Related
I'm trying to work out how to replace the values 1,2 and 3
inside this GUI interface with a range of numbers based on simply how many objects there are in an array, so that it dynamically keeps track of it without me having to hardcore it in everytime the array is updated.
What is preventing me from simply using an array with numbers itself is that each of the numbers is an object inside of this GUI.
The line of code is this.
gui.add(data, 'system', {
"1": 0,
"2": 1,
"3": 2
})
Suppose I am using any random array holding any random information like so:
var arr = [];
for (var i = 0; i<51;i++){
arr.push("element");}
There are 51 elements in this array and what I'm going for is that the above array holding these elements is used to output the numbers "1", "2", "3", "4", "5", "6" (divided by 10 while still having the 1 leftover) with the GUI interface not breaking down on me. The reason I want it to be divided by 10 is because I don't want to have 51 elements clickable but rather 6 because that's more manageable.
What should I do or look into? Thanks. The fiddle with my code is this. FIDDLE
EDIT: It's a relatively difficult question, so I'll try to be more clear. I have the numerical values 1,2,3 hardcoded in this object:
gui.add(data, 'system', {
"1": 0,
"2": 1,
"3": 2
}).name('system #').onChange(function(value) {
updateSets(value);
});
But what I'm trying to achieve is so that that I don't need to write in the "1", "2", "3" but that it looks to an array's length (which if it holds 3 elements) it'll write them in in the object for me. But if an array has 51 elements, I don't want to have 51 numbers under the system object, but that it divides it by 10 so that I have 6. Not 5, because I have the 1 leftover.
You can use Array.prototype.reduce() to create an object having equal amount of properties and values to .length of an array.
Create a plain object. Utilize Object.entries(), for..of loop to set properties, values of object up to variable n.
You can use Array.prototype.slice() to get n through x properties of values from initially created object.
var arr = [];
for (var i = 0; i < 51; i++) {
arr.push("element");
}
var obj = arr.reduce(function(obj, element, index) {
obj[index + 1] = index;
return obj
}, {});
function getObjProps(from, to, obj, res) {
let it = void 0;
if (from) {
it = Object.entries(obj).slice(from, to);
} else {
it = Object.entries(obj);
}
for (let [key, value] of it) {
if (+key <= to) {
res[key] = value;
} else {
break;
}
};
return res
}
var curr = getObjProps(null, 6, obj, {});
console.log(curr);
var next = getObjProps(6, 12, obj, {});
console.log(next);
for (let i = 12; i < arr.length; i += 6) {
console.log(getObjProps(i, i + 6, obj, {}))
}
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
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!
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");
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;
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.
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)
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.
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
})
If you are using underscore js, it is easy to remove duplicate object.
http://underscorejs.org/#uniq
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;
}
I have an array like this:
[{prop1:"abc",prop2:"qwe"},{prop1:"bnmb",prop2:"yutu"},{prop1:"zxvz",prop2:"qwrq"},...]
How can I get the index of the object that matches a condition, without iterating over the entire array?
For instance, given prop2=="yutu", I want to get index 1.
I saw .indexOf() but think it's used for simple arrays like ["a1","a2",...]. I also checked $.grep() but this returns objects, not the index.
As of 2016, you're supposed to use Array.findIndex (an ES2015/ES6 standard) for this:
a = [
{prop1:"abc",prop2:"qwe"},
{prop1:"bnmb",prop2:"yutu"},
{prop1:"zxvz",prop2:"qwrq"}];
index = a.findIndex(x => x.prop2 ==="yutu");
console.log(index);
It's supported in Google Chrome, Firefox and Edge. For Internet Explorer, there's a polyfill on the linked page.
Performance note
Function calls are expensive, therefore with really big arrays a simple loop will perform much better than findIndex:
let test = [];
for (let i = 0; i < 1e6; i++)
test.push({prop: i});
let search = test.length - 1;
let count = 100;
console.time('findIndex/predefined function');
let fn = obj => obj.prop === search;
for (let i = 0; i < count; i++)
test.findIndex(fn);
console.timeEnd('findIndex/predefined function');
console.time('findIndex/dynamic function');
for (let i = 0; i < count; i++)
test.findIndex(obj => obj.prop === search);
console.timeEnd('findIndex/dynamic function');
console.time('loop');
for (let i = 0; i < count; i++) {
for (let index = 0; index < test.length; index++) {
if (test[index].prop === search) {
break;
}
}
}
console.timeEnd('loop');
As with most optimizations, this should be applied with care and only when actually needed.
How can I get the index of the object tha match a condition (without iterate along the array)?
You cannot, something has to iterate through the array (at least once).
If the condition changes a lot, then you'll have to loop through and look at the objects therein to see if they match the condition. However, on a system with ES5 features (or if you install a shim), that iteration can be done fairly concisely:
var index;
yourArray.some(function(entry, i) {
if (entry.prop2 == "yutu") {
index = i;
return true;
}
});
That uses the new(ish) Array#some function, which loops through the entries in the array until the function you give it returns true. The function I've given it saves the index of the matching entry, then returns true to stop the iteration.
Or of course, just use a for loop. Your various iteration options are covered in this other answer.
But if you're always going to be using the same property for this lookup, and if the property values are unique, you can loop just once and create an object to map them:
var prop2map = {};
yourArray.forEach(function(entry) {
prop2map[entry.prop2] = entry;
});
(Or, again, you could use a for loop or any of your other options.)
