If I clone an array, I use cloneArr = arr.slice()
I want to know how to clone an object in nodejs.
For utilities and classes where there is no need to squeeze every drop of performance, I often cheat and just use JSON to perform a deep copy:
function clone(a) {
return JSON.parse(JSON.stringify(a));
}
This isn't the only answer or the most elegant answer; all of the other answers should be considered for production bottlenecks. However, this is a quick and dirty solution, quite effective, and useful in most situations where I would clone a simple hash of properties.
Object.assign hasn't been mentioned in any of above answers.
let cloned = Object.assign({}, source);
If you're on ES6 you can use the spread operator:
let cloned = { ... source };
Reference: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Object/assign
There are some Node modules out there if don't want to "roll your own". This one looks good: https://www.npmjs.com/package/clone
Looks like it handles all kinds of stuff, including circular references. From the github page:
clone masters cloning objects, arrays, Date objects, and RegEx
objects. Everything is cloned recursively, so that you can clone dates
in arrays in objects, for example. [...] Circular references? Yep!
You can use lodash as well. It has a clone and cloneDeep methods.
var _= require('lodash');
var objects = [{ 'a': 1 }, { 'b': 2 }];
var shallow = _.clone(objects);
console.log(shallow[0] === objects[0]);
// => true
var deep = _.cloneDeep(objects);
console.log(deep[0] === objects[0]);
It's hard to do a generic but useful clone operation because what should be cloned recursively and what should be just copied depends on how the specific object is supposed to work.
Something that may be useful is
function clone(x)
{
if (x === null || x === undefined)
return x;
if (typeof x.clone === "function")
return x.clone();
if (x.constructor == Array)
{
var r = [];
for (var i=0,n=x.length; i<n; i++)
r.push(clone(x[i]));
return r;
}
return x;
}
In this code the logic is
in case of null or undefined just return the same (the special case is needed because it's an error to try to see if a clone method is present)
does the object have a clone method ? then use that
is the object an array ? then do a recursive cloning operation
otherwise just return the same value
This clone function should allow implementing custom clone methods easily... for example
function Point(x, y)
{
this.x = x;
this.y = y;
...
}
Point.prototype.clone = function()
{
return new Point(this.x, this.y);
};
function Polygon(points, style)
{
this.points = points;
this.style = style;
...
}
Polygon.prototype.clone = function()
{
return new Polygon(clone(this.points),
this.style);
};
When in the object you know that a correct cloning operation for a specific array is just a shallow copy then you can call values.slice() instead of clone(values).
For example in the above code I am explicitly requiring that a cloning of a polygon object will clone the points, but will share the same style object. If I want to clone the style object too instead then I can just pass clone(this.style).
There is no native method for cloning objects. Underscore implements _.clone which is a shallow clone.
_.clone = function(obj) {
return _.isArray(obj) ? obj.slice() : _.extend({}, obj);
};
It either slices it or extends it.
Here's _.extend
// extend the obj (first parameter)
_.extend = function(obj) {
// for each other parameter
each(slice.call(arguments, 1), function(source) {
// loop through all properties of the other objects
for (var prop in source) {
// if the property is not undefined then add it to the object.
if (source[prop] !== void 0) obj[prop] = source[prop];
}
});
// return the object (first parameter)
return obj;
};
Extend simply iterates through all the items and creates a new object with the items in it.
You can roll out your own naive implementation if you want
function clone(o) {
var ret = {};
Object.keys(o).forEach(function (val) {
ret[val] = o[val];
});
return ret;
}
There are good reasons to avoid deep cloning because closures cannot be cloned.
I've personally asked a question about deep cloning objects before and the conclusion I came to is that you just don't do it.
My recommendation is use underscore and it's _.clone method for shallow clones
For a shallow copy, I like to use the reduce pattern (usually in a module or such), like so:
var newObject = Object.keys(original).reduce(function (obj, item) {
obj[item] = original[item];
return obj;
},{});
Here's a jsperf for a couple of the options: http://jsperf.com/shallow-copying
Old question, but there's a more elegant answer than what's been suggested so far; use the built-in utils._extend:
var extend = require("util")._extend;
var varToCopy = { test: 12345, nested: { val: 6789 } };
var copiedObject = extend({}, varToCopy);
console.log(copiedObject);
// outputs:
// { test: 12345, nested: { val: 6789 } }
Note the use of the first parameter with an empty object {} - this tells extend that the copied object(s) need to be copied to a new object. If you use an existing object as the first parameter, then the second (and all subsequent) parameters will be deep-merge-copied over the first parameter variable.
