Commander.js

The complete solution for node.js command-line interfaces.

Read this in other languages: English | 简体中文

• Commander.js
  • Installation
  • Quick Start
  • Declaring program variable
  • Options
    • Common option types, boolean and value
    • Default option value
    • Other option types, negatable boolean and boolean|value
    • Required option
    • Variadic option
    • Version option
    • More configuration
    • Custom option processing
  • Commands
    • Command-arguments
      • More configuration
      • Custom argument processing
    • Action handler
    • Stand-alone executable (sub)commands
    • Life cycle hooks
  • Automated help
    • Custom help
    • Display help after errors
    • Display help from code
    • .name
    • .usage
    • .description and .summary
    • .helpOption(flags, description)
    • .helpCommand()
    • Help Groups
    • More configuration
  • Custom event listeners
  • Bits and pieces
    • .parse() and .parseAsync()
    • Parsing Configuration
    • Legacy options as properties
    • TypeScript
    • createCommand()
    • Node options such as --harmony
    • Debugging stand-alone executable subcommands
    • npm run-script
    • Display error
    • Override exit and output handling
    • Additional documentation
  • Support
    • Commander for enterprise

For information about terms used in this document see: terminology

Installation

1npm install commander

Quick Start

You write code to describe your command line interface. Commander looks after parsing the arguments into options and command-arguments, displays usage errors for problems, and implements a help system.

Commander is strict and displays an error for unrecognised options. The two most used option types are a boolean option, and an option which takes its value from the following argument.

Example file: split.js

1const { program } = require('commander');
2
3program
4  .option('--first')
5  .option('-s, --separator <char>')
6  .argument('<string>');
7
8program.parse();
9
10const options = program.opts();
11const limit = options.first ? 1 : undefined;
12console.log(program.args[0].split(options.separator, limit));

1$ node split.js -s / --fits a/b/c
2error: unknown option '--fits'
3(Did you mean --first?)
4$ node split.js -s / --first a/b/c
5[ 'a' ]

Here is a more complete program using a subcommand and with descriptions for the help. In a multi-command program, you have an action handler for each command (or stand-alone executables for the commands).

Example file: string-util.js

1const { Command } = require('commander');
2const program = new Command();
3
4program
5  .name('string-util')
6  .description('CLI to some JavaScript string utilities')
7  .version('0.8.0');
8
9program.command('split')
10  .description('Split a string into substrings and display as an array')
11  .argument('<string>', 'string to split')
12  .option('--first', 'display just the first substring')
13  .option('-s, --separator <char>', 'separator character', ',')
14  .action((str, options) => {
15    const limit = options.first ? 1 : undefined;
16    console.log(str.split(options.separator, limit));
17  });
18
19program.parse();

1$ node string-util.js help split
2Usage: string-util split [options] <string>
3
4Split a string into substrings and display as an array.
5
6Arguments:
7  string                  string to split
8
9Options:
10  --first                 display just the first substring
11  -s, --separator <char>  separator character (default: ",")
12  -h, --help              display help for command
13
14$ node string-util.js split --separator=/ a/b/c
15[ 'a', 'b', 'c' ]

More samples can be found in the examples directory.

Declaring program variable

Commander exports a global object which is convenient for quick programs. This is used in the examples in this README for brevity.

1// CommonJS (.cjs)
2const { program } = require('commander');

For larger programs which may use commander in multiple ways, including unit testing, it is better to create a local Command object to use.

1// CommonJS (.cjs)
2const { Command } = require('commander');
3const program = new Command();

1// ECMAScript (.mjs)
2import { Command } from 'commander';
3const program = new Command();

1// TypeScript (.ts)
2import { Command } from 'commander';
3const program = new Command();

Options

Options are defined with the .option() method, also serving as documentation for the options. Each option can have a short flag (single character) and a long name, separated by a comma or space or vertical bar ('|'). To allow a wider range of short-ish flags than just single characters, you may also have two long options. Examples:

1program
2  .option('-p, --port <number>', 'server port number')
3  .option('--trace', 'add extra debugging output')
4  .option('--ws, --workspace <name>', 'use a custom workspace')

The parsed options can be accessed by calling .opts() on a Command object, and are passed to the action handler.