Then if you need to find the entry with prop2 = "yutu", you can do this:
var entry = prop2map["yutu"];
I call this "cross-indexing" the array. Naturally, if you remove or add entries (or change their prop2 values), you need to update your mapping object as well.
What TJ Crowder said, everyway will have some kind of hidden iteration, with lodash this becomes:
var index = _.findIndex(array, {prop2: 'yutu'})
var CarId = 23;
//x.VehicleId property to match in the object array
var carIndex = CarsList.map(function (x) { return x.VehicleId; }).indexOf(CarId);
And for basic array numbers you can also do this:
var numberList = [100,200,300,400,500];
var index = numberList.indexOf(200); // 1
You will get -1 if it cannot find a value in the array.
var index;
yourArray.some(function (elem, i) {
return elem.prop2 === 'yutu' ? (index = i, true) : false;
});
Iterate over all elements of array.
It returns either the index and true or false if the condition does not match.
Important is the explicit return value of true (or a value which boolean result is true). The single assignment is not sufficient, because of a possible index with 0 (Boolean(0) === false), which would not result an error but disables the break of the iteration.
Edit
An even shorter version of the above:
yourArray.some(function (elem, i) {
return elem.prop2 === 'yutu' && ~(index = i);
});
Using Array.map() and Array.indexOf(string)
const arr = [{
prop1: "abc",
prop2: "qwe"
}, {
prop1: "bnmb",
prop2: "yutu"
}, {
prop1: "zxvz",
prop2: "qwrq"
}]
const index = arr.map(i => i.prop2).indexOf("yutu");
console.log(index);
The best & fastest way to do this is:
const products = [
{ prop1: 'telephone', prop2: 996 },
{ prop1: 'computadora', prop2: 1999 },
{ prop1: 'bicicleta', prop2: 995 },
];
const index = products.findIndex(el => el.prop2 > 1000);
console.log(index); // 1
I have seen many solutions in the above.
Here I am using map function to find the index of the search text in an array object.
I am going to explain my answer with using students data.
step 1: create array object for the students(optional you can create your own array object).
var students = [{name:"Rambabu",htno:"1245"},{name:"Divya",htno:"1246"},{name:"poojitha",htno:"1247"},{name:"magitha",htno:"1248"}];
step 2: Create variable to search text
var studentNameToSearch = "Divya";
step 3: Create variable to store matched index(here we use map function to iterate).
var matchedIndex = students.map(function (obj) { return obj.name; }).indexOf(studentNameToSearch);
var students = [{name:"Rambabu",htno:"1245"},{name:"Divya",htno:"1246"},{name:"poojitha",htno:"1247"},{name:"magitha",htno:"1248"}];
var studentNameToSearch = "Divya";
var matchedIndex = students.map(function (obj) { return obj.name; }).indexOf(studentNameToSearch);
console.log(matchedIndex);
alert("Your search name index in array is:"+matchedIndex)
You can use the Array.prototype.some() in the following way (as mentioned in the other answers):
https://jsfiddle.net/h1d69exj/2/
function findIndexInData(data, property, value) {
var result = -1;
data.some(function (item, i) {
if (item[property] === value) {
result = i;
return true;
}
});
return result;
}
var data = [{prop1:"abc",prop2:"qwe"},{prop1:"bnmb",prop2:"yutu"},{prop1:"zxvz",prop2:"qwrq"}]
alert(findIndexInData(data, 'prop2', "yutu")); // shows index of 1
function findIndexByKeyValue(_array, key, value) {
for (var i = 0; i < _array.length; i++) {
if (_array[i][key] == value) {
return i;
}
}
return -1;
}
var a = [
{prop1:"abc",prop2:"qwe"},
{prop1:"bnmb",prop2:"yutu"},
{prop1:"zxvz",prop2:"qwrq"}];
var index = findIndexByKeyValue(a, 'prop2', 'yutu');
console.log(index);
Try this code
var x = [{prop1:"abc",prop2:"qwe"},{prop1:"bnmb",prop2:"yutu"},{prop1:"zxvz",prop2:"qwrq"}]
let index = x.findIndex(x => x.prop1 === 'zxvz')
Another easy way is :
function getIndex(items) {
for (const [index, item] of items.entries()) {
if (item.prop2 === 'yutu') {
return index;
}
}
}
const myIndex = getIndex(myArray);
Georg have already mentioned ES6 have Array.findIndex for this.
And some other answers are workaround for ES5 using Array.some method.
One more elegant approach can be
var index;
for(index = yourArray.length; index-- > 0 && yourArray[index].prop2 !== "yutu";);
At the same time I will like to emphasize, Array.some may be implemented with binary or other efficient searching technique. So, it might perform better over for loop in some browser.
Why do you not want to iterate exactly ? The new Array.prototype.forEach are great for this purpose!
You can use a Binary Search Tree to find via a single method call if you want. This is a neat implementation of BTree and Red black Search tree in JS - https://github.com/vadimg/js_bintrees - but I'm not sure whether you can find the index at the same time.