Using the example variables above, you can also do this:
var anotherMergeVar = { foo: "bar" };
extend(copiedObject, { anotherParam: 'value' }, anotherMergeVar);
console.log(copiedObject);
// outputs:
// { test: 12345, nested: { val: 6789 }, anotherParam: 'value', foo: 'bar' }
Very handy utility, especially where I'm used to extend in AngularJS and jQuery.
Hope this helps someone else; object reference overwrites are a misery, and this solves it every time!
In Node.js 17.x was added the method structuredClone() to allow made a deep clone.
Documentation of reference: https://developer.mozilla.org/en-US/docs/Web/API/structuredClone
I implemented a full deep copy. I believe its the best pick for a generic clone method, but it does not handle cyclical references.
Usage example:
parent = {'prop_chain':3}
obj = Object.create(parent)
obj.a=0; obj.b=1; obj.c=2;
obj2 = copy(obj)
console.log(obj, obj.prop_chain)
// '{'a':0, 'b':1, 'c':2} 3
console.log(obj2, obj2.prop_chain)
// '{'a':0, 'b':1, 'c':2} 3
parent.prop_chain=4
obj2.a = 15
console.log(obj, obj.prop_chain)
// '{'a':0, 'b':1, 'c':2} 4
console.log(obj2, obj2.prop_chain)
// '{'a':15, 'b':1, 'c':2} 4
The code itself:
This code copies objects with their prototypes, it also copy functions (might be useful for someone).
function copy(obj) {
// (F.prototype will hold the object prototype chain)
function F() {}
var newObj;
if(typeof obj.clone === 'function')
return obj.clone()
// To copy something that is not an object, just return it:
if(typeof obj !== 'object' && typeof obj !== 'function' || obj == null)
return obj;
if(typeof obj === 'object') {
// Copy the prototype:
newObj = {}
var proto = Object.getPrototypeOf(obj)
Object.setPrototypeOf(newObj, proto)
} else {
// If the object is a function the function evaluate it:
var aux
newObj = eval('aux='+obj.toString())
// And copy the prototype:
newObj.prototype = obj.prototype
}
// Copy the object normal properties with a deep copy:
for(var i in obj) {
if(obj.hasOwnProperty(i)) {
if(typeof obj[i] !== 'object')
newObj[i] = obj[i]
else
newObj[i] = copy(obj[i])
}
}
return newObj;
}
With this copy I can't find any difference between the original and the copied one except if the original used closures on its construction, so i think its a good implementation.
I hope it helps
Depending on what you want to do with your cloned object you can utilize the prototypal inheritence mechanism of javascript and achieve a somewhat cloned object through:
var clonedObject = Object.create(originalObject);
Just remember that this isn't a full clone - for better or worse.
A good thing about that is that you actually haven't duplicated the object so the memory footprint will be low.
Some tricky things to remember though about this method is that iteration of properties defined in the prototype chain sometimes works a bit different and the fact that any changes to the original object will affect the cloned object as well unless that property has been set on itself also.
Objects and Arrays in JavaScript use call by reference, if you update copied value it might reflect on the original object.
To prevent this you can deep clone the object, to prevent the reference to be passed, using lodash library cloneDeep method
run command
npm install lodash
const ld = require('lodash')
const objectToCopy = {name: "john", age: 24}
const clonedObject = ld.cloneDeep(objectToCopy)
Try this module for complex structures, developed especially for nodejs - https://github.com/themondays/deppcopy
Works faster than JSON stringify and parse on large structures and supports BigInt.
for array, one can use
var arr = [1,2,3];
var arr_2 = arr ;
print ( arr_2 );
arr=arr.slice(0);
print ( arr );
arr[1]=9999;
print ( arr_2 );
How about this method
const v8 = require('v8');
const structuredClone = obj => {
return v8.deserialize(v8.serialize(obj));
};
Given an object:
let myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*"
};
How do I remove the property regex to end up with the following myObject?
let myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI"
};
To remove a property from an object (mutating the object), you can do it like this:
delete myObject.regex;
// or,
delete myObject['regex'];
// or,
var prop = "regex";
delete myObject[prop];
Demo
var myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*"
};
delete myObject.regex;
console.log(myObject);
For anyone interested in reading more about it, Stack Overflow user kangax has written an incredibly in-depth blog post about the delete statement on their blog, Understanding delete. It is highly recommended.
If you'd like a new object with all the keys of the original except some, you could use destructuring.