Multi-word options such as "--template-engine" are camel-cased, becoming program.opts().templateEngine etc.

An option and its option-argument can be separated by a space, or combined into the same argument. The option-argument can follow the short option directly or follow an = for a long option.

1serve -p 80
2serve -p80
3serve --port 80
4serve --port=80

You can use -- to indicate the end of the options, and any remaining arguments will be used without being interpreted.

By default, options on the command line are not positional, and can be specified before or after other arguments.

There are additional related routines for when .opts() is not enough:

• .optsWithGlobals() returns merged local and global option values
• .getOptionValue() and .setOptionValue() work with a single option value
• .getOptionValueSource() and .setOptionValueWithSource() include where the option value came from

Common option types, boolean and value

The two most used option types are a boolean option, and an option which takes its value from the following argument (declared with angle brackets like --expect <value>). Both are undefined unless specified on command line.

Example file: options-common.js

1program
2  .option('-d, --debug', 'output extra debugging')
3  .option('-s, --small', 'small pizza size')
4  .option('-p, --pizza-type <type>', 'flavour of pizza');
5
6program.parse(process.argv);
7
8const options = program.opts();
9if (options.debug) console.log(options);
10console.log('pizza details:');
11if (options.small) console.log('- small pizza size');
12if (options.pizzaType) console.log(`- ${options.pizzaType}`);

1$ pizza-options -p
2error: option '-p, --pizza-type <type>' argument missing
3$ pizza-options -d -s -p vegetarian
4{ debug: true, small: true, pizzaType: 'vegetarian' }
5pizza details:
6- small pizza size
7- vegetarian
8$ pizza-options --pizza-type=cheese
9pizza details:
10- cheese

Multiple boolean short options may be combined following the dash, and may be followed by a single short option taking a value. For example -d -s -p cheese may be written as -ds -p cheese or even -dsp cheese.

Options with an expected option-argument are greedy and will consume the following argument whatever the value. So --id -xyz reads -xyz as the option-argument.

program.parse(arguments) processes the arguments, leaving any args not consumed by the program options in the program.args array. The parameter is optional and defaults to process.argv.

Default option value

You can specify a default value for an option.

Example file: options-defaults.js

1program
2  .option('-c, --cheese <type>', 'add the specified type of cheese', 'blue');
3
4program.parse();
5
6console.log(`cheese: ${program.opts().cheese}`);

1$ pizza-options
2cheese: blue
3$ pizza-options --cheese stilton
4cheese: stilton

Other option types, negatable boolean and boolean|value

You can define a boolean option long name with a leading no- to set the option value to false when used. Defined alone this also makes the option true by default.

If you define --foo first, adding --no-foo does not change the default value from what it would otherwise be.

Example file: options-negatable.js

1program
2  .option('--no-sauce', 'Remove sauce')
3  .option('--cheese <flavour>', 'cheese flavour', 'mozzarella')
4  .option('--no-cheese', 'plain with no cheese')
5  .parse();
6
7const options = program.opts();
8const sauceStr = options.sauce ? 'sauce' : 'no sauce';
9const cheeseStr = (options.cheese === false) ? 'no cheese' : `${options.cheese} cheese`;
10console.log(`You ordered a pizza with ${sauceStr} and ${cheeseStr}`);

1$ pizza-options
2You ordered a pizza with sauce and mozzarella cheese
3$ pizza-options --sauce
4error: unknown option '--sauce'
5$ pizza-options --cheese=blue
6You ordered a pizza with sauce and blue cheese
7$ pizza-options --no-sauce --no-cheese
8You ordered a pizza with no sauce and no cheese

You can specify an option which may be used as a boolean option but may optionally take an option-argument (declared with square brackets like --optional [value]).