One step using Array.reduce() - no jQuery
var items = [{id: 331}, {id: 220}, {id: 872}];
var searchIndexForId = 220;
var index = items.reduce(function(searchIndex, item, index){
if(item.id === searchIndexForId) {
console.log('found!');
searchIndex = index;
}
return searchIndex;
}, null);
will return null if index was not found.
var list = [
{prop1:"abc",prop2:"qwe"},
{prop1:"bnmb",prop2:"yutu"},
{prop1:"zxvz",prop2:"qwrq"}
];
var findProp = p => {
var index = -1;
$.each(list, (i, o) => {
if(o.prop2 == p) {
index = i;
return false; // break
}
});
return index; // -1 == not found, else == index
}
I have an array like this:
employees = [
{
"id": 1,
"shift_id": 1,
"days": {
"2012-03-01": 1,
"2012-03-02": 1,
"2012-03-03": 1,
"2012-03-04": 0,
"2012-03-05": 0,
"2012-03-06": 0
}},
{
"id": 2,
"shift_id": 1,
"days": {
"2012-03-01": 0,
"2012-03-02": 1,
"2012-03-03": 1,
"2012-03-04": 1,
"2012-03-05": 1,
"2012-03-06": 0
}},
{
"id": 3,
"shift_id": 2,
"days": {
"2012-03-01": 0,
"2012-03-02": 0,
"2012-03-03": 1,
"2012-03-04": 1,
"2012-03-05": 1,
"2012-03-06": 1
}}
];
is there a way to access an element in this array using the id value?
maybe something in jquery?
like $(employees('id = 1');
Just loop through your array and check for the id:
var yourId = 1;
for (var i = 0, len = employees.length; i < len; i++) {
if (employees[i].id == yourId) {
// ...
}
}
You can use a function like this, which filters the array appropriately:
var getEmployee = function (id) {
return employees.filter(function(i) { return i.id == id; });
};
You can use .grep() method documented here:
var employee = $.grep(employees, function(e) { return e.id == 1 })[0];
Well, there's a jQuery way of doing it:
var findElementById = function(elements, id) {
return $.grep(elements, function(e) { return e.id === id; })[0];
}
Still I wonder why don't you just index the source array by id instead.
Maybe you are looking for something like the below:
$.grep(employees, function(n){return n.id==1});
Or this:
$.each(employee, function(){
if(this["id"] == 2){
console.log(this);
}
});
As far as I am aware, in order to achieve that you would have to loop through them
Array.prototype.getObjectById = function(x){
var catcher = false, i = 0;
while(!catcher){
catcher = this[i].id == x ? this[i] : false;
i++;
}
return catcher;
}
This function should help. It will extend the array object so you can use it as myArray.getObjectbyId(id);
By design, this will return the first object that meets the criteria. You could extend it like so:
Array.prototype.getObjectsById = function(x){
var catcher = [], i = 0;
for(var i = 0; i < this.length; i++){
if(this[i].id == value){
catcher.push(this[i]);
}
i++;
}
return catcher.length == 1 ? catcher[0] : catcher;
}
This will return an array of objects if more than one object matches the criteria.
Array.prototype.getObjectsByAttribute = function(x, criteria){
if(!criteria){criteria = 'id';}
var catcher = [], i = 0;
for(var i = 0; i < this.length; i++){
if(this[i].criteria == value){
catcher.push(this[i]);
}
i++;
}
return catcher.length == 1 ? catcher[0] : catcher;
}
This extends it further to look for any criteria.
I know this question is old, but for future reference if anyone else stumbles upon this question ...
Instead of trying to over-engineer a function to parse/examine your JSON, consider changing the structure of your data to suit its purpose.
Consider the example in the question:
data = [ {
"id": 1,
"shift_id": 1,
"days": {
"2012-03-01": 1,
"2012-03-02": 1,
"2012-03-03": 1,
"2012-03-04": 0,
"2012-03-05": 0,
"2012-03-06": 0
}}, { .. }, {...} ]
Structuring the data in this way only gives you sequential access to the objects with no way to lookup an object by a particular index. Array indices are generally meaningless in this context.
data[0] => { id : 1, .. }
data[1] => { id : 2, .. }
What happens if the id is non-sequential or alphanumeric?
Using an array wont help you search any faster, you'll still have to loop...
Instead consider using a hash table/object:
{
'id1' => { id : 1 , data : { ... } },
'id99' => { id : 99, data : { ... } },
'id2' => { id : 2 , data : { ... } },
}
You can use a string value for the key and get direct access to the data by doing something like:
data['id2'] => { id : 2, ... }
Which will give you direct access to the data you want to find (by id). By simply re-organizing the structure of the data we were able to go from a O(n) search to an O(1) search.
Keep in mind that this method may work better for some solutions than others, and there are a number of other considerations to make.
This is just one approach to how you might solve a problem when you want to lookup data by a unique property.
The accepted answer is great - modified a bit for an AngularJS app:
$rootScope.getObjectsByAttribute = function(inarry,infldnm,infldval){
// This will iterate through a fetchAll sql result set and return rows where fldnm==fldval
// If it finds 1 row it returns a single object, if more than that it returns an array of objects
// Usage: result = $rootScope.getObjectsByAttribute(myarray,'myfldnm',myfldval);
if(!infldnm){infldnm = 'id';}
var catcher = [], i = 0;
for(i = 0; i < inarry.length; i++){
if(inarry[i][infldnm] == infldval){
catcher.push(inarry[i]);
}
}
return catcher.length == 1 ? catcher[0] : catcher;
}
What is the most efficient way to create an arbitrary length zero filled array in JavaScript?