Demo
let myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*"
};
// assign the key regex to the variable _ indicating it will be unused
const {regex: _, ...newObj} = myObject;
console.log(newObj); // has no 'regex' key
console.log(myObject); // remains unchanged
Objects in JavaScript can be thought of as maps between keys and values. The delete operator is used to remove these keys, more commonly known as object properties, one at a time.
var obj = {
myProperty: 1
}
console.log(obj.hasOwnProperty('myProperty')) // true
delete obj.myProperty
console.log(obj.hasOwnProperty('myProperty')) // false
The delete operator does not directly free memory, and it differs from simply assigning the value of null or undefined to a property, in that the property itself is removed from the object. Note that if the value of a deleted property was a reference type (an object), and another part of your program still holds a reference to that object, then that object will, of course, not be garbage collected until all references to it have disappeared.
delete will only work on properties whose descriptor marks them as configurable.
Old question, modern answer. Using object destructuring, an ECMAScript 6 feature, it's as simple as:
const { a, ...rest } = { a: 1, b: 2, c: 3 };
Or with the questions sample:
const myObject = {"ircEvent": "PRIVMSG", "method": "newURI", "regex": "^http://.*"};
const { regex, ...newObject } = myObject;
console.log(newObject);
You can see it in action in the Babel try-out editor.
Edit:
To reassign to the same variable, use a let:
let myObject = {"ircEvent": "PRIVMSG", "method": "newURI", "regex": "^http://.*"};
({ regex, ...myObject } = myObject);
console.log(myObject);
var myObject = {"ircEvent": "PRIVMSG", "method": "newURI", "regex": "^http://.*"};
delete myObject.regex;
console.log ( myObject.regex); // logs: undefined
This works in Firefox and Internet Explorer, and I think it works in all others.
The delete operator is used to remove properties from objects.
const obj = { foo: "bar" };
delete obj.foo;
obj.hasOwnProperty("foo"); // false
Note that, for arrays, this is not the same as removing an element. To remove an element from an array, use Array#splice or Array#pop. For example:
arr; // [0, 1, 2, 3, 4]
arr.splice(3,1); // 3
arr; // [0, 1, 2, 4]
Details
Strictly speaking, it's impossible to truly delete anything in JavaScript. The delete operator neither deletes objects nor frees memory. Rather, it sets its operand to undefined and manipulates the parent object so that the member is gone.
let parent = {
member: { str: "Hello" }
};
let secondref = parent.member;
delete parent.member;
parent.member; // undefined
secondref; // { str: "Hello" }
The object is not deleted. Only the reference is. Memory is only freed
by the garbage collector when all references to an object are removed.
Another important caveat is that the delete operator will not reorganize structures for you, which has results that can seem counterintuitive. Deleting an array index, for example, will leave a "hole" in it.
let array = [0, 1, 2, 3]; // [0, 1, 2, 3]
delete array[2]; // [0, 1, empty, 3]
This is because arrays are objects. So indices are the same as keys.
let fauxarray = {0: 1, 1: 2, length: 2};
fauxarray.__proto__ = [].__proto__;
fauxarray.push(3);
fauxarray; // [1, 2, 3]
Array.isArray(fauxarray); // false
Array.isArray([1, 2, 3]); // true
Different built-in functions in JavaScript handle arrays with holes in them differently.
for..in statements will skip the empty index completely.
A naive for loop will yield undefined for the value at the index.
Any method using Symbol.iterator will return undefined for the value at the index.
forEach, map and reduce will simply skip the missing index, but will not remove it
Example:
let array = [1, 2, 3]; // [1,2,3]
delete array[1]; // [1, empty, 3]
array.map(x => 0); // [0, empty, 0]
So, the delete operator should not be used for the common use-case of removing elements from an array. Arrays have a dedicated methods for removing elements and reallocating memory: Array#splice() and Array#pop.
Array#splice(start[, deleteCount[, item1[, item2[, ...]]]])
Array#splice mutates the array, and returns any removed indices. deleteCount elements are removed from index start, and item1, item2... itemN are inserted into the array from index start. If deleteCount is omitted then elements from startIndex are removed to the end of the array.
let a = [0,1,2,3,4]
a.splice(2,2) // returns the removed elements [2,3]
// ...and `a` is now [0,1,4]
There is also a similarly named, but different, function on Array.prototype: Array#slice.
Array#slice([begin[, end]])
Array#slice is non-destructive, and returns a new array containing the indicated indices from start to end. If end is left unspecified, it defaults to the end of the array. If end is positive, it specifies the zero-based non-inclusive index to stop at. If end is negative it, it specifies the index to stop at by counting back from the end of the array (eg. -1 will omit the final index). If end <= start, the result is an empty array.
let a = [0,1,2,3,4]
let slices = [
a.slice(0,2),
a.slice(2,2),
a.slice(2,3),
a.slice(2,5) ]
// a [0,1,2,3,4]
// slices[0] [0 1]- - -
// slices[1] - - - - -
// slices[2] - -[3]- -
// slices[3] - -[2 4 5]
Array#pop
Array#pop removes the last element from an array, and returns that element. This operation changes the length of the array. The opposite operation is push
Array#shift
Array#shift is similar to pop, except it removes the first element. The opposite operation is unshift.
