TSConfig Reference #

Specifies an allowlist of files to include in the program. An error occurs if any of the files can’t be found.

{
  "compilerOptions": {},
  "files": [
    "core.ts",
    "sys.ts",
    "types.ts",
    "scanner.ts",
    "parser.ts",
    "utilities.ts",
    "binder.ts",
    "checker.ts",
    "tsc.ts"
  ]
}

This is useful when you only have a small number of files and don’t need to use a glob to reference many files. If you need that then use include.

The value of extends is a string which contains a path to another configuration file to inherit from. The path may use Node.js style resolution.

The configuration from the base file are loaded first, then overridden by those in the inheriting config file. All relative paths found in the configuration file will be resolved relative to the configuration file they originated in.

It’s worth noting that files, include, and exclude from the inheriting config file overwrite those from the base config file, and that circularity between configuration files is not allowed.

Currently, the only top-level property that is excluded from inheritance is references.

Example #

configs/base.json:

{
  "compilerOptions": {
    "noImplicitAny": true,
    "strictNullChecks": true
  }
}

tsconfig.json:

{
  "extends": "./configs/base",
  "files": ["main.ts", "supplemental.ts"]
}

tsconfig.nostrictnull.json:

{
  "extends": "./tsconfig",
  "compilerOptions": {
    "strictNullChecks": false
  }
}

Properties with relative paths found in the configuration file, which aren’t excluded from inheritance, will be resolved relative to the configuration file they originated in.

Specifies an array of filenames or patterns to include in the program. These filenames are resolved relative to the directory containing the tsconfig.json file.

{
  "include": ["src/**/*", "tests/**/*"]
}

Which would include:

.
├── scripts                ⨯
│   ├── lint.ts            ⨯
│   ├── update_deps.ts     ⨯
│   └── utils.ts           ⨯
├── src                    ✓
│   ├── client             ✓
│   │    ├── index.ts      ✓
│   │    └── utils.ts      ✓
│   ├── server             ✓
│   │    └── index.ts      ✓
├── tests                  ✓
│   ├── app.test.ts        ✓
│   ├── utils.ts           ✓
│   └── tests.d.ts         ✓
├── package.json
├── tsconfig.json
└── yarn.lock

include and exclude support wildcard characters to make glob patterns:

• * matches zero or more characters (excluding directory separators)
• ? matches any one character (excluding directory separators)
• **/ matches any directory nested to any level

If a glob pattern doesn’t include a file extension, then only files with supported extensions are included (e.g. .ts, .tsx, and .d.ts by default, with .js and .jsx if allowJs is set to true).

Specifies an array of filenames or patterns that should be skipped when resolving include.

Important: exclude only changes which files are included as a result of the include setting. A file specified by exclude can still become part of your codebase due to an import statement in your code, a types inclusion, a /// <reference directive, or being specified in the files list.

It is not a mechanism that prevents a file from being included in the codebase - it simply changes what the include setting finds.

Project references are a way to structure your TypeScript programs into smaller pieces. Using Project References can greatly improve build and editor interaction times, enforce logical separation between components, and organize your code in new and improved ways.

You can read more about how references works in the Project References section of the handbook

When:

• undefined (default) provide suggestions as warnings to editors
• true unreachable code is ignored
• false raises compiler errors about unreachable code

These warnings are only about code which is provably unreachable due to the use of JavaScript syntax, for example:

function fn(n: number) {
  if (n > 5) {
    return true;
  } else {
    return false;
  }
  return true;
}

With "allowUnreachableCode": false:

function fn(n: number) {
  if (n > 5) {
    return true;
  } else {
    return false;
  }
  return true;
}

This does not affect errors on the basis of code which appears to be unreachable due to type analysis.

When:

• undefined (default) provide suggestions as warnings to editors
• true unused labels are ignored
• false raises compiler errors about unused labels

Labels are very rare in JavaScript and typically indicate an attempt to write an object literal:

function verifyAge(age: number) {
  // Forgot 'return' statement
  if (age > 18) {
    verified: true;
  }
}

Ensures that your files are parsed in the ECMAScript strict mode, and emit “use strict” for each source file.

ECMAScript strict mode was introduced in ES5 and provides behavior tweaks to the runtime of the JavaScript engine to improve performance, and makes a set of errors throw instead of silently ignoring them.

With exactOptionalPropertyTypes enabled, TypeScript applies stricter rules around how it handles properties on type or interfaces which have a ? prefix.

For example, this interface declares that there is a property which can be one of two strings: ‘dark’ or ’light’ or it should not be in the object.

interface UserDefaults {
  // The absence of a value represents 'system'
  colorThemeOverride?: "dark" | "light";
}

Without this flag enabled, there are three values which you can set colorThemeOverride to be: “dark”, “light” and undefined.

Setting the value to undefined will allow most JavaScript runtime checks for the existence to fail, which is effectively falsy. However, this isn’t quite accurate; colorThemeOverride: undefined is not the same as colorThemeOverride not being defined. For example, "colorThemeOverride" in settings would have different behavior with undefined as the key compared to not being defined.

exactOptionalPropertyTypes makes TypeScript truly enforce the definition provided as an optional property:

const settings = getUserSettings();
settings.colorThemeOverride = "dark";
settings.colorThemeOverride = "light";
 
// But not:
settings.colorThemeOverride = undefined;

Report errors for fallthrough cases in switch statements. Ensures that any non-empty case inside a switch statement includes either break, return, or throw. This means you won’t accidentally ship a case fallthrough bug.

const a: number = 6;
 
switch (a) {
  case 0:
    console.log("even");
  case 1:
    console.log("odd");
    break;
}

In some cases where no type annotations are present, TypeScript will fall back to a type of any for a variable when it cannot infer the type.

This can cause some errors to be missed, for example:

function fn(s) {
  // No error?
  console.log(s.subtr(3));
}
fn(42);

Turning on noImplicitAny however TypeScript will issue an error whenever it would have inferred any:

function fn(s) {
  console.log(s.subtr(3));
}

When working with classes which use inheritance, it’s possible for a sub-class to get “out of sync” with the functions it overloads when they are renamed in the base class.

For example, imagine you are modeling a music album syncing system:

class Album {
  download() {
    // Default behavior
  }
}
 
class SharedAlbum extends Album {
  download() {
    // Override to get info from many sources
  }
}

Then when you add support for machine-learning generated playlists, you refactor the Album class to have a ‘setup’ function instead:

class Album {
  setup() {
    // Default behavior
  }
}
 
class MLAlbum extends Album {
  setup() {
    // Override to get info from algorithm
  }
}
 
class SharedAlbum extends Album {
  download() {
    // Override to get info from many sources
  }
}

In this case, TypeScript has provided no warning that download on SharedAlbum expected to override a function in the base class.

Using noImplicitOverride you can ensure that the sub-classes never go out of sync, by ensuring that functions which override include the keyword override.

The following example has noImplicitOverride enabled, and you can see the error received when override is missing:

class Album {
  setup() {}
}
 
class MLAlbum extends Album {
  override setup() {}
}
 
class SharedAlbum extends Album {
  setup() {}
}

When enabled, TypeScript will check all code paths in a function to ensure they return a value.

function lookupHeadphonesManufacturer(color: "blue" | "black"): string {
  if (color === "blue") {
    return "beats";
  } else {
    "bose";
  }
}

Raise error on ’this’ expressions with an implied ‘any’ type.

For example, the class below returns a function which tries to access this.width and this.height – but the context for this inside the function inside getAreaFunction is not the instance of the Rectangle.

class Rectangle {
  width: number;
  height: number;
 
  constructor(width: number, height: number) {
    this.width = width;
    this.height = height;
  }
 
  getAreaFunction() {
    return function () {
      return this.width * this.height;
    };
  }
}

This setting ensures consistency between accessing a field via the “dot” (obj.key) syntax, and “indexed” (obj["key"]) and the way which the property is declared in the type.