ES6 introduces Array.prototype.fill. It can be used like this:
new Array(len).fill(0);
Not sure if it's fast, but I like it because it's short and self-describing.
It's still not in IE (check compatibility), but there's a polyfill available.
Although this is an old thread, I wanted to add my 2 cents to it. Not sure how slow/fast this is, but it's a quick one liner. Here is what I do:
If I want to pre-fill with a number:
Array.apply(null, Array(5)).map(Number.prototype.valueOf,0);
// [0, 0, 0, 0, 0]
If I want to pre-fill with a string:
Array.apply(null, Array(3)).map(String.prototype.valueOf,"hi")
// ["hi", "hi", "hi"]
Other answers have suggested:
new Array(5+1).join('0').split('')
// ["0", "0", "0", "0", "0"]
but if you want 0 (the number) and not "0" (zero inside a string), you can do:
new Array(5+1).join('0').split('').map(parseFloat)
// [0, 0, 0, 0, 0]
In short
Fastest solution:
let a = new Array(n); for (let i=0; i<n; ++i) a[i] = 0;
Shortest (handy) solution (3x slower for small arrays, slightly slower for big (slowest on Firefox))
Array(n).fill(0)
Details
Today 2020.06.09 I perform tests on macOS High Sierra 10.13.6 on browsers Chrome 83.0, Firefox 77.0, and Safari 13.1. I test chosen solutions for two test cases
small array - with 10 elements - you can perform test HERE
big arrays - with 1M elements - you can perform test HERE
Conclusions
solution based on new Array(n)+for (N) is fastest solution for small arrays and big arrays (except Chrome but still very fast there) and it is recommended as fast cross-browser solution
solution based on new Float32Array(n) (I) returns non typical array (e.g. you cannot call push(..) on it) so I not compare its results with other solutions - however this solution is about 10-20x faster than other solutions for big arrays on all browsers
solutions based on for (L,M,N,O) are fast for small arrays
solutions based on fill (B,C) are fast on Chrome and Safari but surprisingly slowest on Firefox for big arrays. They are medium fast for small arrays
solution based on Array.apply (P) throws error for big arrays
function P(n) {
return Array.apply(null, Array(n)).map(Number.prototype.valueOf,0);
}
try {
P(1000000);
} catch(e) {
console.error(e.message);
}
Code and example
Below code presents solutions used in measurements
function A(n) {
return [...new Array(n)].fill(0);
}
function B(n) {
return new Array(n).fill(0);
}
function C(n) {
return Array(n).fill(0);
}
function D(n) {
return Array.from({length: n}, () => 0);
}
function E(n) {
return [...new Array(n)].map(x => 0);
}
// arrays with type
function F(n) {
return Array.from(new Int32Array(n));
}
function G(n) {
return Array.from(new Float32Array(n));
}
function H(n) {
return Array.from(new Float64Array(n)); // needs 2x more memory than float32
}
function I(n) {
return new Float32Array(n); // this is not typical array
}
function J(n) {
return [].slice.apply(new Float32Array(n));
}
// Based on for
function K(n) {
let a = [];
a.length = n;
let i = 0;
while (i < n) {
a[i] = 0;
i++;
}
return a;
}
function L(n) {
let a=[]; for(let i=0; i<n; i++) a[i]=0;
return a;
}
function M(n) {
let a=[]; for(let i=0; i<n; i++) a.push(0);
return a;
}
function N(n) {
let a = new Array(n); for (let i=0; i<n; ++i) a[i] = 0;
return a;
}
function O(n) {
let a = new Array(n); for (let i=n; i--;) a[i] = 0;
return a;
}
// other
function P(n) {
return Array.apply(null, Array(n)).map(Number.prototype.valueOf,0);
}
function Q(n) {
return "0".repeat( n ).split("").map( parseFloat );
}
function R(n) {
return new Array(n+1).join('0').split('').map(parseFloat)
}
// ---------
// TEST
// ---------
[A,B,C,D,E,F,G,H,I,J,K,L,M,N,O,P,Q,R].forEach(f => {
let a = f(10);
console.log(`${f.name} length=${a.length}, arr[0]=${a[0]}, arr[9]=${a[9]}`)
});
This snippets only present used codes
Example results for Chrome:
Elegant way to fill an array with precomputed values
Here is another way to do it using ES6 that nobody has mentioned so far:
> Array.from(Array(3), () => 0)
< [0, 0, 0]
It works by passing a map function as the second parameter of Array.from.
In the example above, the first parameter allocates an array of 3 positions filled with the value undefined and then the lambda function maps each one of them to the value 0.
Although Array(len).fill(0) is shorter, it doesn't work if you need to fill the array by doing some computation first (I know the question didn't ask for it, but a lot of people end up here looking for this).
For instance, if you need an array with 10 random numbers:
> Array.from(Array(10), () => Math.floor(10 * Math.random()))
< [3, 6, 8, 1, 9, 3, 0, 6, 7, 1]
It's more concise (and elegant) than the equivalent:
const numbers = Array(10);
for (let i = 0; i < numbers.length; i++) {
numbers[i] = Math.round(10 * Math.random());
}
This method can also be used to generate sequences of numbers by taking advantage of the index parameter provided in the callback:
> Array.from(Array(10), (d, i) => i)
< [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
Bonus answer: fill an array using String repeat()
Since this answer is getting a good deal of attention, I also wanted to show this cool trick. Although not as useful as my main answer, will introduce the still not very known, but very useful String repeat() method. Here's the trick:
> "?".repeat(10).split("").map(() => Math.floor(10 * Math.random()))
< [5, 6, 3, 5, 0, 8, 2, 7, 4, 1]
Cool, huh? repeat() is a very useful method to create a string that is the repetition of the original string a certain number of times. After that, split() creates an array for us, which is then map()ped to the values we want. Breaking it down in steps:
> "?".repeat(10)
< "??????????"