Spread Syntax (ES6)
To complete Koen's answer, in case you want to remove a dynamic variable using the spread syntax, you can do it like so:
const key = 'a';
const { [key]: foo, ...rest } = { a: 1, b: 2, c: 3 };
console.log(foo); // 1
console.log(rest); // { b: 2, c: 3 }
* foo will be a new variable with the value of a (which is 1).
Extended answer 😇
There are a few common ways to remove a property from an object. Each one has its own pros and cons (check this performance comparison):
Delete Operator
It is readable and short, however, it might not be the best choice if you are operating on a large number of objects as its performance is not optimized.
delete obj[key];
Reassignment
It is more than two times faster than delete, however the property is not deleted and can be iterated.
obj[key] = null;
obj[key] = false;
obj[key] = undefined;
Spread Operator
This ES6 operator allows us to return a brand new object, excluding any properties, without mutating the existing object. The downside is that it has the worse performance out of the above and is not suggested to be used when you need to remove many properties at a time.
{ [key]: val, ...rest } = obj;
Another alternative is to use the Underscore.js library.
Note that _.pick() and _.omit() both return a copy of the object and don't directly modify the original object. Assigning the result to the original object should do the trick (not shown).
Reference: link _.pick(object, *keys)
Return a copy of the object, filtered to only have values for the
whitelisted keys (or array of valid keys).
var myJSONObject =
{"ircEvent": "PRIVMSG", "method": "newURI", "regex": "^http://.*"};
_.pick(myJSONObject, "ircEvent", "method");
=> {"ircEvent": "PRIVMSG", "method": "newURI"};
Reference: link _.omit(object, *keys)
Return a copy of the object, filtered to omit the
blacklisted keys (or array of keys).
var myJSONObject =
{"ircEvent": "PRIVMSG", "method": "newURI", "regex": "^http://.*"};
_.omit(myJSONObject, "regex");
=> {"ircEvent": "PRIVMSG", "method": "newURI"};
For arrays, _.filter() and _.reject() can be used in a similar manner.
To clone an object without a property:
For example:
let object = { a: 1, b: 2, c: 3 };
And we need to delete a.
With an explicit prop key:
const { a, ...rest } = object;
object = rest;
With a variable prop key:
const propKey = 'a';
const { [propKey]: propValue, ...rest } = object;
object = rest;
A cool arrow function 😎:
const removeProperty = (propKey, { [propKey]: propValue, ...rest }) => rest;
object = removeProperty('a', object);
For multiple properties
const removeProperties = (object, ...keys) => (keys.length ? removeProperties(removeProperty(keys.pop(), object), ...keys) : object);
Usage
object = removeProperties(object, 'a', 'b') // result => { c: 3 }
Or
const propsToRemove = ['a', 'b']
object = removeProperties(object, ...propsToRemove) // result => { c: 3 }
The term you have used in your question title, Remove a property from a JavaScript object, can be interpreted in some different ways. The one is to remove it for whole the memory and the list of object keys or the other is just to remove it from your object. As it has been mentioned in some other answers, the delete keyword is the main part. Let's say you have your object like:
myJSONObject = {"ircEvent": "PRIVMSG", "method": "newURI", "regex": "^http://.*"};
If you do:
console.log(Object.keys(myJSONObject));
the result would be:
["ircEvent", "method", "regex"]
You can delete that specific key from your object keys like:
delete myJSONObject["regex"];
Then your objects key using Object.keys(myJSONObject) would be:
["ircEvent", "method"]
But the point is if you care about memory and you want to whole the object gets removed from the memory, it is recommended to set it to null before you delete the key:
myJSONObject["regex"] = null;
delete myJSONObject["regex"];
The other important point here is to be careful about your other references to the same object. For instance, if you create a variable like:
var regex = myJSONObject["regex"];
Or add it as a new pointer to another object like:
var myOtherObject = {};
myOtherObject["regex"] = myJSONObject["regex"];
Then even if you remove it from your object myJSONObject, that specific object won't get deleted from the memory, since the regex variable and myOtherObject["regex"] still have their values. Then how could we remove the object from the memory for sure?
The answer would be to delete all the references you have in your code, pointed to that very object and also not use var statements to create new references to that object. This last point regarding var statements, is one of the most crucial issues that we are usually faced with, because using var statements would prevent the created object from getting removed.