Without this flag, TypeScript will allow you to use the dot syntax to access fields which are not defined:

interface GameSettings {
  // Known up-front properties
  speed: "fast" | "medium" | "slow";
  quality: "high" | "low";
 
  // Assume anything unknown to the interface
  // is a string.
  [key: string]: string;
}
 
const settings = getSettings();
settings.speed;
settings.quality;
 
// Unknown key accessors are allowed on
// this object, and are `string`
settings.username;

Turning the flag on will raise an error because the unknown field uses dot syntax instead of indexed syntax.

const settings = getSettings();
settings.speed;
settings.quality;
 
// This would need to be settings["username"];
settings.username;

The goal of this flag is to signal intent in your calling syntax about how certain you are this property exists.

TypeScript has a way to describe objects which have unknown keys but known values on an object, via index signatures.

interface EnvironmentVars {
  NAME: string;
  OS: string;
 
  // Unknown properties are covered by this index signature.
  [propName: string]: string;
}
 
declare const env: EnvironmentVars;
 
// Declared as existing
const sysName = env.NAME;
const os = env.OS;
 
// Not declared, but because of the index
// signature, then it is considered a string
const nodeEnv = env.NODE_ENV;

Turning on noUncheckedIndexedAccess will add undefined to any un-declared field in the type.

declare const env: EnvironmentVars;
 
// Declared as existing
const sysName = env.NAME;
const os = env.OS;
 
// Not declared, but because of the index
// signature, then it is considered a string
const nodeEnv = env.NODE_ENV;

Report errors on unused local variables.

const createKeyboard = (modelID: number) => {
  const defaultModelID = 23;
  return { type: "keyboard", modelID };
};

Report errors on unused parameters in functions.

const createDefaultKeyboard = (modelID: number) => {
  const defaultModelID = 23;
  return { type: "keyboard", modelID: defaultModelID };
};

The strict flag enables a wide range of type checking behavior that results in stronger guarantees of program correctness. Turning this on is equivalent to enabling all of the strict mode family options, which are outlined below. You can then turn off individual strict mode family checks as needed.

Future versions of TypeScript may introduce additional stricter checking under this flag, so upgrades of TypeScript might result in new type errors in your program. When appropriate and possible, a corresponding flag will be added to disable that behavior.

When set, TypeScript will check that the built-in methods of functions call, bind, and apply are invoked with correct argument for the underlying function:

// With strictBindCallApply on
function fn(x: string) {
  return parseInt(x);
}
 
const n1 = fn.call(undefined, "10");
 
const n2 = fn.call(undefined, false);

Otherwise, these functions accept any arguments and will return any:

// With strictBindCallApply off
function fn(x: string) {
  return parseInt(x);
}
 
// Note: No error; return type is 'any'
const n = fn.call(undefined, false);

When enabled, this flag causes functions parameters to be checked more correctly.

Here’s a basic example with strictFunctionTypes off:

function fn(x: string) {
  console.log("Hello, " + x.toLowerCase());
}
 
type StringOrNumberFunc = (ns: string | number) => void;
 
// Unsafe assignment
let func: StringOrNumberFunc = fn;
// Unsafe call - will crash
func(10);

With strictFunctionTypes on, the error is correctly detected:

function fn(x: string) {
  console.log("Hello, " + x.toLowerCase());
}
 
type StringOrNumberFunc = (ns: string | number) => void;
 
// Unsafe assignment is prevented
let func: StringOrNumberFunc = fn;

During development of this feature, we discovered a large number of inherently unsafe class hierarchies, including some in the DOM. Because of this, the setting only applies to functions written in function syntax, not to those in method syntax:

type Methodish = {
  func(x: string | number): void;
};
 
function fn(x: string) {
  console.log("Hello, " + x.toLowerCase());
}
 
// Ultimately an unsafe assignment, but not detected
const m: Methodish = {
  func: fn,
};
m.func(10);

When strictNullChecks is false, null and undefined are effectively ignored by the language. This can lead to unexpected errors at runtime.

When strictNullChecks is true, null and undefined have their own distinct types and you’ll get a type error if you try to use them where a concrete value is expected.

For example with this TypeScript code, users.find has no guarantee that it will actually find a user, but you can write code as though it will:

declare const loggedInUsername: string;
 
const users = [
  { name: "Oby", age: 12 },
  { name: "Heera", age: 32 },
];
 
const loggedInUser = users.find((u) => u.name === loggedInUsername);
console.log(loggedInUser.age);

Setting strictNullChecks to true will raise an error that you have not made a guarantee that the loggedInUser exists before trying to use it.

declare const loggedInUsername: string;
 
const users = [
  { name: "Oby", age: 12 },
  { name: "Heera", age: 32 },
];
 
const loggedInUser = users.find((u) => u.name === loggedInUsername);
console.log(loggedInUser.age);

The second example failed because the array’s find function looks a bit like this simplification:

// When strictNullChecks: true
type Array = {
  find(predicate: (value: any, index: number) => boolean): S | undefined;
};

// When strictNullChecks: false the undefined is removed from the type system,
// allowing you to write code which assumes it always found a result
type Array = {
  find(predicate: (value: any, index: number) => boolean): S;
};

When set to true, TypeScript will raise an error when a class property was declared but not set in the constructor.

class UserAccount {
  name: string;
  accountType = "user";
 
  email: string;
  address: string | undefined;
 
  constructor(name: string) {
    this.name = name;
    // Note that this.email is not set
  }
}

In the above case:

• this.name is set specifically.
• this.accountType is set by default.
• this.email is not set and raises an error.
• this.address is declared as potentially undefined which means it does not have to be set.

In TypeScript 4.0, support was added to allow changing the type of the variable in a catch clause from any to unknown. Allowing for code like:

try {
  // ...
} catch (err) {
  // We have to verify err is an
  // error before using it as one.
  if (err instanceof Error) {
    console.log(err.message);
  }
}

This pattern ensures that error handling code becomes more comprehensive because you cannot guarantee that the object being thrown is a Error subclass ahead of time. With the flag useUnknownInCatchVariables enabled, then you do not need the additional syntax (: unknown) nor a linter rule to try enforce this behavior.

In TypeScript 5.0, when an import path ends in an extension that isn’t a known JavaScript or TypeScript file extension, the compiler will look for a declaration file for that path in the form of {file basename}.d.{extension}.ts. For example, if you are using a CSS loader in a bundler project, you might want to write (or generate) declaration files for those stylesheets:

/* app.css */
.cookie-banner {
  display: none;
}

// app.d.css.ts
declare const css: {
  cookieBanner: string;
};
export default css;

// App.tsx
import styles from "./app.css";

styles.cookieBanner; // string

By default, this import will raise an error to let you know that TypeScript doesn’t understand this file type and your runtime might not support importing it. But if you’ve configured your runtime or bundler to handle it, you can suppress the error with the new --allowArbitraryExtensions compiler option.

Note that historically, a similar effect has often been achievable by adding a declaration file named app.css.d.ts instead of app.d.css.ts - however, this just worked through Node’s require resolution rules for CommonJS. Strictly speaking, the former is interpreted as a declaration file for a JavaScript file named app.css.js. Because relative files imports need to include extensions in Node’s ESM support, TypeScript would error on our example in an ESM file under --moduleResolution node16 or nodenext.

For more information, read up the proposal for this feature and its corresponding pull request.

--allowImportingTsExtensions allows TypeScript files to import each other with a TypeScript-specific extension like .ts, .mts, or .tsx.

This flag is only allowed when --noEmit or --emitDeclarationOnly is enabled, since these import paths would not be resolvable at runtime in JavaScript output files. The expectation here is that your resolver (e.g. your bundler, a runtime, or some other tool) is going to make these imports between .ts files work.

When set to true, allowUmdGlobalAccess lets you access UMD exports as globals from inside module files. A module file is a file that has imports and/or exports. Without this flag, using an export from a UMD module requires an import declaration.

An example use case for this flag would be a web project where you know the particular library (like jQuery or Lodash) will always be available at runtime, but you can’t access it with an import.

Sets a base directory from which to resolve non-relative module names. For example, in the directory structure:

project
├── ex.ts
├── hello
│   └── world.ts
└── tsconfig.json

With "baseUrl": "./", TypeScript will look for files starting at the same folder as the tsconfig.json:

import { helloWorld } from "hello/world";

console.log(helloWorld);

This resolution has higher priority than lookups from node_modules.