> "?".repeat(10).split("")
< ["?", "?", "?", "?", "?", "?", "?", "?", "?", "?"]
> "?".repeat(10).split("").map(() => Math.floor(10 * Math.random()))
< [5, 6, 3, 5, 0, 8, 2, 7, 4, 1]
The already mentioned ES 6 fill method takes care of this nicely. Most modern desktop browsers already support the required Array prototype methods as of today (Chromium, FF, Edge and Safari) [1]. You can look up details on MDN. A simple usage example is
a = new Array(10).fill(0);
Given the current browser support you should be cautious to use this unless you are sure your audience uses modern Desktop browsers.
Note added August 2013, updated February 2015: The answer below from 2009 relates to JavaScript's generic Array type. It doesn't relate to the newer typed arrays defined in ES2015 [and available now in many browsers], like Int32Array and such. Also note that ES2015 adds a fill method to both Arrays and typed arrays, which is likely to be the most efficient way to fill them...
Also, it can make a big difference to some implementations how you create the array. Chrome's V8 engine, in particular, tries to use a highly-efficient, contiguous-memory array if it thinks it can, shifting to the object-based array only when necessary.
With most languages, it would be pre-allocate, then zero-fill, like this:
function newFilledArray(len, val) {
var rv = new Array(len);
while (--len >= 0) {
rv[len] = val;
}
return rv;
}
But, JavaScript arrays aren't really arrays, they're key/value maps just like all other JavaScript objects, so there's no "pre-allocate" to do (setting the length doesn't allocate that many slots to fill), nor is there any reason to believe that the benefit of counting down to zero (which is just to make the comparison in the loop fast) isn't outweighed by adding the keys in reverse order when the implementation may well have optimized their handling of the keys related to arrays on the theory you'll generally do them in order.
In fact, Matthew Crumley pointed out that counting down is markedly slower on Firefox than counting up, a result I can confirm — it's the array part of it (looping down to zero is still faster than looping up to a limit in a var). Apparently adding the elements to the array in reverse order is a slow op on Firefox. In fact, the results vary quite a bit by JavaScript implementation (which isn't all that surprising). Here's a quick and dirty test page (below) for browser implementations (very dirty, doesn't yield during tests, so provides minimal feedback and will run afoul of script time limits). I recommend refreshing between tests; FF (at least) slows down on repeated tests if you don't.
The fairly complicated version that uses Array#concat is faster than a straight init on FF as of somewhere between 1,000 and 2,000 element arrays. On Chrome's V8 engine, though, straight init wins out every time...
Here's a test:
const tests = [
{
name: "downpre",
total: 0,
desc: "Count down, pre-decrement",
func: makeWithCountDownPre
},
{
name: "downpost",
total: 0,
desc: "Count down, post-decrement",
func: makeWithCountDownPost
},
{
name: "up",
total: 0,
desc: "Count up (normal)",
func: makeWithCountUp
},
{
name: "downandup",
total: 0,
desc: "Count down (for loop) and up (for filling)",
func: makeWithCountDownArrayUp
},
{
name: "concat",
total: 0,
desc: "Concat",
func: makeWithConcat
}
];
const q = sel => document.querySelector(sel);
let markup = "";
for (const {name, desc} of tests) {
markup += `
<div><input type="checkbox" id="chk_${name}" checked>
<label for="chk_${name}">${desc}</label></div>`;
}
q("#checkboxes").innerHTML = markup;
q("#btnTest").addEventListener("click", btnTestClick);
function btnTestClick() {
// Clear log
q("#log").innerHTML = "Testing...";
// Show running
q("#btnTest").disabled = true;
// Run after a pause while the browser updates display
setTimeout(btnTestClickPart2, 0);
}
function btnTestClickPart2() {
try {
runTests();
} catch (e) {
log(`Exception: ${e.message}`);
}
// Re-enable the button
q("#btnTest").disabled = false;
}
function getNumField(name) {
const val = q("#" + name).value.trim();
const num = /^\d+$/.test(val) ? parseInt(val) : NaN;
if (isNaN(num) || num <= 0) {
throw new Error(`Invalid ${name} value ${JSON.stringify(val)}`);
}
return num;
}
function runTests() {
try {
// Clear log
q("#log").innerHTML = "";
const runCount = getNumField("loops");
const length = getNumField("length");
// Do it (we run runCount + 1 times, first time is a warm up)
for (let counter = 0; counter <= runCount; ++counter) {
for (const test of tests) {
if (q("#chk_" + test.name).checked) {
const start = Date.now();
const a = test.func(length);
const time = Date.now() - start;
if (counter == 0) {
// Don't count (warm up), but do check the algorithm works
