Variable Declaration

let and const are two relatively new concepts for variable declarations in JavaScript. As we mentioned earlier, let is similar to var in some respects, but allows users to avoid some of the common “gotchas” that users run into in JavaScript.

const is an augmentation of let in that it prevents re-assignment to a variable.

With TypeScript being an extension of JavaScript, the language naturally supports let and const. Here we’ll elaborate more on these new declarations and why they’re preferable to var.

If you’ve used JavaScript offhandedly, the next section might be a good way to refresh your memory. If you’re intimately familiar with all the quirks of var declarations in JavaScript, you might find it easier to skip ahead.

var declarations

Declaring a variable in JavaScript has always traditionally been done with the var keyword.

ts
var a = 10;

As you might’ve figured out, we just declared a variable named a with the value 10.

We can also declare a variable inside of a function:

ts
function f() {
var message = "Hello, world!";
return message;
}

and we can also access those same variables within other functions:

ts
function f() {
var a = 10;
return function g() {
var b = a + 1;
return b;
};
}
var g = f();
g(); // returns '11'

In this above example, g captured the variable a declared in f. At any point that g gets called, the value of a will be tied to the value of a in f. Even if g is called once f is done running, it will be able to access and modify a.

ts
function f() {
var a = 1;
a = 2;
var b = g();
a = 3;
return b;
function g() {
return a;
}
}
f(); // returns '2'

Scoping rules

var declarations have some odd scoping rules for those used to other languages. Take the following example:

ts
function f(shouldInitialize: boolean) {
if (shouldInitialize) {
var x = 10;
}
return x;
}
f(true); // returns '10'
f(false); // returns 'undefined'

Some readers might do a double-take at this example. The variable x was declared within the if block, and yet we were able to access it from outside that block. That’s because var declarations are accessible anywhere within their containing function, module, namespace, or global scope - all which we’ll go over later on - regardless of the containing block. Some people call this var-scoping or function-scoping. Parameters are also function scoped.

These scoping rules can cause several types of mistakes. One problem they exacerbate is the fact that it is not an error to declare the same variable multiple times:

ts
function sumMatrix(matrix: number[][]) {
var sum = 0;
for (var i = 0; i < matrix.length; i++) {
var currentRow = matrix[i];
for (var i = 0; i < currentRow.length; i++) {
sum += currentRow[i];
}
}
return sum;
}

Maybe it was easy to spot out for some experienced JavaScript developers, but the inner for-loop will accidentally overwrite the variable i because i refers to the same function-scoped variable. As experienced developers know by now, similar sorts of bugs slip through code reviews and can be an endless source of frustration.

Variable capturing quirks

Take a quick second to guess what the output of the following snippet is:

ts
for (var i = 0; i < 10; i++) {
setTimeout(function () {
console.log(i);
}, 100 * i);
}

For those unfamiliar, setTimeout will try to execute a function after a certain number of milliseconds (though waiting for anything else to stop running).

Ready? Take a look:

10
10
10
10
10
10
10
10
10
10

Many JavaScript developers are intimately familiar with this behavior, but if you’re surprised, you’re certainly not alone. Most people expect the output to be

0
1
2
3
4
5
6
7
8
9

Remember what we mentioned earlier about variable capturing? Every function expression we pass to setTimeout actually refers to the same i from the same scope.

Let’s take a minute to consider what that means. setTimeout will run a function after some number of milliseconds, but only after the for loop has stopped executing; By the time the for loop has stopped executing, the value of i is 10. So each time the given function gets called, it will print out 10!

A common work around is to use an IIFE - an Immediately Invoked Function Expression - to capture i at each iteration:

ts
for (var i = 0; i < 10; i++) {
// capture the current state of 'i'
// by invoking a function with its current value
(function (i) {
setTimeout(function () {
console.log(i);
}, 100 * i);
})(i);
}

This odd-looking pattern is actually pretty common. The i in the parameter list actually shadows the i declared in the for loop, but since we named them the same, we didn’t have to modify the loop body too much.

let declarations

By now you’ve figured out that var has some problems, which is precisely why let statements were introduced. Apart from the keyword used, let statements are written the same way var statements are.

ts
let hello = "Hello!";

The key difference is not in the syntax, but in the semantics, which we’ll now dive into.