This feature was designed for use in conjunction with AMD module loaders in the browser, and is not recommended in any other context. As of TypeScript 4.1, baseUrl is no longer required to be set when using paths.

--customConditions takes a list of additional conditions that should succeed when TypeScript resolves from an [exports] or ( https://nodejs.org/api/packages.html#exports) or imports field of a package.json. These conditions are added to whatever existing conditions a resolver will use by default.

For example, when this field is set in a tsconfig.json as so:

{
  "compilerOptions": {
    "target": "es2022",
    "moduleResolution": "bundler",
    "customConditions": ["my-condition"]
  }
}

Any time an exports or imports field is referenced in package.json, TypeScript will consider conditions called my-condition.

So when importing from a package with the following package.json

{
  // ...
  "exports": {
    ".": {
      "my-condition": "./foo.mjs",
      "node": "./bar.mjs",
      "import": "./baz.mjs",
      "require": "./biz.mjs"
    }
  }
}

TypeScript will try to look for files corresponding to foo.mjs.

This field is only valid under the node16, nodenext, and bundler options for --moduleResolution.

Sets the module system for the program. See the Modules reference page for more information. You very likely want "CommonJS" for node projects.

Changing module affects moduleResolution which also has a reference page.

Here’s some example output for this file:

// @filename: index.ts
import { valueOfPi } from "./constants";
 
export const twoPi = valueOfPi * 2;

CommonJS #

"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
exports.twoPi = void 0;
const constants_1 = require("./constants");
exports.twoPi = constants_1.valueOfPi * 2;
 

UMD #

(function (factory) {
    if (typeof module === "object" && typeof module.exports === "object") {
        var v = factory(require, exports);
        if (v !== undefined) module.exports = v;
    }
    else if (typeof define === "function" && define.amd) {
        define(["require", "exports", "./constants"], factory);
    }
})(function (require, exports) {
    "use strict";
    Object.defineProperty(exports, "__esModule", { value: true });
    exports.twoPi = void 0;
    const constants_1 = require("./constants");
    exports.twoPi = constants_1.valueOfPi * 2;
});
 

AMD #

define(["require", "exports", "./constants"], function (require, exports, constants_1) {
    "use strict";
    Object.defineProperty(exports, "__esModule", { value: true });
    exports.twoPi = void 0;
    exports.twoPi = constants_1.valueOfPi * 2;
});
 

System #

System.register(["./constants"], function (exports_1, context_1) {
    "use strict";
    var constants_1, twoPi;
    var __moduleName = context_1 && context_1.id;
    return {
        setters: [
            function (constants_1_1) {
                constants_1 = constants_1_1;
            }
        ],
        execute: function () {
            exports_1("twoPi", twoPi = constants_1.valueOfPi * 2);
        }
    };
});
 

ESNext #

import { valueOfPi } from "./constants";
export const twoPi = valueOfPi * 2;
 

ES2020 #

import { valueOfPi } from "./constants";
export const twoPi = valueOfPi * 2;
 

ES2015/ES6/ES2020/ES2022 #

import { valueOfPi } from "./constants";
export const twoPi = valueOfPi * 2;
 

In addition to the base functionality of ES2015/ES6, ES2020 adds support for dynamic imports, and import.meta while ES2022 further adds support for top level await.

node16/nodenext (nightly builds) #

Available from 4.7+, the node16 and nodenext modes integrate with Node’s native ECMAScript Module support. The emitted JavaScript uses either CommonJS or ES2020 output depending on the file extension and the value of the type setting in the nearest package.json. Module resolution also works differently. You can learn more in the handbook.

None #

"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
exports.twoPi = void 0;
const constants_1 = require("./constants");
exports.twoPi = constants_1.valueOfPi * 2;
 

Specify the module resolution strategy:

• 'node' for Node.js’ CommonJS implementation
• 'node16' or 'nodenext' for Node.js’ ECMAScript Module Support from TypeScript 4.7 onwards
• 'classic' used in TypeScript before the release of 1.6. You probably won’t need to use classic in modern code

There is a handbook reference page on Module Resolution

Provides a way to override the default list of file name suffixes to search when resolving a module.

{
  "compilerOptions": {
    "moduleSuffixes": [".ios", ".native", ""]
  }
}

Given the above configuration, an import like the following:

import * as foo from "./foo";

TypeScript will look for the relative files ./foo.ios.ts, ./foo.native.ts, and finally ./foo.ts.

Note the empty string "" in moduleSuffixes which is necessary for TypeScript to also look-up ./foo.ts.

This feature can be useful for React Native projects where each target platform can use a separate tsconfig.json with differing moduleSuffixes.

By default, TypeScript will examine the initial set of files for import and <reference directives and add these resolved files to your program.

If noResolve is set, this process doesn’t happen. However, import statements are still checked to see if they resolve to a valid module, so you’ll need to make sure this is satisfied by some other means.

A series of entries which re-map imports to lookup locations relative to the baseUrl if set, or to the tsconfig file itself otherwise. There is a larger coverage of paths in the handbook.

paths lets you declare how TypeScript should resolve an import in your require/imports.

{
  "compilerOptions": {
    "paths": {
      "jquery": ["./vendor/jquery/dist/jquery"]
    }
  }
}

This would allow you to be able to write import "jquery", and get all of the correct typing locally.

{
  "compilerOptions": {
    "paths": {
        "app/*": ["./src/app/*"],
        "config/*": ["./src/app/_config/*"],
        "environment/*": ["./src/environments/*"],
        "shared/*": ["./src/app/_shared/*"],
        "helpers/*": ["./src/helpers/*"],
        "tests/*": ["./src/tests/*"]
    },
}

In this case, you can tell the TypeScript file resolver to support a number of custom prefixes to find code.

Note that this feature does not change how import paths are emitted by tsc, so paths should only be used to inform TypeScript that another tool has this mapping and will use it at runtime or when bundling.

Allows importing modules with a ‘.json’ extension, which is a common practice in node projects. This includes generating a type for the import based on the static JSON shape.

TypeScript does not support resolving JSON files by default:

// @filename: settings.json
{
    "repo": "TypeScript",
    "dry": false,
    "debug": false
}
// @filename: index.ts
import settings from "./settings.json";
 
settings.debug === true;
settings.dry === 2;

Enabling the option allows importing JSON, and validating the types in that JSON file.

// @filename: settings.json
{
    "repo": "TypeScript",
    "dry": false,
    "debug": false
}
// @filename: index.ts
import settings from "./settings.json";
 
settings.debug === true;
settings.dry === 2;

--resolvePackageJsonExports forces TypeScript to consult the exports field of package.json files if it ever reads from a package in node_modules.

This option defaults to true under the node16, nodenext, and bundler options for --moduleResolution.

--resolvePackageJsonImports forces TypeScript to consult the imports field of package.json files when performing a lookup that starts with # from a file whose ancestor directory contains a package.json.

This option defaults to true under the node16, nodenext, and bundler options for --moduleResolution.

Default: The longest common path of all non-declaration input files. If composite is set, the default is instead the directory containing the tsconfig.json file.

When TypeScript compiles files, it keeps the same directory structure in the output directory as exists in the input directory.

For example, let’s say you have some input files:

MyProj
├── tsconfig.json
├── core
│   ├── a.ts
│   ├── b.ts
│   ├── sub
│   │   ├── c.ts
├── types.d.ts

The inferred value for rootDir is the longest common path of all non-declaration input files, which in this case is core/.

If your outDir was dist, TypeScript would write this tree:

MyProj
├── dist
│   ├── a.js
│   ├── b.js
│   ├── sub
│   │   ├── c.js

However, you may have intended for core to be part of the output directory structure. By setting rootDir: "." in tsconfig.json, TypeScript would write this tree:

MyProj
├── dist
│   ├── core
│   │   ├── a.js
│   │   ├── b.js
│   │   ├── sub
│   │   │   ├── c.js

Importantly, rootDir does not affect which files become part of the compilation. It has no interaction with the include, exclude, or files tsconfig.json settings.