const invalid = validateResult(a, length);
if (invalid) {
log(`<span class=error>FAILURE</span> with test ${test.name}: ${invalid}`);
return;
}
} else {
// Count this one
log(`#${counter}: ${test.desc}: ${time}ms`);
test.total += time;
}
}
}
}
for (const test of tests) {
if (q("#chk_" + test.name).checked) {
test.avg = test.total / runCount;
if (typeof lowest != "number" || lowest > test.avg) {
lowest = test.avg;
}
}
}
let results =
"<p>Results:" +
"<br>Length: " + length +
"<br>Loops: " + runCount +
"</p>";
for (const test of tests) {
if (q("#chk_" + test.name).checked) {
results +=
`<p ${lowest == test.avg ? " class=winner" : ""}>${test.desc}, average time: ${test.avg}ms</p>`;
}
}
results += "<hr>";
q("#log").insertAdjacentHTML("afterbegin", results);
} catch (e) {
log(e.message);
return;
}
}
function validateResult(a, length) {
if (a.length != length) {
return "Length is wrong";
}
for (let n = length - 1; n >= 0; --n) {
if (a[n] != 0) {
return "Index " + n + " is not zero";
}
}
return undefined;
}
function makeWithCountDownPre(len) {
const a = new Array(len);
while (--len >= 0) {
a[len] = 0;
}
return a;
}
function makeWithCountDownPost(len) {
const a = new Array(len);
while (len-- > 0) {
a[len] = 0;
}
return a;
}
function makeWithCountUp(len) {
const a = new Array(len);
for (let i = 0; i < len; ++i) {
a[i] = 0;
}
return a;
}
function makeWithCountDownArrayUp(len) {
const a = new Array(len);
let i = 0;
while (--len >= 0) {
a[i++] = 0;
}
return a;
}
function makeWithConcat(len) {
if (len == 0) {
return [];
}
let a = [0];
let currlen = 1;
while (currlen < len) {
const rem = len - currlen;
if (rem < currlen) {
a = a.concat(a.slice(0, rem));
} else {
a = a.concat(a);
}
currlen = a.length;
}
return a;
}
function log(msg) {
const p = document.createElement("p");
p.textContent = msg;
q("#log").appendChild(p);
}
body {
font-family: sans-serif;
}
#log p {
margin: 0;
padding: 0;
}
.error {
color: red;
}
.winner {
color: green;
}
<div>
<label for='txtLength'>Length:</label><input type='text' id='length' value='1000'>
<br><label for='txtLoops'>Loops:</label><input type='text' id='loops' value='100000'>
<div id='checkboxes'></div>
<br><input type='button' id='btnTest' value='Test'>
<hr>
<div id='log'></div>
</div>
If you use ES6, you can use Array.from() like this:
Array.from({ length: 3 }, () => 0);
//[0, 0, 0]
Has the same result as
Array.from({ length: 3 }).map(() => 0)
//[0, 0, 0]
Because
Array.from({ length: 3 })
//[undefined, undefined, undefined]
By default Uint8Array, Uint16Array and Uint32Array classes keep zeros as its values, so you don't need any complex filling techniques, just do:
var ary = new Uint8Array(10);
all elements of array ary will be zeros by default.
function makeArrayOf(value, length) {
var arr = [], i = length;
while (i--) {
arr[i] = value;
}
return arr;
}
makeArrayOf(0, 5); // [0, 0, 0, 0, 0]
makeArrayOf('x', 3); // ['x', 'x', 'x']
Note that while is usually more efficient than for-in, forEach, etc.
using object notation
var x = [];
zero filled? like...
var x = [0,0,0,0,0,0];
filled with 'undefined'...
var x = new Array(7);
obj notation with zeros
var x = [];
for (var i = 0; i < 10; i++) x[i] = 0;
As a side note, if you modify Array's prototype, both
var x = new Array();
and
var y = [];
will have those prototype modifications
At any rate, I wouldn't be overly concerned with the efficiency or speed of this operation, there are plenty of other things that you will likely be doing that are far more wasteful and expensive than instanciating an array of arbitrary length containing zeros.
I've tested all combinations of pre-allocating/not pre-allocating, counting up/down, and for/while loops in IE 6/7/8, Firefox 3.5, Chrome, and Opera.
The functions below was consistently the fastest or extremely close in Firefox, Chrome, and IE8, and not much slower than the fastest in Opera and IE 6. It's also the simplest and clearest in my opinion. I've found several browsers where the while loop version is slightly faster, so I'm including it too for reference.
function newFilledArray(length, val) {
var array = [];
for (var i = 0; i < length; i++) {
array[i] = val;
}
return array;
}
or
function newFilledArray(length, val) {
var array = [];
var i = 0;
while (i < length) {
array[i++] = val;
}
return array;
}
ES6 solution:
[...new Array(5)].map(x => 0); // [0, 0, 0, 0, 0]
const arr = Array.from({ length: 10 }).fill(0);
console.log(arr)
If you need to create many zero filled arrays of different lengths during the execution of your code, the fastest way I've found to achieve this is to create a zero array once, using one of the methods mentioned on this topic, of a length which you know will never be exceeded, and then slice that array as necessary.