Which means in this case you won't be able to remove that object because you have created the regex variable via a var statement, and if you do:
delete regex; //False
The result would be false, which means that your delete statement haven't been executed as you expected. But if you had not created that variable before, and you only had myOtherObject["regex"] as your last existing reference, you could have done this just by removing it like:
myOtherObject["regex"] = null;
delete myOtherObject["regex"];
In other words, a JavaScript object is eligible to be killed as soon as there is no reference left in your code pointed to that object.
Update:
Thanks to #AgentME:
Setting a property to null before deleting it doesn't accomplish
anything (unless the object has been sealed by Object.seal and the
delete fails. That's not usually the case unless you specifically
try).
To get more information on Object.seal: Object.seal()
ECMAScript 2015 (or ES6) came with built-in Reflect object. It is possible to delete object property by calling Reflect.deleteProperty() function with target object and property key as parameters:
Reflect.deleteProperty(myJSONObject, 'regex');
which is equivalent to:
delete myJSONObject['regex'];
But if the property of the object is not configurable it cannot be deleted neither with deleteProperty function nor delete operator:
let obj = Object.freeze({ prop: "value" });
let success = Reflect.deleteProperty(obj, "prop");
console.log(success); // false
console.log(obj.prop); // value
Object.freeze() makes all properties of object not configurable (besides other things). deleteProperty function (as well as delete operator) returns false when tries to delete any of it's properties. If property is configurable it returns true, even if property does not exist.
The difference between delete and deleteProperty is when using strict mode:
"use strict";
let obj = Object.freeze({ prop: "value" });
Reflect.deleteProperty(obj, "prop"); // false
delete obj["prop"];
// TypeError: property "prop" is non-configurable and can't be deleted
Suppose you have an object that looks like this:
var Hogwarts = {
staff : [
'Argus Filch',
'Filius Flitwick',
'Gilderoy Lockhart',
'Minerva McGonagall',
'Poppy Pomfrey',
...
],
students : [
'Hannah Abbott',
'Katie Bell',
'Susan Bones',
'Terry Boot',
'Lavender Brown',
...
]
};
Deleting an object property
If you want to use the entire staff array, the proper way to do this, would be to do this:
delete Hogwarts.staff;
Alternatively, you could also do this:
delete Hogwarts['staff'];
Similarly, removing the entire students array would be done by calling delete Hogwarts.students; or delete Hogwarts['students'];.
Deleting an array index
Now, if you want to remove a single staff member or student, the procedure is a bit different, because both properties are arrays themselves.
If you know the index of your staff member, you could simply do this:
Hogwarts.staff.splice(3, 1);
If you do not know the index, you'll also have to do an index search:
Hogwarts.staff.splice(Hogwarts.staff.indexOf('Minerva McGonnagall') - 1, 1);
Note
While you technically can use delete for an array, using it would result in getting incorrect results when calling for example Hogwarts.staff.length later on. In other words, delete would remove the element, but it wouldn't update the value of length property. Using delete would also mess up your indexing.
So, when deleting values from an object, always first consider whether you're dealing with object properties or whether you're dealing with array values, and choose the appropriate strategy based on that.
If you want to experiment with this, you can use this Fiddle as a starting point.
I personally use Underscore.js or Lodash for object and array manipulation:
myObject = _.omit(myObject, 'regex');
Using delete method is the best way to do that, as per MDN description, the delete operator removes a property from an object. So you can simply write:
delete myObject.regex;
// OR
delete myObject['regex'];
The delete operator removes a given property from an object. On
successful deletion, it will return true, else false will be returned.
However, it is important to consider the following scenarios:
If the property which you are trying to delete does not exist, delete
will not have any effect and will return true
If a property with the same name exists on the object's prototype
chain, then, after deletion, the object will use the property from the
prototype chain (in other words, delete only has an effect on own
properties).
Any property declared with var cannot be deleted from the global scope
or from a function's scope.
As such, delete cannot delete any functions in the global scope (whether this is part of a function definition or a function (expression).
Functions which are part of an object (apart from the
global scope) can be deleted with delete.
Any property declared with let or const cannot be deleted from the scope within which they were defined. Non-configurable properties cannot be removed. This includes properties of built-in objects like Math, Array, Object and properties that are created as non-configurable with methods like Object.defineProperty().