Note that TypeScript will never write an output file to a directory outside of outDir, and will never skip emitting a file. For this reason, rootDir also enforces that all files which need to be emitted are underneath the rootDir path.

For example, let’s say you had this tree:

MyProj
├── tsconfig.json
├── core
│   ├── a.ts
│   ├── b.ts
├── helpers.ts

It would be an error to specify rootDir as core and include as * because it creates a file (helpers.ts) that would need to be emitted outside the outDir (i.e. ../helpers.js).

Using rootDirs, you can inform the compiler that there are many “virtual” directories acting as a single root. This allows the compiler to resolve relative module imports within these “virtual” directories, as if they were merged in to one directory.

For example:

 src
 └── views
     └── view1.ts (can import "./template1", "./view2`)
     └── view2.ts (can import "./template1", "./view1`)

 generated
 └── templates
         └── views
             └── template1.ts (can import "./view1", "./view2")

{
  "compilerOptions": {
    "rootDirs": ["src/views", "generated/templates/views"]
  }
}

This does not affect how TypeScript emits JavaScript, it only emulates the assumption that they will be able to work via those relative paths at runtime.

rootDirs can be used to provide a separate “type layer” to files that are not TypeScript or JavaScript by providing a home for generated .d.ts files in another folder. This technique is useful for bundled applications where you use import of files that aren’t necessarily code:

 src
 └── index.ts
 └── css
     └── main.css
     └── navigation.css

 generated
 └── css
     └── main.css.d.ts
     └── navigation.css.d.ts

{
  "compilerOptions": {
    "rootDirs": ["src", "generated"]
  }
}

This technique lets you generate types ahead of time for the non-code source files. Imports then work naturally based off the source file’s location. For example ./src/index.ts can import the file ./src/css/main.css and TypeScript will be aware of the bundler’s behavior for that filetype via the corresponding generated declaration file.

// @filename: index.ts
import { appClass } from "./main.css";

By default all visible “@types” packages are included in your compilation. Packages in node_modules/@types of any enclosing folder are considered visible. For example, that means packages within ./node_modules/@types/, ../node_modules/@types/, ../../node_modules/@types/, and so on.

If typeRoots is specified, only packages under typeRoots will be included. For example:

{
  "compilerOptions": {
    "typeRoots": ["./typings", "./vendor/types"]
  }
}

This config file will include all packages under ./typings and ./vendor/types, and no packages from ./node_modules/@types. All paths are relative to the tsconfig.json.

By default all visible “@types” packages are included in your compilation. Packages in node_modules/@types of any enclosing folder are considered visible. For example, that means packages within ./node_modules/@types/, ../node_modules/@types/, ../../node_modules/@types/, and so on.

If types is specified, only packages listed will be included in the global scope. For instance:

{
  "compilerOptions": {
    "types": ["node", "jest", "express"]
  }
}

This tsconfig.json file will only include ./node_modules/@types/node, ./node_modules/@types/jest and ./node_modules/@types/express. Other packages under node_modules/@types/* will not be included.

What does this affect? #

This option does not affect how @types/* are included in your application code, for example if you had the above compilerOptions example with code like:

import * as moment from "moment";

moment().format("MMMM Do YYYY, h:mm:ss a");

The moment import would be fully typed.

When you have this option set, by not including a module in the types array it:

• Will not add globals to your project (e.g process in node, or expect in Jest)
• Will not have exports appear as auto-import recommendations

This feature differs from typeRoots in that it is about specifying only the exact types you want included, whereas typeRoots supports saying you want particular folders.

Generate .d.ts files for every TypeScript or JavaScript file inside your project. These .d.ts files are type definition files which describe the external API of your module. With .d.ts files, tools like TypeScript can provide intellisense and accurate types for un-typed code.

When declaration is set to true, running the compiler with this TypeScript code:

export let helloWorld = "hi";

Will generate an index.js file like this:

export let helloWorld = "hi";
 

With a corresponding helloWorld.d.ts:

export declare let helloWorld: string;
 

When working with .d.ts files for JavaScript files you may want to use emitDeclarationOnly or use outDir to ensure that the JavaScript files are not overwritten.

Offers a way to configure the root directory for where declaration files are emitted.

example
├── index.ts
├── package.json
└── tsconfig.json

with this tsconfig.json:

{
  "compilerOptions": {
    "declaration": true,
    "declarationDir": "./types"
  }
}

Would place the d.ts for the index.ts in a types folder:

example
├── index.js
├── index.ts
├── package.json
├── tsconfig.json
└── types
    └── index.d.ts

Generates a source map for .d.ts files which map back to the original .ts source file. This will allow editors such as VS Code to go to the original .ts file when using features like Go to Definition.

You should strongly consider turning this on if you’re using project references.

Downleveling is TypeScript’s term for transpiling to an older version of JavaScript. This flag is to enable support for a more accurate implementation of how modern JavaScript iterates through new concepts in older JavaScript runtimes.

ECMAScript 6 added several new iteration primitives: the for / of loop (for (el of arr)), Array spread ([a, ...b]), argument spread (fn(...args)), and Symbol.iterator. downlevelIteration allows for these iteration primitives to be used more accurately in ES5 environments if a Symbol.iterator implementation is present.

Example: Effects on for / of #

With this TypeScript code:

const str = "Hello!";
for (const s of str) {
  console.log(s);
}

Without downlevelIteration enabled, a for / of loop on any object is downleveled to a traditional for loop:

"use strict";
var str = "Hello!";
for (var _i = 0, str_1 = str; _i < str_1.length; _i++) {
    var s = str_1[_i];
    console.log(s);
}
 

This is often what people expect, but it’s not 100% compliant with ECMAScript iteration protocol. Certain strings, such as emoji (😜), have a .length of 2 (or even more!), but should iterate as 1 unit in a for-of loop. See this blog post by Jonathan New for a longer explanation.

When downlevelIteration is enabled, TypeScript will use a helper function that checks for a Symbol.iterator implementation (either native or polyfill). If this implementation is missing, you’ll fall back to index-based iteration.

"use strict";
var __values = (this && this.__values) || function(o) {
    var s = typeof Symbol === "function" && Symbol.iterator, m = s && o[s], i = 0;
    if (m) return m.call(o);
    if (o && typeof o.length === "number") return {
        next: function () {
            if (o && i >= o.length) o = void 0;
            return { value: o && o[i++], done: !o };
        }
    };
    throw new TypeError(s ? "Object is not iterable." : "Symbol.iterator is not defined.");
};
var e_1, _a;
var str = "Hello!";
try {
    for (var str_1 = __values(str), str_1_1 = str_1.next(); !str_1_1.done; str_1_1 = str_1.next()) {
        var s = str_1_1.value;
        console.log(s);
    }
}
catch (e_1_1) { e_1 = { error: e_1_1 }; }
finally {
    try {
        if (str_1_1 && !str_1_1.done && (_a = str_1.return)) _a.call(str_1);
    }
    finally { if (e_1) throw e_1.error; }
}
 

You can use tslib via importHelpers to reduce the amount of inline JavaScript too:

"use strict";
var __values = (this && this.__values) || function(o) {
    var s = typeof Symbol === "function" && Symbol.iterator, m = s && o[s], i = 0;
    if (m) return m.call(o);
    if (o && typeof o.length === "number") return {
        next: function () {
            if (o && i >= o.length) o = void 0;
            return { value: o && o[i++], done: !o };
        }
    };
    throw new TypeError(s ? "Object is not iterable." : "Symbol.iterator is not defined.");
};
var e_1, _a;
var str = "Hello!";
try {
    for (var str_1 = __values(str), str_1_1 = str_1.next(); !str_1_1.done; str_1_1 = str_1.next()) {
        var s = str_1_1.value;
        console.log(s);
    }
}
catch (e_1_1) { e_1 = { error: e_1_1 }; }
finally {
    try {
        if (str_1_1 && !str_1_1.done && (_a = str_1.return)) _a.call(str_1);
    }
    finally { if (e_1) throw e_1.error; }
}
 

Note: enabling downlevelIteration does not improve compliance if Symbol.iterator is not present in the runtime.