Example file: options-boolean-or-value.js

1program
2  .option('-c, --cheese [type]', 'Add cheese with optional type');
3
4program.parse(process.argv);
5
6const options = program.opts();
7if (options.cheese === undefined) console.log('no cheese');
8else if (options.cheese === true) console.log('add cheese');
9else console.log(`add cheese type ${options.cheese}`);

1$ pizza-options
2no cheese
3$ pizza-options --cheese
4add cheese
5$ pizza-options --cheese mozzarella
6add cheese type mozzarella

Options with an optional option-argument are not greedy and will ignore arguments starting with a dash. So id behaves as a boolean option for --id -ABCD, but you can use a combined form if needed like --id=-ABCD. Negative numbers are special and are accepted as an option-argument.

For information about possible ambiguous cases, see options taking varying arguments.

Required option

You may specify a required (mandatory) option using .requiredOption(). The option must have a value after parsing, usually specified on the command line, or perhaps from a default value (say from environment). The method is otherwise the same as .option() in format, taking flags and description, and optional default value or custom processing.

Example file: options-required.js

1program
2  .requiredOption('-c, --cheese <type>', 'pizza must have cheese');
3
4program.parse();

1$ pizza
2error: required option '-c, --cheese <type>' not specified

Variadic option

You may make an option variadic by appending ... to the value placeholder when declaring the option. On the command line you can then specify multiple option-arguments, and the parsed option value will be an array. The extra arguments are read until the first argument starting with a dash. The special argument -- stops option processing entirely. If a value is specified in the same argument as the option then no further values are read.

Example file: options-variadic.js

1program
2  .option('-n, --number <numbers...>', 'specify numbers')
3  .option('-l, --letter [letters...]', 'specify letters');
4
5program.parse();
6
7console.log('Options: ', program.opts());
8console.log('Remaining arguments: ', program.args);

1$ collect -n 1 2 3 --letter a b c
2Options:  { number: [ '1', '2', '3' ], letter: [ 'a', 'b', 'c' ] }
3Remaining arguments:  []
4$ collect --letter=A -n80 operand
5Options:  { number: [ '80' ], letter: [ 'A' ] }
6Remaining arguments:  [ 'operand' ]
7$ collect --letter -n 1 -n 2 3 -- operand
8Options:  { number: [ '1', '2', '3' ], letter: true }
9Remaining arguments:  [ 'operand' ]

For information about possible ambiguous cases, see options taking varying arguments.

Version option

The optional version method adds handling for displaying the command version. The default option flags are -V and --version, and when present the command prints the version number and exits.

1program.version('0.0.1');

1$ ./examples/pizza -V
20.0.1

You may change the flags and description by passing additional parameters to the version method, using the same syntax for flags as the option method.

1program.version('0.0.1', '-v, --vers', 'output the current version');

More configuration

You can add most options using the .option() method, but there are some additional features available by constructing an Option explicitly for less common cases.

Example files: options-extra.js, options-env.js, options-conflicts.js, options-implies.js

1program
2  .addOption(new Option('-s, --secret').hideHelp())
3  .addOption(new Option('-t, --timeout <delay>', 'timeout in seconds').default(60, 'one minute'))
4  .addOption(new Option('-d, --drink <size>', 'drink size').choices(['small', 'medium', 'large']))
5  .addOption(new Option('-p, --port <number>', 'port number').env('PORT'))
6  .addOption(new Option('--donate [amount]', 'optional donation in dollars').preset('20').argParser(parseFloat))
7  .addOption(new Option('--disable-server', 'disables the server').conflicts('port'))
8  .addOption(new Option('--free-drink', 'small drink included free ').implies({ drink: 'small' }));

1$ extra --help
2Usage: help [options]
3
4Options:
5  -t, --timeout <delay>  timeout in seconds (default: one minute)
6  -d, --drink <size>     drink cup size (choices: "small", "medium", "large")
7  -p, --port <number>    port number (env: PORT)
8  --donate [amount]      optional donation in dollars (preset: "20")
9  --disable-server       disables the server
10  --free-drink           small drink included free
11  -h, --help             display help for command
12
13$ extra --drink huge
14error: option '-d, --drink <size>' argument 'huge' is invalid. Allowed choices are small, medium, large.
15
16$ PORT=80 extra --donate --free-drink
17Options:  { timeout: 60, donate: 20, port: '80', freeDrink: true, drink: 'small' }
18
19$ extra --disable-server --port 8000
20error: option '--disable-server' cannot be used with option '-p, --port <number>'

Specify a required (mandatory) option using the Option method .makeOptionMandatory(). This matches the Command method .requiredOption().