The following snippet gives another simple example:
var Employee = {
age: 28,
name: 'Alireza',
designation: 'developer'
}
console.log(delete Employee.name); // returns true
console.log(delete Employee.age); // returns true
// When trying to delete a property that does
// not exist, true is returned
console.log(delete Employee.salary); // returns true
For more info about and seeing more examples visit the link below:
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/delete
Another solution, using Array#reduce.
var myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*"
};
myObject = Object.keys(myObject).reduce(function(obj, key) {
if (key != "regex") { //key you want to remove
obj[key] = myObject[key];
}
return obj;
}, {});
console.log(myObject);
However, it will mutate the original object. If you want to create a new object without the specified key, just assign the reduce function to a new variable, e.g.:
(ES6)
const myObject = {
ircEvent: 'PRIVMSG',
method: 'newURI',
regex: '^http://.*',
};
const myNewObject = Object.keys(myObject).reduce((obj, key) => {
key !== 'regex' ? obj[key] = myObject[key] : null;
return obj;
}, {});
console.log(myNewObject);
There are a lot of good answers here but I just want to chime in that when using delete to remove a property in JavaScript, it is often wise to first check if that property exists to prevent errors.
E.g
var obj = {"property":"value", "property2":"value"};
if (obj && obj.hasOwnProperty("property2")) {
delete obj.property2;
} else {
//error handling
}
Due to the dynamic nature of JavaScript there are often cases where you simply don't know if the property exists or not. Checking if obj exists before the && also makes sure you don't throw an error due to calling the hasOwnProperty() function on an undefined object.
Sorry if this didn't add to your specific use case but I believe this to be a good design to adapt when managing objects and their properties.
This post is very old and I find it very helpful so I decided to share the unset function I wrote in case someone else see this post and think why it's not so simple as it in PHP unset function.
The reason for writing this new unset function, is to keep the index of all other variables in this hash_map. Look at the following example, and see how the index of "test2" did not change after removing a value from the hash_map.
function unset(unsetKey, unsetArr, resort) {
var tempArr = unsetArr;
var unsetArr = {};
delete tempArr[unsetKey];
if (resort) {
j = -1;
}
for (i in tempArr) {
if (typeof(tempArr[i]) !== 'undefined') {
if (resort) {
j++;
} else {
j = i;
}
unsetArr[j] = tempArr[i];
}
}
return unsetArr;
}
var unsetArr = ['test', 'deletedString', 'test2'];
console.log(unset('1', unsetArr, true)); // output Object {0: "test", 1: "test2"}
console.log(unset('1', unsetArr, false)); // output Object {0: "test", 2: "test2"}
There are a couple of ways to remove properties from an object:
1) Remove using a dot property accessor (mutable)
const myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*",
};
delete myObject.regex;
console.log(myObject);
2. Remove using square brackets property accessor (mutable)
const myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*",
};
delete myObject['regex'];
console.log(myObject);
// or
const name = 'ircEvent';
delete myObject[name];
console.log(myObject);
3) Alternative option but without altering the original object, is using object destructuring and rest syntax (immutable)
const myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*",
};
const { regex, ...myObjectRest} = myObject;
console.log(myObjectRest);
Using ramda#dissoc you will get a new object without the attribute regex:
const newObject = R.dissoc('regex', myObject);
// newObject !== myObject
You can also use other functions to achieve the same effect - omit, pick, ...
Try the following method. Assign the Object property value to undefined. Then stringify the object and parse.
var myObject = {"ircEvent": "PRIVMSG", "method": "newURI", "regex": "^http://.*"};
myObject.regex = undefined;
myObject = JSON.parse(JSON.stringify(myObject));
console.log(myObject);
Using Lodash
import omit from 'lodash/omit';
const prevObject = {test: false, test2: true};
// Removes test2 key from previous object
const nextObject = omit(prevObject, 'test2');
Using Ramda
R.omit(['a', 'd'], {a: 1, b: 2, c: 3, d: 4}); //=> {b: 2, c: 3}
If you want to delete a property deeply nested in the object then you can use the following recursive function with path to the property as the second argument:
var deepObjectRemove = function(obj, path_to_key){
if(path_to_key.length === 1){
delete obj[path_to_key[0]];
return true;
}else{
if(obj[path_to_key[0]])
return deepObjectRemove(obj[path_to_key[0]], path_to_key.slice(1));
else
return false;
}
};
Example:
var a = {
level1:{
level2:{
level3: {
level4: "yolo"
}
}
}
};
deepObjectRemove(a, ["level1", "level2", "level3"]);
console.log(a);
//Prints {level1: {level2: {}}}
Object.assign() & Object.keys() & Array.map()
const obj = {
"Filters":[
{
"FilterType":"between",
"Field":"BasicInformationRow.A0",
"MaxValue":"2017-10-01",
"MinValue":"2017-09-01",
"Value":"Filters value"
}
]
};
let new_obj1 = Object.assign({}, obj.Filters[0]);
let new_obj2 = Object.assign({}, obj.Filters[0]);
/*
// old version
let shaped_obj1 = Object.keys(new_obj1).map(
(key, index) => {
switch (key) {
case "MaxValue":
delete new_obj1["MaxValue"];
break;
case "MinValue":
delete new_obj1["MinValue"];
break;
}
return new_obj1;
}
)[0];
let shaped_obj2 = Object.keys(new_obj2).map(
(key, index) => {
if(key === "Value"){
delete new_obj2["Value"];
}
return new_obj2;
}
)[0];
*/
// new version!