Example: Effects on Array Spreads #

This is an array spread:

// Make a new array who elements are 1 followed by the elements of arr2
const arr = [1, ...arr2];

Based on the description, it sounds easy to downlevel to ES5:

// The same, right?
const arr = [1].concat(arr2);

However, this is observably different in certain rare cases.

For example, if a source array is missing one or more items (contains a hole), the spread syntax will replace each empty item with undefined, whereas .concat will leave them intact.

// Make an array where the element at index 1 is missing
let arrayWithHole = ["a", , "c"];
let spread = [...arrayWithHole];
let concatenated = [].concat(arrayWithHole);

console.log(arrayWithHole);
// [ 'a', <1 empty item>, 'c' ]
console.log(spread);
// [ 'a', undefined, 'c' ]
console.log(concatenated);
// [ 'a', <1 empty item>, 'c' ]

Just as with for / of, downlevelIteration will use Symbol.iterator (if present) to more accurately emulate ES 6 behavior.

Controls whether TypeScript will emit a byte order mark (BOM) when writing output files. Some runtime environments require a BOM to correctly interpret a JavaScript files; others require that it is not present. The default value of false is generally best unless you have a reason to change it.

Only emit .d.ts files; do not emit .js files.

This setting is useful in two cases:

• You are using a transpiler other than TypeScript to generate your JavaScript.
• You are using TypeScript to only generate d.ts files for your consumers.

For certain downleveling operations, TypeScript uses some helper code for operations like extending class, spreading arrays or objects, and async operations. By default, these helpers are inserted into files which use them. This can result in code duplication if the same helper is used in many different modules.

If the importHelpers flag is on, these helper functions are instead imported from the tslib module. You will need to ensure that the tslib module is able to be imported at runtime. This only affects modules; global script files will not attempt to import modules.

For example, with this TypeScript:

export function fn(arr: number[]) {
  const arr2 = [1, ...arr];
}

Turning on downlevelIteration and importHelpers is still false:

var __read = (this && this.__read) || function (o, n) {
    var m = typeof Symbol === "function" && o[Symbol.iterator];
    if (!m) return o;
    var i = m.call(o), r, ar = [], e;
    try {
        while ((n === void 0 || n-- > 0) && !(r = i.next()).done) ar.push(r.value);
    }
    catch (error) { e = { error: error }; }
    finally {
        try {
            if (r && !r.done && (m = i["return"])) m.call(i);
        }
        finally { if (e) throw e.error; }
    }
    return ar;
};
var __spreadArray = (this && this.__spreadArray) || function (to, from, pack) {
    if (pack || arguments.length === 2) for (var i = 0, l = from.length, ar; i < l; i++) {
        if (ar || !(i in from)) {
            if (!ar) ar = Array.prototype.slice.call(from, 0, i);
            ar[i] = from[i];
        }
    }
    return to.concat(ar || Array.prototype.slice.call(from));
};
export function fn(arr) {
    var arr2 = __spreadArray([1], __read(arr), false);
}
 

Then turning on both downlevelIteration and importHelpers:

import { __read, __spreadArray } from "tslib";
export function fn(arr) {
    var arr2 = __spreadArray([1], __read(arr), false);
}
 

You can use noEmitHelpers when you provide your own implementations of these functions.

Deprecated in favor of verbatimModuleSyntax.

This flag controls how import works, there are 3 different options:

• remove: The default behavior of dropping import statements which only reference types.

• preserve: Preserves all import statements whose values or types are never used. This can cause imports/side-effects to be preserved.

• error: This preserves all imports (the same as the preserve option), but will error when a value import is only used as a type. This might be useful if you want to ensure no values are being accidentally imported, but still make side-effect imports explicit.

This flag works because you can use import type to explicitly create an import statement which should never be emitted into JavaScript.

When set, instead of writing out a .js.map file to provide source maps, TypeScript will embed the source map content in the .js files. Although this results in larger JS files, it can be convenient in some scenarios. For example, you might want to debug JS files on a webserver that doesn’t allow .map files to be served.

Mutually exclusive with sourceMap.

For example, with this TypeScript:

const helloWorld = "hi";
console.log(helloWorld);

Converts to this JavaScript:

"use strict";
const helloWorld = "hi";
console.log(helloWorld);
 

Then enable building it with inlineSourceMap enabled there is a comment at the bottom of the file which includes a source-map for the file.

"use strict";
const helloWorld = "hi";
console.log(helloWorld);
//# sourceMappingURL=data:application/json;base64,eyJ2ZXJzaW9uIjozLCJmaWxlIjoiaW5kZXguanMiLCJzb3VyY2VSb290IjoiIiwic291cmNlcyI6WyJpbmRleC50cyJdLCJuYW1lcyI6W10sIm1hcHBpbmdzIjoiO0FBQUEsTUFBTSxVQUFVLEdBQUcsSUFBSSxDQUFDO0FBQ3hCLE9BQU8sQ0FBQyxHQUFHLENBQUMsVUFBVSxDQUFDLENBQUMifQ==

When set, TypeScript will include the original content of the .ts file as an embedded string in the source map (using the source map’s sourcesContent property). This is often useful in the same cases as inlineSourceMap.

Requires either sourceMap or inlineSourceMap to be set.

For example, with this TypeScript:

const helloWorld = "hi";
console.log(helloWorld);

By default converts to this JavaScript:

"use strict";
const helloWorld = "hi";
console.log(helloWorld);
 

Then enable building it with inlineSources and inlineSourceMap enabled there is a comment at the bottom of the file which includes a source-map for the file. Note that the end is different from the example in inlineSourceMap because the source-map now contains the original source code also.

"use strict";
const helloWorld = "hi";
console.log(helloWorld);
//# sourceMappingURL=data:application/json;base64,eyJ2ZXJzaW9uIjozLCJmaWxlIjoiaW5kZXguanMiLCJzb3VyY2VSb290IjoiIiwic291cmNlcyI6WyJpbmRleC50cyJdLCJuYW1lcyI6W10sIm1hcHBpbmdzIjoiO0FBQUEsTUFBTSxVQUFVLEdBQUcsSUFBSSxDQUFDO0FBQ3hCLE9BQU8sQ0FBQyxHQUFHLENBQUMsVUFBVSxDQUFDLENBQUMiLCJzb3VyY2VzQ29udGVudCI6WyJjb25zdCBoZWxsb1dvcmxkID0gXCJoaVwiO1xuY29uc29sZS5sb2coaGVsbG9Xb3JsZCk7Il19

Specify the location where debugger should locate map files instead of generated locations. This string is treated verbatim inside the source-map, for example:

{
  "compilerOptions": {
    "sourceMap": true,
    "mapRoot": "https://my-website.com/debug/sourcemaps/"
  }
}

Would declare that index.js will have sourcemaps at https://my-website.com/debug/sourcemaps/index.js.map.

Specify the end of line sequence to be used when emitting files: ‘CRLF’ (dos) or ‘LF’ (unix).

Do not emit compiler output files like JavaScript source code, source-maps or declarations.

This makes room for another tool like Babel, or swc to handle converting the TypeScript file to a file which can run inside a JavaScript environment.

You can then use TypeScript as a tool for providing editor integration, and as a source code type-checker.

Instead of importing helpers with importHelpers, you can provide implementations in the global scope for the helpers you use and completely turn off emitting of helper functions.