For example (using the function from the chosen answer above to initialize the array), create a zero filled array of length maxLength, as a variable visible to the code that needs zero arrays:
var zero = newFilledArray(maxLength, 0);
Now slice this array everytime you need a zero filled array of length requiredLength < maxLength:
zero.slice(0, requiredLength);
I was creating zero filled arrays thousands of times during execution of my code, this speeded up the process tremendously.
function zeroFilledArray(size) {
return new Array(size + 1).join('0').split('');
}
Using lodash or underscore
_.range(0, length - 1, 0);
Or if you have an array existing and you want an array of the same length
array.map(_.constant(0));
I have nothing against:
Array.apply(null, Array(5)).map(Number.prototype.valueOf,0);
new Array(5+1).join('0').split('').map(parseFloat);
suggested by Zertosh, but in a new ES6 array extensions allow you to do this natively with fill method. Now IE edge, Chrome and FF supports it, but check the compatibility table
new Array(3).fill(0) will give you [0, 0, 0]. You can fill the array with any value like new Array(5).fill('abc') (even objects and other arrays).
On top of that you can modify previous arrays with fill:
arr = [1, 2, 3, 4, 5, 6]
arr.fill(9, 3, 5) # what to fill, start, end
which gives you: [1, 2, 3, 9, 9, 6]
To create an all new Array
new Array(arrayLength).fill(0);
To add some values at the end of an existing Array
[...existingArray, ...new Array(numberOfElementsToAdd).fill(0)]
Example
//**To create an all new Array**
console.log(new Array(5).fill(0));
//**To add some values at the end of an existing Array**
let existingArray = [1,2,3]
console.log([...existingArray, ...new Array(5).fill(0)]);
The way I usually do it (and is amazing fast) is using Uint8Array. For example, creating a zero filled vector of 1M elements:
var zeroFilled = [].slice.apply(new Uint8Array(1000000))
I'm a Linux user and always have worked for me, but once a friend using a Mac had some non-zero elements. I thought his machine was malfunctioning, but still here's the safest way we found to fix it:
var zeroFilled = [].slice.apply(new Uint8Array(new Array(1000000))
Edited
Chrome 25.0.1364.160
Frederik Gottlieb - 6.43
Sam Barnum - 4.83
Eli - 3.68
Joshua 2.91
Mathew Crumley - 2.67
bduran - 2.55
Allen Rice - 2.11
kangax - 0.68
Tj. Crowder - 0.67
zertosh - ERROR
Firefox 20.0
Allen Rice - 1.85
Joshua - 1.82
Mathew Crumley - 1.79
bduran - 1.37
Frederik Gottlieb - 0.67
Sam Barnum - 0.63
Eli - 0.59
kagax - 0.13
Tj. Crowder - 0.13
zertosh - ERROR
Missing the most important test (at least for me): the Node.js one. I suspect it close to Chrome benchmark.
As of ECMAScript2016, there is one clear choice for large arrays.
Since this answer still shows up near the top on google searches, here's an answer for 2017.
Here's a current jsbench with a few dozen popular methods, including many proposed up to now on this question. If you find a better method please add, fork and share.
I want to note that there is no true most efficient way to create an arbitrary length zero filled array. You can optimize for speed, or for clarity and maintainability - either can be considered the more efficient choice depending on the needs of the project.
When optimizing for speed, you want to: create the array using literal syntax; set the length, initialize iterating variable, and iterate through the array using a while loop. Here's an example.
const arr = [];
arr.length = 120000;
let i = 0;
while (i < 120000) {
arr[i] = 0;
i++;
}
Another possible implementation would be:
(arr = []).length = n;
let i = 0;
while (i < n) {
arr[i] = 0;
i++;
}
But I strongly discourage using this second implantation in practice as it's less clear and doesn't allow you to maintain block scoping on your array variable.
These are significantly faster than filling with a for loop, and about 90% faster than the standard method of
const arr = Array(n).fill(0);
But this fill method is still the most efficient choice for smaller arrays due to it's clarity, conciseness and maintainability. The performance difference likely won't kill you unless you're making a lot of arrays with lengths on the order of thousands or more.
A few other important notes. Most style guides recommend you no longer use varwithout a very special reason when using ES6 or later. Use const for variables that won't be redefined and let for variables that will. The MDN and Airbnb's Style Guide are great places to go for more information on best practices. The questions wasn't about syntax, but it's important that people new to JS know about these new standards when searching through these reams of old and new answers.
Didn't see this method in answers, so here it is:
"0".repeat( 200 ).split("").map( parseFloat )
In result you will get zero-valued array of length 200:
[ 0, 0, 0, 0, ... 0 ]
I'm not sure about the performance of this code, but it shouldn't be an issue if you use it for relatively small arrays.
What about new Array(51).join('0').split('')?
let filled = [];
filled.length = 10;
filled.fill(0);
console.log(filled);
This concat version is much faster in my tests on Chrome (2013-03-21). About 200ms for 10,000,000 elements vs 675 for straight init.
function filledArray(len, value) {
if (len <= 0) return [];
var result = [value];
while (result.length < len/2) {
result = result.concat(result);
}
return result.concat(result.slice(0, len-result.length));
}
Bonus: if you want to fill your array with Strings, this is a concise way to do it (not quite as fast as concat though):
function filledArrayString(len, value) {
return new Array(len+1).join(value).split('');
}
I was testing out the great answer by T.J. Crowder, and came up with a recursive merge based on the concat solution that outperforms any in his tests in Chrome (i didn't test other browsers).
function makeRec(len, acc) {
if (acc == null) acc = [];
if (len <= 1) return acc;
var b = makeRec(len >> 1, [0]);
b = b.concat(b);
if (len & 1) b = b.concat([0]);
return b;
},
call the method with makeRec(29).