let shaped_obj1 = Object.keys(new_obj1).forEach(
(key, index) => {
switch (key) {
case "MaxValue":
delete new_obj1["MaxValue"];
break;
case "MinValue":
delete new_obj1["MinValue"];
break;
default:
break;
}
}
);
let shaped_obj2 = Object.keys(new_obj2).forEach(
(key, index) => {
if(key === "Value"){
delete new_obj2["Value"];
}
}
);
Dan's assertion that 'delete' is very slow and the benchmark he posted were doubted. So I carried out the test myself in Chrome 59. It does seem that 'delete' is about 30 times slower:
var iterationsTotal = 10000000; // 10 million
var o;
var t1 = Date.now(),t2;
for (let i=0; i<iterationsTotal; i++) {
o = {a:1,b:2,c:3,d:4,e:5};
delete o.a; delete o.b; delete o.c; delete o.d; delete o.e;
}
console.log ((t2=Date.now())-t1); // 6135
for (let i=0; i<iterationsTotal; i++) {
o = {a:1,b:2,c:3,d:4,e:5};
o.a = o.b = o.c = o.d = o.e = undefined;
}
console.log (Date.now()-t2); // 205
Note that I purposely carried out more than one 'delete' operations in one loop cycle to minimize the effect caused by the other operations.
Property Removal in JavaScript
There are many different options presented on this page, not because most of the options are wrong—or because the answers are duplicates—but because the appropriate technique depends on the situation you're in and the goals of the tasks you and/or you team are trying to fulfill. To answer you question unequivocally, one needs to know:
The version of ECMAScript you're targeting
The range of object types you want to remove properties on and the type of property names you need to be able to omit (Strings only? Symbols? Weak references mapped from arbitrary objects? These have all been types of property pointers in JavaScript for years now)
The programming ethos/patterns you and your team use. Do you favor functional approaches and mutation is verboten on your team, or do you employ wild west mutative object-oriented techniques?
Are you looking to achieve this in pure JavaScript or are you willing & able to use a 3rd-party library?
Once those four queries have been answered, there are essentially four categories of "property removal" in JavaScript to chose from in order to meet your goals. They are:
Mutative object property deletion, unsafe
This category is for operating on object literals or object instances when you want to retain/continue to use the original reference and aren't using stateless functional principles in your code. An example piece of syntax in this category:
'use strict'
const iLikeMutatingStuffDontI = { myNameIs: 'KIDDDDD!', [Symbol.for('amICool')]: true }
delete iLikeMutatingStuffDontI[Symbol.for('amICool')] // true
Object.defineProperty({ myNameIs: 'KIDDDDD!', 'amICool', { value: true, configurable: false })
delete iLikeMutatingStuffDontI['amICool'] // throws
This category is the oldest, most straightforward & most widely supported category of property removal. It supports Symbol & array indexes in addition to strings and works in every version of JavaScript except for the very first release. However, it's mutative which violates some programming principles and has performance implications. It also can result in uncaught exceptions when used on non-configurable properties in strict mode.
Rest-based string property omission
This category is for operating on plain object or array instances in newer ECMAScript flavors when a non-mutative approach is desired and you don't need to account for Symbol keys:
const foo = { name: 'KIDDDDD!', [Symbol.for('isCool')]: true }
const { name, ...coolio } = foo // coolio doesn't have "name"
const { isCool, ...coolio2 } = foo // coolio2 has everything from `foo` because `isCool` doesn't account for Symbols :(
Mutative object property deletion, safe
This category is for operating on object literals or object instances when you want to retain/continue to use the original reference while guarding against exceptions being thrown on unconfigurable properties:
'use strict'
const iLikeMutatingStuffDontI = { myNameIs: 'KIDDDDD!', [Symbol.for('amICool')]: true }
Reflect.deleteProperty(iLikeMutatingStuffDontI, Symbol.for('amICool')) // true
Object.defineProperty({ myNameIs: 'KIDDDDD!', 'amICool', { value: true, configurable: false })
Reflect.deleteProperty(iLikeMutatingStuffDontI, 'amICool') // false
In addition, while mutating objects in-place isn't stateless, you can use the functional nature of Reflect.deleteProperty to do partial application and other functional techniques that aren't possible with delete statements.