Block-scoping

When a variable is declared using let, it uses what some call lexical-scoping or block-scoping. Unlike variables declared with var whose scopes leak out to their containing function, block-scoped variables are not visible outside of their nearest containing block or for-loop.

ts
function f(input: boolean) {
let a = 100;
if (input) {
// Still okay to reference 'a'
let b = a + 1;
return b;
}
// Error: 'b' doesn't exist here
return b;
}

Here, we have two local variables a and b. a’s scope is limited to the body of f while b’s scope is limited to the containing if statement’s block.

Variables declared in a catch clause also have similar scoping rules.

ts
try {
throw "oh no!";
} catch (e) {
console.log("Oh well.");
}
// Error: 'e' doesn't exist here
console.log(e);

Another property of block-scoped variables is that they can’t be read or written to before they’re actually declared. While these variables are “present” throughout their scope, all points up until their declaration are part of their temporal dead zone. This is just a sophisticated way of saying you can’t access them before the let statement, and luckily TypeScript will let you know that.

ts
a++; // illegal to use 'a' before it's declared;
let a;

Something to note is that you can still capture a block-scoped variable before it’s declared. The only catch is that it’s illegal to call that function before the declaration. If targeting ES2015, a modern runtime will throw an error; however, right now TypeScript is permissive and won’t report this as an error.

ts
function foo() {
// okay to capture 'a'
return a;
}
// illegal call 'foo' before 'a' is declared
// runtimes should throw an error here
foo();
let a;

For more information on temporal dead zones, see relevant content on the Mozilla Developer Network.

Re-declarations and Shadowing

With var declarations, we mentioned that it didn’t matter how many times you declared your variables; you just got one.

ts
function f(x) {
var x;
var x;
if (true) {
var x;
}
}

In the above example, all declarations of x actually refer to the same x, and this is perfectly valid. This often ends up being a source of bugs. Thankfully, let declarations are not as forgiving.

ts
let x = 10;
let x = 20; // error: can't re-declare 'x' in the same scope

The variables don’t necessarily need to both be block-scoped for TypeScript to tell us that there’s a problem.

ts
function f(x) {
let x = 100; // error: interferes with parameter declaration
}
function g() {
let x = 100;
var x = 100; // error: can't have both declarations of 'x'
}

That’s not to say that a block-scoped variable can never be declared with a function-scoped variable. The block-scoped variable just needs to be declared within a distinctly different block.

ts
function f(condition, x) {
if (condition) {
let x = 100;
return x;
}
return x;
}
f(false, 0); // returns '0'
f(true, 0); // returns '100'

The act of introducing a new name in a more nested scope is called shadowing. It is a bit of a double-edged sword in that it can introduce certain bugs on its own in the event of accidental shadowing, while also preventing certain bugs. For instance, imagine we had written our earlier sumMatrix function using let variables.

ts
function sumMatrix(matrix: number[][]) {
let sum = 0;
for (let i = 0; i < matrix.length; i++) {
var currentRow = matrix[i];
for (let i = 0; i < currentRow.length; i++) {
sum += currentRow[i];
}
}
return sum;
}

This version of the loop will actually perform the summation correctly because the inner loop’s i shadows i from the outer loop.

Shadowing should usually be avoided in the interest of writing clearer code. While there are some scenarios where it may be fitting to take advantage of it, you should use your best judgement.

Block-scoped variable capturing

When we first touched on the idea of variable capturing with var declaration, we briefly went into how variables act once captured. To give a better intuition of this, each time a scope is run, it creates an “environment” of variables. That environment and its captured variables can exist even after everything within its scope has finished executing.

ts
function theCityThatAlwaysSleeps() {
let getCity;
if (true) {
let city = "Seattle";
getCity = function () {
return city;
};
}
return getCity();
}

Because we’ve captured city from within its environment, we’re still able to access it despite the fact that the if block finished executing.

Recall that with our earlier setTimeout example, we ended up needing to use an IIFE to capture the state of a variable for every iteration of the for loop. In effect, what we were doing was creating a new variable environment for our captured variables. That was a bit of a pain, but luckily, you’ll never have to do that again in TypeScript.

let declarations have drastically different behavior when declared as part of a loop. Rather than just introducing a new environment to the loop itself, these declarations sort of create a new scope per iteration. Since this is what we were doing anyway with our IIFE, we can change our old setTimeout example to just use a let declaration.

ts
for (let i = 0; i < 10; i++) {
setTimeout(function () {
console.log(i);
}, 100 * i);
}

and as expected, this will print out

0
1
2
3
4
5
6
7
8
9

const declarations

const declarations are another way of declaring variables.

ts
const numLivesForCat = 9;

They are like let declarations but, as their name implies, their value cannot be changed once they are bound. In other words, they have the same scoping rules as let, but you can’t re-assign to them.