For example, using this async function in ES5 requires a await-like function and generator-like function to run:

const getAPI = async (url: string) => {
  // Get API
  return {};
};

Which creates quite a lot of JavaScript:

"use strict";
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
    function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
    return new (P || (P = Promise))(function (resolve, reject) {
        function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
        function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
        function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
        step((generator = generator.apply(thisArg, _arguments || [])).next());
    });
};
var __generator = (this && this.__generator) || function (thisArg, body) {
    var _ = { label: 0, sent: function() { if (t[0] & 1) throw t[1]; return t[1]; }, trys: [], ops: [] }, f, y, t, g;
    return g = { next: verb(0), "throw": verb(1), "return": verb(2) }, typeof Symbol === "function" && (g[Symbol.iterator] = function() { return this; }), g;
    function verb(n) { return function (v) { return step([n, v]); }; }
    function step(op) {
        if (f) throw new TypeError("Generator is already executing.");
        while (g && (g = 0, op[0] && (_ = 0)), _) try {
            if (f = 1, y && (t = op[0] & 2 ? y["return"] : op[0] ? y["throw"] || ((t = y["return"]) && t.call(y), 0) : y.next) && !(t = t.call(y, op[1])).done) return t;
            if (y = 0, t) op = [op[0] & 2, t.value];
            switch (op[0]) {
                case 0: case 1: t = op; break;
                case 4: _.label++; return { value: op[1], done: false };
                case 5: _.label++; y = op[1]; op = [0]; continue;
                case 7: op = _.ops.pop(); _.trys.pop(); continue;
                default:
                    if (!(t = _.trys, t = t.length > 0 && t[t.length - 1]) && (op[0] === 6 || op[0] === 2)) { _ = 0; continue; }
                    if (op[0] === 3 && (!t || (op[1] > t[0] && op[1] < t[3]))) { _.label = op[1]; break; }
                    if (op[0] === 6 && _.label < t[1]) { _.label = t[1]; t = op; break; }
                    if (t && _.label < t[2]) { _.label = t[2]; _.ops.push(op); break; }
                    if (t[2]) _.ops.pop();
                    _.trys.pop(); continue;
            }
            op = body.call(thisArg, _);
        } catch (e) { op = [6, e]; y = 0; } finally { f = t = 0; }
        if (op[0] & 5) throw op[1]; return { value: op[0] ? op[1] : void 0, done: true };
    }
};
var getAPI = function (url) { return __awaiter(void 0, void 0, void 0, function () {
    return __generator(this, function (_a) {
        // Get API
        return [2 /*return*/, {}];
    });
}); };
 

Which can be switched out with your own globals via this flag:

"use strict";
var getAPI = function (url) { return __awaiter(void 0, void 0, void 0, function () {
    return __generator(this, function (_a) {
        // Get API
        return [2 /*return*/, {}];
    });
}); };
 

Do not emit compiler output files like JavaScript source code, source-maps or declarations if any errors were reported.

This defaults to false, making it easier to work with TypeScript in a watch-like environment where you may want to see results of changes to your code in another environment before making sure all errors are resolved.

If specified, .js (as well as .d.ts, .js.map, etc.) files will be emitted into this directory. The directory structure of the original source files is preserved; see rootDir if the computed root is not what you intended.

If not specified, .js files will be emitted in the same directory as the .ts files they were generated from:

$ tsc

example
├── index.js
└── index.ts

With a tsconfig.json like this:

{
  "compilerOptions": {
    "outDir": "dist"
  }
}

Running tsc with these settings moves the files into the specified dist folder:

$ tsc

example
├── dist
│   └── index.js
├── index.ts
└── tsconfig.json

If specified, all global (non-module) files will be concatenated into the single output file specified.

If module is system or amd, all module files will also be concatenated into this file after all global content.

Note: outFile cannot be used unless module is None, System, or AMD. This option cannot be used to bundle CommonJS or ES6 modules.

Do not erase const enum declarations in generated code. const enums provide a way to reduce the overall memory footprint of your application at runtime by emitting the enum value instead of a reference.

For example with this TypeScript:

const enum Album {
  JimmyEatWorldFutures = 1,
  TubRingZooHypothesis = 2,
  DogFashionDiscoAdultery = 3,
}
 
const selectedAlbum = Album.JimmyEatWorldFutures;
if (selectedAlbum === Album.JimmyEatWorldFutures) {
  console.log("That is a great choice.");
}

The default const enum behavior is to convert any Album.Something to the corresponding number literal, and to remove a reference to the enum from the JavaScript completely.

"use strict";
const selectedAlbum = 1 /* Album.JimmyEatWorldFutures */;
if (selectedAlbum === 1 /* Album.JimmyEatWorldFutures */) {
    console.log("That is a great choice.");
}
 

With preserveConstEnums set to true, the enum exists at runtime and the numbers are still emitted.

"use strict";
var Album;
(function (Album) {
    Album[Album["JimmyEatWorldFutures"] = 1] = "JimmyEatWorldFutures";
    Album[Album["TubRingZooHypothesis"] = 2] = "TubRingZooHypothesis";
    Album[Album["DogFashionDiscoAdultery"] = 3] = "DogFashionDiscoAdultery";
})(Album || (Album = {}));
const selectedAlbum = 1 /* Album.JimmyEatWorldFutures */;
if (selectedAlbum === 1 /* Album.JimmyEatWorldFutures */) {
    console.log("That is a great choice.");
}
 

This essentially makes such const enums a source-code feature only, with no runtime traces.

Deprecated in favor of verbatimModuleSyntax.

There are some cases where TypeScript can’t detect that you’re using an import. For example, take the following code:

import { Animal } from "./animal.js";

eval("console.log(new Animal().isDangerous())");

or code using ‘Compiles to HTML’ languages like Svelte or Vue. preserveValueImports will prevent TypeScript from removing the import, even if it appears unused.

When combined with isolatedModules: imported types must be marked as type-only because compilers that process single files at a time have no way of knowing whether imports are values that appear unused, or a type that must be removed in order to avoid a runtime crash.

Strips all comments from TypeScript files when converting into JavaScript. Defaults to false.

For example, this is a TypeScript file which has a JSDoc comment:

/** The translation of 'Hello world' into Portuguese */
export const helloWorldPTBR = "Olá Mundo";

When removeComments is set to true:

export const helloWorldPTBR = "Olá Mundo";
 

Without setting removeComments or having it as false:

/** The translation of 'Hello world' into Portuguese */
export const helloWorldPTBR = "Olá Mundo";
 

This means that your comments will show up in the JavaScript code.

Enables the generation of sourcemap files. These files allow debuggers and other tools to display the original TypeScript source code when actually working with the emitted JavaScript files. Source map files are emitted as .js.map (or .jsx.map) files next to the corresponding .js output file.

The .js files will in turn contain a sourcemap comment to indicate where the files are to external tools, for example:

// helloWorld.ts
export declare const helloWorld = "hi";

Compiling with sourceMap set to true creates the following JavaScript file:

// helloWorld.js
"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
exports.helloWorld = "hi";
//# sourceMappingURL=// helloWorld.js.map

And this also generates this json map:

// helloWorld.js.map
{
  "version": 3,
  "file": "ex.js",
  "sourceRoot": "",
  "sources": ["../ex.ts"],
  "names": [],
  "mappings": ";;AAAa,QAAA,UAAU,GAAG,IAAI,CAAA"
}

Specify the location where a debugger should locate TypeScript files instead of relative source locations. This string is treated verbatim inside the source-map where you can use a path or a URL:

{
  "compilerOptions": {
    "sourceMap": true,
    "sourceRoot": "https://my-website.com/debug/source/"
  }
}

Would declare that index.js will have a source file at https://my-website.com/debug/source/index.ts.

Do not emit declarations for code that has an @internal annotation in its JSDoc comment. This is an internal compiler option; use at your own risk, because the compiler does not check that the result is valid. If you are searching for a tool to handle additional levels of visibility within your d.ts files, look at api-extractor.

/**
 * Days available in a week
 * @internal
 */
export const daysInAWeek = 7;
 
/** Calculate how much someone earns in a week */
export function weeklySalary(dayRate: number) {
  return daysInAWeek * dayRate;
}

With the flag set to false (default):

/**
 * Days available in a week
 * @internal
 */
export declare const daysInAWeek = 7;
/** Calculate how much someone earns in a week */
export declare function weeklySalary(dayRate: number): number;
 

With stripInternal set to true the d.ts emitted will be redacted.

/** Calculate how much someone earns in a week */
export declare function weeklySalary(dayRate: number): number;
 

The JavaScript output is still the same.