Syntax-based string property omission
This category is for operating on plain object or array instances in newer ECMAScript flavors when a non-mutative approach is desired and you don't need to account for Symbol keys:
const foo = { name: 'KIDDDDD!', [Symbol.for('isCool')]: true }
const { name, ...coolio } = foo // coolio doesn't have "name"
const { isCool, ...coolio2 } = foo // coolio2 has everything from `foo` because `isCool` doesn't account for Symbols :(
Library-based property omission
This category is generally allows for greater functional flexibility, including accounting for Symbols & omitting more than one property in one statement:
const o = require("lodash.omit")
const foo = { [Symbol.for('a')]: 'abc', b: 'b', c: 'c' }
const bar = o(foo, 'a') // "'a' undefined"
const baz = o(foo, [ Symbol.for('a'), 'b' ]) // Symbol supported, more than one prop at a time, "Symbol.for('a') undefined"
Here's an ES6 way to remove the entry easily:
let myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*"
};
const removeItem = 'regex';
const { [removeItem]: remove, ...rest } = myObject;
console.log(remove); // "^http://.*"
console.log(rest); // Object { ircEvent: "PRIVMSG", method: "newURI" }
#johnstock, we can also use JavaScript's prototyping concept to add method to objects to delete any passed key available in calling object.
Above answers are appreciated.
var myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*"
};
// 1st and direct way
delete myObject.regex; // delete myObject["regex"]
console.log(myObject); // { ircEvent: 'PRIVMSG', method: 'newURI' }
// 2 way - by using the concept of JavaScript's prototyping concept
Object.prototype.removeFromObjectByKey = function(key) {
// If key exists, remove it and return true
if (this[key] !== undefined) {
delete this[key]
return true;
}
// Else return false
return false;
}
var isRemoved = myObject.removeFromObjectByKey('method')
console.log(myObject) // { ircEvent: 'PRIVMSG' }
// More examples
var obj = {
a: 45,
b: 56,
c: 67
}
console.log(obj) // { a: 45, b: 56, c: 67 }
// Remove key 'a' from obj
isRemoved = obj.removeFromObjectByKey('a')
console.log(isRemoved); //true
console.log(obj); // { b: 56, c: 67 }
// Remove key 'd' from obj which doesn't exist
var isRemoved = obj.removeFromObjectByKey('d')
console.log(isRemoved); // false
console.log(obj); // { b: 56, c: 67 }
You can use a filter like below
var myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*"
};
// Way 1
let filter1 = {}
Object.keys({...myObject}).filter(d => {
if(d !== 'regex'){
filter1[d] = myObject[d];
}
})
console.log(filter1)
// Way 2
let filter2 = Object.fromEntries(Object.entries({...myObject}).filter(d =>
d[0] !== 'regex'
))
console.log(filter2)
I have used Lodash "unset" to make it happen for a nested object also... only this needs to write small logic to get the path of the property key which is expected by the omit method.
Method which returns the property path as an array
var a = {"bool":{"must":[{"range":{"price_index.final_price":{"gt":"450", "lt":"500"}}}, {"bool":{"should":[{"term":{"color_value.keyword":"Black"}}]}}]}};
function getPathOfKey(object,key,currentPath, t){
var currentPath = currentPath || [];
for(var i in object){
if(i == key){
t = currentPath;
}
else if(typeof object[i] == "object"){
currentPath.push(i)
return getPathOfKey(object[i], key,currentPath)
}
}
t.push(key);
return t;
}
document.getElementById("output").innerHTML =JSON.stringify(getPathOfKey(a,"price_index.final_price"))
<div id="output">
</div>
Then just using Lodash unset method remove property from object.
var unset = require('lodash.unset');
unset(a, getPathOfKey(a, "price_index.final_price"));
let myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*"
};
obj = Object.fromEntries(
Object.entries(myObject).filter(function (m){
return m[0] != "regex"/*or whatever key to delete*/
}
))
console.log(obj)
You can also just treat the object like a2d array using Object.entries, and use splice to remove an element as you would in a normal array, or simply filter through the object, as one would an array, and assign the reconstructed object back to the original variable
If you don't want to modify the original object.
Remove a property without mutating the object
If mutability is a concern, you can create a completely new object by copying all the properties from the old, except the one you want to remove.
let myObject = {
"ircEvent": "PRIVMSG",
"method": "newURI",
"regex": "^http://.*"
};
let prop = 'regex';
const updatedObject = Object.keys(myObject).reduce((object, key) => {
if (key !== prop) {
object[key] = myObject[key]
}
return object
}, {})
console.log(updatedObject);