This should not be confused with the idea that the values they refer to are immutable.

ts
const numLivesForCat = 9;
const kitty = {
name: "Aurora",
numLives: numLivesForCat,
};
// Error
kitty = {
name: "Danielle",
numLives: numLivesForCat,
};
// all "okay"
kitty.name = "Rory";
kitty.name = "Kitty";
kitty.name = "Cat";
kitty.numLives--;

Unless you take specific measures to avoid it, the internal state of a const variable is still modifiable. Fortunately, TypeScript allows you to specify that members of an object are readonly. The chapter on Interfaces has the details.

let vs. const

Given that we have two types of declarations with similar scoping semantics, it’s natural to find ourselves asking which one to use. Like most broad questions, the answer is: it depends.

Applying the principle of least privilege, all declarations other than those you plan to modify should use const. The rationale is that if a variable didn’t need to get written to, others working on the same codebase shouldn’t automatically be able to write to the object, and will need to consider whether they really need to reassign to the variable. Using const also makes code more predictable when reasoning about flow of data.

Use your best judgement, and if applicable, consult the matter with the rest of your team.

The majority of this handbook uses let declarations.

Destructuring

Another ECMAScript 2015 feature that TypeScript has is destructuring. For a complete reference, see the article on the Mozilla Developer Network. In this section, we’ll give a short overview.

Array destructuring

The simplest form of destructuring is array destructuring assignment:

ts
let input = [1, 2];
let [first, second] = input;
console.log(first); // outputs 1
console.log(second); // outputs 2

This creates two new variables named first and second. This is equivalent to using indexing, but is much more convenient:

ts
first = input[0];
second = input[1];

Destructuring works with already-declared variables as well:

ts
// swap variables
[first, second] = [second, first];

And with parameters to a function:

ts
function f([first, second]: [number, number]) {
console.log(first);
console.log(second);
}
f([1, 2]);

You can create a variable for the remaining items in a list using the syntax ...:

ts
let [first, ...rest] = [1, 2, 3, 4];
console.log(first); // outputs 1
console.log(rest); // outputs [ 2, 3, 4 ]

Of course, since this is JavaScript, you can just ignore trailing elements you don’t care about:

ts
let [first] = [1, 2, 3, 4];
console.log(first); // outputs 1

Or other elements:

ts
let [, second, , fourth] = [1, 2, 3, 4];
console.log(second); // outputs 2
console.log(fourth); // outputs 4

Tuple destructuring

Tuples may be destructured like arrays; the destructuring variables get the types of the corresponding tuple elements:

ts
let tuple: [number, string, boolean] = [7, "hello", true];
let [a, b, c] = tuple; // a: number, b: string, c: boolean

It’s an error to destructure a tuple beyond the range of its elements:

ts
let [a, b, c, d] = tuple; // Error, no element at index 3

As with arrays, you can destructure the rest of the tuple with ..., to get a shorter tuple:

ts
let [a, ...bc] = tuple; // bc: [string, boolean]
let [a, b, c, ...d] = tuple; // d: [], the empty tuple

Or ignore trailing elements, or other elements:

ts
let [a] = tuple; // a: number
let [, b] = tuple; // b: string

Object destructuring

You can also destructure objects:

ts
let o = {
a: "foo",
b: 12,
c: "bar",
};
let { a, b } = o;

This creates new variables a and b from o.a and o.b. Notice that you can skip c if you don’t need it.

Like array destructuring, you can have assignment without declaration:

ts
({ a, b } = { a: "baz", b: 101 });

Notice that we had to surround this statement with parentheses. JavaScript normally parses a { as the start of block.

You can create a variable for the remaining items in an object using the syntax ...:

ts
let { a, ...passthrough } = o;
let total = passthrough.b + passthrough.c.length;

Property renaming

You can also give different names to properties:

ts
let { a: newName1, b: newName2 } = o;

Here the syntax starts to get confusing. You can read a: newName1 as ”a as newName1”. The direction is left-to-right, as if you had written:

ts
let newName1 = o.a;
let newName2 = o.b;

Confusingly, the colon here does not indicate the type. The type, if you specify it, still needs to be written after the entire destructuring:

ts
let { a: newName1, b: newName2 }: { a: string; b: number } = o;

Default values

Default values let you specify a default value in case a property is undefined:

ts
function keepWholeObject(wholeObject: { a: string; b?: number }) {
let { a, b = 1001 } = wholeObject;
}

In this example the b? indicates that b is optional, so it may be undefined. keepWholeObject now has a variable for wholeObject as well as the properties a and b, even if b is undefined.

