfiddle test). Please try replacing my for loop with the following:

var checker;
var uniqueResults = [];
for (var i = 0; i < arr.length; ++i) {
    if (!checker || comparer(checker, arr[i]) != 0) {
        checker = arr[i];
        uniqueResults.push(checker);
    }
}
return uniqueResults;

Upvotes: 2

Related Questions