It might be worth pointing out, that Array.prototype.fill had been added as part of the ECMAScript 6 (Harmony) proposal. I would rather go with the polyfill written below, before considering other options mentioned on the thread.
if (!Array.prototype.fill) {
Array.prototype.fill = function(value) {
// Steps 1-2.
if (this == null) {
throw new TypeError('this is null or not defined');
}
var O = Object(this);
// Steps 3-5.
var len = O.length >>> 0;
// Steps 6-7.
var start = arguments[1];
var relativeStart = start >> 0;
// Step 8.
var k = relativeStart < 0 ?
Math.max(len + relativeStart, 0) :
Math.min(relativeStart, len);
// Steps 9-10.
var end = arguments[2];
var relativeEnd = end === undefined ?
len : end >> 0;
// Step 11.
var final = relativeEnd < 0 ?
Math.max(len + relativeEnd, 0) :
Math.min(relativeEnd, len);
// Step 12.
while (k < final) {
O[k] = value;
k++;
}
// Step 13.
return O;
};
}
Shortest for loop code
a=i=[];for(;i<100;)a[i++]=0;
edit:
for(a=i=[];i<100;)a[i++]=0;
or
for(a=[],i=100;i--;)a[i]=0;
Safe var version
var a=[],i=0;for(;i<100;)a[i++]=0;
edit:
for(var i=100,a=[];i--;)a[i]=0;
My fastest function would be:
function newFilledArray(len, val) {
var a = [];
while(len--){
a.push(val);
}
return a;
}
var st = (new Date()).getTime();
newFilledArray(1000000, 0)
console.log((new Date()).getTime() - st); // returned 63, 65, 62 milliseconds
Using the native push and shift to add items to the array is much faster (about 10 times) than declaring the array scope and referencing each item to set it's value.
fyi: I consistently get faster times with the first loop, which is counting down, when running this in firebug (firefox extension).
var a = [];
var len = 1000000;
var st = (new Date()).getTime();
while(len){
a.push(0);
len -= 1;
}
console.log((new Date()).getTime() - st); // returned 863, 894, 875 milliseconds
st = (new Date()).getTime();
len = 1000000;
a = [];
for(var i = 0; i < len; i++){
a.push(0);
}
console.log((new Date()).getTime() - st); // returned 1155, 1179, 1163 milliseconds
I'm interested to know what T.J. Crowder makes of that ? :-)
I knew I had this proto'd somewhere :)
Array.prototype.init = function(x,n)
{
if(typeof(n)=='undefined') { n = this.length; }
while (n--) { this[n] = x; }
return this;
}
var a = (new Array(5)).init(0);
var b = [].init(0,4);
Edit: tests
In response to Joshua and others methods I ran my own benchmarking, and I'm seeing completely different results to those reported.
Here's what I tested:
//my original method
Array.prototype.init = function(x,n)
{
if(typeof(n)=='undefined') { n = this.length; }
while (n--) { this[n] = x; }
return this;
}
//now using push which I had previously thought to be slower than direct assignment
Array.prototype.init2 = function(x,n)
{
if(typeof(n)=='undefined') { n = this.length; }
while (n--) { this.push(x); }
return this;
}
//joshua's method
function newFilledArray(len, val) {
var a = [];
while(len--){
a.push(val);
}
return a;
}
//test m1 and m2 with short arrays many times 10K * 10
var a = new Date();
for(var i=0; i<10000; i++)
{
var t1 = [].init(0,10);
}
var A = new Date();
var b = new Date();
for(var i=0; i<10000; i++)
{
var t2 = [].init2(0,10);
}
var B = new Date();
//test m1 and m2 with long array created once 100K
var c = new Date();
var t3 = [].init(0,100000);
var C = new Date();
var d = new Date();
var t4 = [].init2(0,100000);
var D = new Date();
//test m3 with short array many times 10K * 10
var e = new Date();
for(var i=0; i<10000; i++)
{
var t5 = newFilledArray(10,0);
}
var E = new Date();
//test m3 with long array created once 100K
var f = new Date();
var t6 = newFilledArray(100000, 0)
var F = new Date();
Results:
IE7 deltas:
dA=156
dB=359
dC=125
dD=375
dE=468
dF=412
FF3.5 deltas:
dA=6
dB=13
dC=63
dD=8
dE=12
dF=8
So by my reckoning push is indeed slower generally but performs better with longer arrays in FF but worse in IE which just sucks in general (quel surprise).
Anonymous function:
(function(n) { while(n-- && this.push(0)); return this; }).call([], 5);
// => [0, 0, 0, 0, 0]
A bit shorter with for-loop:
(function(n) { for(;n--;this.push(0)); return this; }).call([], 5);
// => [0, 0, 0, 0, 0]
Works with any Object, just change what's inside this.push().
You can even save the function:
function fill(size, content) {
for(;size--;this.push(content));
return this;
}
Call it using:
var helloArray = fill.call([], 5, 'hello');
// => ['hello', 'hello', 'hello', 'hello', 'hello']
Adding elements to an already existing array:
var helloWorldArray = fill.call(helloArray, 5, 'world');
// => ['hello', 'hello', 'hello', 'hello', 'hello', 'world', 'world', 'world', 'world', 'world']
Performance: http://jsperf.com/zero-filled-array-creation/25