Function declarations

Destructuring also works in function declarations. For simple cases this is straightforward:

ts
type C = { a: string; b?: number };
function f({ a, b }: C): void {
// ...
}

But specifying defaults is more common for parameters, and getting defaults right with destructuring can be tricky. First of all, you need to remember to put the pattern before the default value.

ts
function f({ a = "", b = 0 } = {}): void {
// ...
}
f();

The snippet above is an example of type inference, explained earlier in the handbook.

Then, you need to remember to give a default for optional properties on the destructured property instead of the main initializer. Remember that C was defined with b optional:

ts
function f({ a, b = 0 } = { a: "" }): void {
// ...
}
f({ a: "yes" }); // ok, default b = 0
f(); // ok, default to { a: "" }, which then defaults b = 0
f({}); // error, 'a' is required if you supply an argument

Use destructuring with care. As the previous example demonstrates, anything but the simplest destructuring expression is confusing. This is especially true with deeply nested destructuring, which gets really hard to understand even without piling on renaming, default values, and type annotations. Try to keep destructuring expressions small and simple. You can always write the assignments that destructuring would generate yourself.

Spread

The spread operator is the opposite of destructuring. It allows you to spread an array into another array, or an object into another object. For example:

ts
let first = [1, 2];
let second = [3, 4];
let bothPlus = [0, ...first, ...second, 5];

This gives bothPlus the value [0, 1, 2, 3, 4, 5]. Spreading creates a shallow copy of first and second. They are not changed by the spread.

You can also spread objects:

ts
let defaults = { food: "spicy", price: "$$", ambiance: "noisy" };
let search = { ...defaults, food: "rich" };

Now search is { food: "rich", price: "$$", ambiance: "noisy" }. Object spreading is more complex than array spreading. Like array spreading, it proceeds from left-to-right, but the result is still an object. This means that properties that come later in the spread object overwrite properties that come earlier. So if we modify the previous example to spread at the end:

ts
let defaults = { food: "spicy", price: "$$", ambiance: "noisy" };
let search = { food: "rich", ...defaults };

Then the food property in defaults overwrites food: "rich", which is not what we want in this case.

Object spread also has a couple of other surprising limits. First, it only includes an objects’ own, enumerable properties. Basically, that means you lose methods when you spread instances of an object:

ts
class C {
p = 12;
m() {}
}
let c = new C();
let clone = { ...c };
clone.p; // ok
clone.m(); // error!

Second, the TypeScript compiler doesn’t allow spreads of type parameters from generic functions. That feature is expected in future versions of the language.

using declarations

using declarations are an upcoming feature for JavaScript that are part of the Stage 3 Explicit Resource Management proposal. A using declaration is much like a const declaration, except that it couples the lifetime of the value bound to the declaration with the scope of the variable.

When control exits the block containing a using declaration, the [Symbol.dispose]() method of the declared value is executed, which allows that value to perform cleanup:

ts
function f() {
using x = new C();
doSomethingWith(x);
} // `x[Symbol.dispose]()` is called

At runtime, this has an effect roughly equivalent to the following:

ts
function f() {
const x = new C();
try {
doSomethingWith(x);
}
finally {
x[Symbol.dispose]();
}
}

using declarations are extremely useful for avoiding memory leaks when working with JavaScript objects that hold on to native references like file handles

ts
{
using file = await openFile();
file.write(text);
doSomethingThatMayThrow();
} // `file` is disposed, even if an error is thrown

or scoped operations like tracing

ts
function f() {
using activity = new TraceActivity("f"); // traces entry into function
// ...
} // traces exit of function

Unlike var, let, and const, using declarations do not support destructuring.

null and undefined

It’s important to note that the value can be null or undefined, in which case nothing is disposed at the end of the block:

ts
{
using x = b ? new C() : null;
// ...
}

which is roughly equivalent to:

ts
{
const x = b ? new C() : null;
try {
// ...
}
finally {
x?.[Symbol.dispose]();
}
}

This allows you to conditionally acquire resources when declaring a using declaration without the need for complex branching or repetition.