Allow JavaScript files to be imported inside your project, instead of just .ts and .tsx files. For example, this JS file:

// @filename: card.js
export const defaultCardDeck = "Heart";

When imported into a TypeScript file will raise an error:

// @filename: index.ts
import { defaultCardDeck } from "./card";
 
console.log(defaultCardDeck);

Imports fine with allowJs enabled:

// @filename: index.ts
import { defaultCardDeck } from "./card";
 
console.log(defaultCardDeck);

This flag can be used as a way to incrementally add TypeScript files into JS projects by allowing the .ts and .tsx files to live along-side existing JavaScript files.

It can also be used along-side declaration and emitDeclarationOnly to create declarations for JS files.

Works in tandem with allowJs. When checkJs is enabled then errors are reported in JavaScript files. This is the equivalent of including // @ts-check at the top of all JavaScript files which are included in your project.

For example, this is incorrect JavaScript according to the parseFloat type definition which comes with TypeScript:

// parseFloat only takes a string
module.exports.pi = parseFloat(3.142);

When imported into a TypeScript module:

// @filename: constants.js
module.exports.pi = parseFloat(3.142);
 
// @filename: index.ts
import { pi } from "./constants";
console.log(pi);

You will not get any errors. However, if you turn on checkJs then you will get error messages from the JavaScript file.

// @filename: constants.js
module.exports.pi = parseFloat(3.142);
 
// @filename: index.ts
import { pi } from "./constants";
console.log(pi);

The maximum dependency depth to search under node_modules and load JavaScript files.

This flag can only be used when allowJs is enabled, and is used if you want to have TypeScript infer types for all of the JavaScript inside your node_modules.

Ideally this should stay at 0 (the default), and d.ts files should be used to explicitly define the shape of modules. However, there are cases where you may want to turn this on at the expense of speed and potential accuracy.

To avoid a possible memory bloat issues when working with very large JavaScript projects, there is an upper limit to the amount of memory TypeScript will allocate. Turning this flag on will remove the limit.

List of language service plugins to run inside the editor.

Language service plugins are a way to provide additional information to a user based on existing TypeScript files. They can enhance existing messages between TypeScript and an editor, or to provide their own error messages.

For example:

• ts-sql-plugin — Adds SQL linting with a template strings SQL builder.
• typescript-styled-plugin — Provides CSS linting inside template strings .
• typescript-eslint-language-service — Provides eslint error messaging and fix-its inside the compiler’s output.
• ts-graphql-plugin — Provides validation and auto-completion inside GraphQL query template strings.

VS Code has the ability for a extension to automatically include language service plugins, and so you may have some running in your editor without needing to define them in your tsconfig.json.

When set to true, allowSyntheticDefaultImports allows you to write an import like:

import React from "react";

instead of:

import * as React from "react";

When the module does not explicitly specify a default export.

For example, without allowSyntheticDefaultImports as true:

// @filename: utilFunctions.js
const getStringLength = (str) => str.length;
 
module.exports = {
  getStringLength,
};
 
// @filename: index.ts
import utils from "./utilFunctions";
 
const count = utils.getStringLength("Check JS");

This code raises an error because there isn’t a default object which you can import. Even though it feels like it should. For convenience, transpilers like Babel will automatically create a default if one isn’t created. Making the module look a bit more like:

// @filename: utilFunctions.js
const getStringLength = (str) => str.length;
const allFunctions = {
  getStringLength,
};

module.exports = allFunctions;
module.exports.default = allFunctions;

This flag does not affect the JavaScript emitted by TypeScript, it’s only for the type checking. This option brings the behavior of TypeScript in-line with Babel, where extra code is emitted to make using a default export of a module more ergonomic.

By default (with esModuleInterop false or not set) TypeScript treats CommonJS/AMD/UMD modules similar to ES6 modules. In doing this, there are two parts in particular which turned out to be flawed assumptions:

• a namespace import like import * as moment from "moment" acts the same as const moment = require("moment")

• a default import like import moment from "moment" acts the same as const moment = require("moment").default

This mis-match causes these two issues:

• the ES6 modules spec states that a namespace import (import * as x) can only be an object, by having TypeScript treating it the same as = require("x") then TypeScript allowed for the import to be treated as a function and be callable. That’s not valid according to the spec.

• while accurate to the ES6 modules spec, most libraries with CommonJS/AMD/UMD modules didn’t conform as strictly as TypeScript’s implementation.

Turning on esModuleInterop will fix both of these problems in the code transpiled by TypeScript. The first changes the behavior in the compiler, the second is fixed by two new helper functions which provide a shim to ensure compatibility in the emitted JavaScript:

import * as fs from "fs";
import _ from "lodash";

fs.readFileSync("file.txt", "utf8");
_.chunk(["a", "b", "c", "d"], 2);

With esModuleInterop disabled:

"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
const fs = require("fs");
const lodash_1 = require("lodash");
fs.readFileSync("file.txt", "utf8");
lodash_1.default.chunk(["a", "b", "c", "d"], 2);
 

With esModuleInterop set to true:

"use strict";
var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
    if (k2 === undefined) k2 = k;
    var desc = Object.getOwnPropertyDescriptor(m, k);
    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
      desc = { enumerable: true, get: function() { return m[k]; } };
    }
    Object.defineProperty(o, k2, desc);
}) : (function(o, m, k, k2) {
    if (k2 === undefined) k2 = k;
    o[k2] = m[k];
}));
var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
    Object.defineProperty(o, "default", { enumerable: true, value: v });
}) : function(o, v) {
    o["default"] = v;
});
var __importStar = (this && this.__importStar) || function (mod) {
    if (mod && mod.__esModule) return mod;
    var result = {};
    if (mod != null) for (var k in mod) if (k !== "default" && Object.prototype.hasOwnProperty.call(mod, k)) __createBinding(result, mod, k);
    __setModuleDefault(result, mod);
    return result;
};
var __importDefault = (this && this.__importDefault) || function (mod) {
    return (mod && mod.__esModule) ? mod : { "default": mod };
};
Object.defineProperty(exports, "__esModule", { value: true });
const fs = __importStar(require("fs"));
const lodash_1 = __importDefault(require("lodash"));
fs.readFileSync("file.txt", "utf8");
lodash_1.default.chunk(["a", "b", "c", "d"], 2);
 

Note: The namespace import import * as fs from "fs" only accounts for properties which are owned (basically properties set on the object and not via the prototype chain) on the imported object. If the module you’re importing defines its API using inherited properties, you need to use the default import form (import fs from "fs"), or disable esModuleInterop.

Note: You can make JS emit terser by enabling importHelpers:

"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
const tslib_1 = require("tslib");
const fs = tslib_1.__importStar(require("fs"));
const lodash_1 = tslib_1.__importDefault(require("lodash"));
fs.readFileSync("file.txt", "utf8");
lodash_1.default.chunk(["a", "b", "c", "d"], 2);
 

Enabling esModuleInterop will also enable allowSyntheticDefaultImports.

TypeScript follows the case sensitivity rules of the file system it’s running on. This can be problematic if some developers are working in a case-sensitive file system and others aren’t. If a file attempts to import fileManager.ts by specifying ./FileManager.ts the file will be found in a case-insensitive file system, but not on a case-sensitive file system.

When this option is set, TypeScript will issue an error if a program tries to include a file by a casing different from the casing on disk.

While you can use TypeScript to produce JavaScript code from TypeScript code, it’s also common to use other transpilers such as Babel to do this. However, other transpilers only operate on a single file at a time, which means they can’t apply code transforms that depend on understanding the full type system. This restriction also applies to TypeScript’s ts.transpileModule API which is used by some build tools.

These limitations can cause runtime problems with some TypeScript features like const enums and namespaces. Setting the isolatedModules flag tells TypeScript to warn you if you write certain code that can’t be correctly interpreted by a single-file transpilation process.

It does not change the behavior of your code, or otherwise change the behavior of TypeScript’s checking and emitting process.

Some examples of code which does not work when isolatedModules is enabled.

Exports of Non-Value Identifiers #

In TypeScript, you can import a type and then subsequently export it:

import { someType, someFunction } from "someModule";
 
someFunction();
 
export { someType, someFunction };

Because there’s no value for someType, the emitted export will not try to export it (this would be a runtime error in JavaScript):

export { someFunction };

Single-file transpilers don’t know whether someType produces a value or not, so it’s an error to export a name that only refers to a type.