Defining a disposable resource

You can indicate the classes or objects you produce are disposable by implementing the Disposable interface:

ts
// from the default lib:
interface Disposable {
[Symbol.dispose](): void;
}
// usage:
class TraceActivity implements Disposable {
readonly name: string;
constructor(name: string) {
this.name = name;
console.log(`Entering: ${name}`);
}
[Symbol.dispose](): void {
console.log(`Exiting: ${name}`);
}
}
function f() {
using _activity = new TraceActivity("f");
console.log("Hello world!");
}
f();
// prints:
// Entering: f
// Hello world!
// Exiting: f

await using declarations

Some resources or operations may have cleanup that needs to be performed asynchronously. To accommodate this, the Explicit Resource Management proposal also introduces the await using declaration:

ts
async function f() {
await using x = new C();
} // `await x[Symbol.asyncDispose]()` is invoked

An await using declaration invokes, and awaits, its value’s [Symbol.asyncDispose]() method as control leaves the containing block. This allows for asynchronous cleanup, such as a database transaction performing a rollback or commit, or a file stream flushing any pending writes to storage before it is closed.

As with await, await using can only be used in an async function or method, or at the top level of a module.

Defining an asynchronously disposable resource

Just as using relies on objects that are Disposable, an await using relies on objects that are AsyncDisposable:

ts
// from the default lib:
interface AsyncDisposable {
[Symbol.asyncDispose]: PromiseLike<void>;
}
// usage:
class DatabaseTransaction implements AsyncDisposable {
public success = false;
private db: Database | undefined;
private constructor(db: Database) {
this.db = db;
}
static async create(db: Database) {
await db.execAsync("BEGIN TRANSACTION");
return new DatabaseTransaction(db);
}
async [Symbol.asyncDispose]() {
if (this.db) {
const db = this.db:
this.db = undefined;
if (this.success) {
await db.execAsync("COMMIT TRANSACTION");
}
else {
await db.execAsync("ROLLBACK TRANSACTION");
}
}
}
}
async function transfer(db: Database, account1: Account, account2: Account, amount: number) {
using tx = await DatabaseTransaction.create(db);
if (await debitAccount(db, account1, amount)) {
await creditAccount(db, account2, amount);
}
// if an exception is thrown before this line, the transaction will roll back
tx.success = true;
// now the transaction will commit
}

await using vs await

The await keyword that is part of the await using declaration only indicates that the disposal of the resource is await-ed. It does not await the value itself:

ts
{
await using x = getResourceSynchronously();
} // performs `await x[Symbol.asyncDispose]()`
{
await using y = await getResourceAsynchronously();
} // performs `await y[Symbol.asyncDispose]()`

await using and return

It’s important to note that there is a small caveat with this behavior if you are using an await using declaration in an async function that returns a Promise without first await-ing it:

ts
function g() {
return Promise.reject("error!");
}
async function f() {
await using x = new C();
return g(); // missing an `await`
}

Because the returned promise isn’t await-ed, it’s possible that the JavaScript runtime may report an unhandled rejection since execution pauses while await-ing the asynchronous disposal of x, without having subscribed to the returned promise. This is not a problem that is unique to await using, however, as this can also occur in an async function that uses try..finally:

ts
async function f() {
try {
return g(); // also reports an unhandled rejection
}
finally {
await somethingElse();
}
}

To avoid this situation, it is recommended that you await your return value if it may be a Promise:

ts
async function f() {
await using x = new C();
return await g();
}

using and await using in for and for..of statements

Both using and await using can be used in a for statement:

ts
for (using x = getReader(); !x.eof; x.next()) {
// ...
}

In this case, the lifetime of x is scoped to the entire for statement and is only disposed when control leaves the loop due to break, return, throw, or when the loop condition is false.

In addition to for statements, both declarations can also be used in for..of statements:

ts
function * g() {
yield createResource1();
yield createResource2();
}
for (using x of g()) {
// ...
}

Here, x is disposed at the end of each iteration of the loop, and is then reinitialized with the next value. This is especially useful when consuming resources produced one at a time by a generator.

using and await using in older runtimes

using and await using declarations can be used when targeting older ECMAScript editions as long as you are using a compatible polyfill for Symbol.dispose/Symbol.asyncDispose, such as the one provided by default in recent editions of NodeJS.

The TypeScript docs are an open source project. Help us improve these pages by sending a Pull Request

Contributors to this page:
DRDaniel Rosenwasser  (58)
OTOrta Therox  (20)
NSNathan Shively-Sanders  (9)
VRVimal Raghubir  (3)
BCBrett Cannon  (3)
24+

Last updated: Oct 09, 2024