Non-Module Files #

If isolatedModules is set, all implementation files must be modules (which means it has some form of import/export). An error occurs if any file isn’t a module:

function fn() {}

This restriction doesn’t apply to .d.ts files.

References to const enum members #

In TypeScript, when you reference a const enum member, the reference is replaced by its actual value in the emitted JavaScript. Changing this TypeScript:

declare const enum Numbers {
  Zero = 0,
  One = 1,
}
console.log(Numbers.Zero + Numbers.One);

To this JavaScript:

"use strict";
console.log(0 + 1);
 

Without knowledge of the values of these members, other transpilers can’t replace the references to Numbers, which would be a runtime error if left alone (since there are no Numbers object at runtime). Because of this, when isolatedModules is set, it is an error to reference an ambient const enum member.

This is to reflect the same flag in Node.js; which does not resolve the real path of symlinks.

This flag also exhibits the opposite behavior to Webpack’s resolve.symlinks option (i.e. setting TypeScript’s preserveSymlinks to true parallels setting Webpack’s resolve.symlinks to false, and vice-versa).

With this enabled, references to modules and packages (e.g. imports and /// <reference type="..." /> directives) are all resolved relative to the location of the symbolic link file, rather than relative to the path that the symbolic link resolves to.

By default, TypeScript does something called import elision. Basically, if you write something like

import { Car } from "./car";

export function drive(car: Car) {
  // ...
}

TypeScript detects that you’re only using an import for types and drops the import entirely. Your output JavaScript might look something like this:

export function drive(car) {
  // ...
}

Most of the time this is good, because if Car isn’t a value that’s exported from ./car, we’ll get a runtime error.

But it does add a layer of complexity for certain edge cases. For example, notice there’s no statement like import "./car"; - the import was dropped entirely. That actually makes a difference for modules that have side-effects or not.

TypeScript’s emit strategy for JavaScript also has another few layers of complexity - import elision isn’t always just driven by how an import is used - it often consults how a value is declared as well. So it’s not always clear whether code like the following

export { Car } from "./car";

should be preserved or dropped. If Car is declared with something like a class, then it can be preserved in the resulting JavaScript file. But if Car is only declared as a type alias or interface, then the JavaScript file shouldn’t export Car at all.

While TypeScript might be able to make these emit decisions based on information from across files, not every compiler can.

The type modifier on imports and exports helps with these situations a bit. We can make it explicit whether an import or export is only being used for type analysis, and can be dropped entirely in JavaScript files by using the type modifier.

// This statement can be dropped entirely in JS output
import type * as car from "./car";

// The named import/export 'Car' can be dropped in JS output
import { type Car } from "./car";
export { type Car } from "./car";

type modifiers are not quite useful on their own - by default, module elision will still drop imports, and nothing forces you to make the distinction between type and plain imports and exports. So TypeScript has the flag --importsNotUsedAsValues to make sure you use the type modifier, --preserveValueImports to prevent some module elision behavior, and --isolatedModules to make sure that your TypeScript code works across different compilers. Unfortunately, understanding the fine details of those 3 flags is hard, and there are still some edge cases with unexpected behavior.

TypeScript 5.0 introduces a new option called --verbatimModuleSyntax to simplify the situation. The rules are much simpler - any imports or exports without a type modifier are left around. Anything that uses the type modifier is dropped entirely.

// Erased away entirely.
import type { A } from "a";

// Rewritten to 'import { b } from "bcd";'
import { b, type c, type d } from "bcd";

// Rewritten to 'import {} from "xyz";'
import { type xyz } from "xyz";

With this new option, what you see is what you get.

That does have some implications when it comes to module interop though. Under this flag, ECMAScript imports and exports won’t be rewritten to require calls when your settings or file extension implied a different module system. Instead, you’ll get an error. If you need to emit code that uses require and module.exports, you’ll have to use TypeScript’s module syntax that predates ES2015:

+———————————–+———————————–+ | Input TypeScript | Output JavaScript | +===================================+===================================+ | ts | js | | import foo = require(“foo”); | const foo = require(“foo”); | | | | +———————————–+———————————–+ | ts | js | | function foo() {} | function foo() {} | | function bar() {} | function bar() {} | | function baz() {} | function baz() {} | | | | | export = { | module.exports = { | | foo, | foo, | | bar, | bar, | | baz, | baz, | | }; | }; | | | | +———————————–+———————————–+

While this is a limitation, it does help make some issues more obvious. For example, it’s very common to forget to set the type field in package.json under --module node16. As a result, developers would start writing CommonJS modules instead of an ES modules without realizing it, giving surprising lookup rules and JavaScript output. This new flag ensures that you’re intentional about the file type you’re using because the syntax is intentionally different.

Because --verbatimModuleSyntax provides a more consistent story than --importsNotUsedAsValues and --preserveValueImports, those two existing flags are being deprecated in its favor.

For more details, read up on the original pull request and its proposal issue.

In prior versions of TypeScript, this controlled what encoding was used when reading text files from disk. Today, TypeScript assumes UTF-8 encoding, but will correctly detect UTF-16 (BE and LE) or UTF-8 BOMs.

This flag changes the keyof type operator to return string instead of string | number when applied to a type with a string index signature.

This flag is used to help people keep this behavior from before TypeScript 2.9’s release.

You shouldn’t need this. By default, when emitting a module file to a non-ES6 target, TypeScript emits a "use strict"; prologue at the top of the file. This setting disables the prologue.

TypeScript will unify type parameters when comparing two generic functions.

type A = <T, U>(x: T, y: U) => [T, U];
type B = <S>(x: S, y: S) => [S, S];
 
function f(a: A, b: B) {
  b = a; // Ok
  a = b; // Error
}

This flag can be used to remove that check.

Use outFile instead.

The out option computes the final file location in a way that is not predictable or consistent. This option is retained for backward compatibility only and is deprecated.

This disables reporting of excess property errors, such as the one shown in the following example:

type Point = { x: number; y: number };
const p: Point = { x: 1, y: 3, m: 10 };

This flag was added to help people migrate to the stricter checking of new object literals in TypeScript 1.6.

We don’t recommend using this flag in a modern codebase, you can suppress one-off cases where you need it using // @ts-ignore.

Turning suppressImplicitAnyIndexErrors on suppresses reporting the error about implicit anys when indexing into objects, as shown in the following example:

const obj = { x: 10 };
console.log(obj["foo"]);

Using suppressImplicitAnyIndexErrors is quite a drastic approach. It is recommended to use a @ts-ignore comment instead:

const obj = { x: 10 };
// @ts-ignore
console.log(obj["foo"]);

Enables experimental support for emitting type metadata for decorators which works with the module reflect-metadata.

For example, here is the TypeScript

function LogMethod(
  target: any,
  propertyKey: string | symbol,
  descriptor: PropertyDescriptor
) {
  console.log(target);
  console.log(propertyKey);
  console.log(descriptor);
}
 
class Demo {
  @LogMethod
  public foo(bar: number) {
    // do nothing
  }
}
 
const demo = new Demo();

With emitDecoratorMetadata not set to true (default) the emitted JavaScript is:

"use strict";
var __decorate = (this && this.__decorate) || function (decorators, target, key, desc) {
    var c = arguments.length, r = c < 3 ? target : desc === null ? desc = Object.getOwnPropertyDescriptor(target, key) : desc, d;
    if (typeof Reflect === "object" && typeof Reflect.decorate === "function") r = Reflect.decorate(decorators, target, key, desc);
    else for (var i = decorators.length - 1; i >= 0; i--) if (d = decorators[i]) r = (c < 3 ? d(r) : c > 3 ? d(target, key, r) : d(target, key)) || r;
    return c > 3 && r && Object.defineProperty(target, key, r), r;
};
function LogMethod(target, propertyKey, descriptor) {
    console.log(target);
    console.log(propertyKey);
    console.log(descriptor);
}
class Demo {
    foo(bar) {
        // do nothing
    }
}
__decorate([
    LogMethod
], Demo.prototype, "foo", null);
const demo = new Demo();