Custom option processing

You may specify a function to do custom processing of option-arguments. The callback function receives two parameters, the user specified option-argument and the previous value for the option. It returns the new value for the option.

This allows you to coerce the option-argument to the desired type, or accumulate values, or do entirely custom processing.

You can optionally specify the default/starting value for the option after the function parameter.

Example file: options-custom-processing.js

1function myParseInt(value, dummyPrevious) {
2  // parseInt takes a string and a radix
3  const parsedValue = parseInt(value, 10);
4  if (isNaN(parsedValue)) {
5    throw new commander.InvalidArgumentError('Not a number.');
6  }
7  return parsedValue;
8}
9
10function increaseVerbosity(dummyValue, previous) {
11  return previous + 1;
12}
13
14function collect(value, previous) {
15  return previous.concat([value]);
16}
17
18function commaSeparatedList(value, dummyPrevious) {
19  return value.split(',');
20}
21
22program
23  .option('-f, --float <number>', 'float argument', parseFloat)
24  .option('-i, --integer <number>', 'integer argument', myParseInt)
25  .option('-v, --verbose', 'verbosity that can be increased', increaseVerbosity, 0)
26  .option('-c, --collect <value>', 'repeatable value', collect, [])
27  .option('-l, --list <items>', 'comma separated list', commaSeparatedList)
28;
29
30program.parse();
31
32const options = program.opts();
33if (options.float !== undefined) console.log(`float: ${options.float}`);
34if (options.integer !== undefined) console.log(`integer: ${options.integer}`);
35if (options.verbose > 0) console.log(`verbosity: ${options.verbose}`);
36if (options.collect.length > 0) console.log(options.collect);
37if (options.list !== undefined) console.log(options.list);

1$ custom -f 1e2
2float: 100
3$ custom --integer 2
4integer: 2
5$ custom -v -v -v
6verbose: 3
7$ custom -c a -c b -c c
8[ 'a', 'b', 'c' ]
9$ custom --list x,y,z
10[ 'x', 'y', 'z' ]

Commands

You can specify (sub)commands using .command() or .addCommand(). There are two ways these can be implemented: using an action handler attached to the command, or as a stand-alone executable file (described in more detail later). The subcommands may be nested (example).

In the first parameter to .command() you specify the command name. You may append the command-arguments after the command name, or specify them separately using .argument(). The arguments may be <required> or [optional], and the last argument may also be variadic....

You can use .addCommand() to add an already configured subcommand to the program.

For example:

1// Command implemented using action handler (description is supplied separately to `.command`)
2// Returns new command for configuring.
3program
4  .command('clone <source> [destination]')
5  .description('clone a repository into a newly created directory')
6  .action((source, destination) => {
7    console.log('clone command called');
8  });
9
10// Command implemented using stand-alone executable file, indicated by adding description as second parameter to `.command`.
11// Returns `this` for adding more commands.
12program
13  .command('start <service>', 'start named service')
14  .command('stop [service]', 'stop named service, or all if no name supplied');
15
16// Command prepared separately.
17// Returns `this` for adding more commands.
18program
19  .addCommand(build.makeBuildCommand());

Configuration options can be passed with the call to .command() and .addCommand(). Specifying hidden: true will remove the command from the generated help output. Specifying isDefault: true will run the subcommand if no other subcommand is specified (example).

You can add alternative names for a command with .alias(). (example)

.command() automatically copies the inherited settings from the parent command to the newly created subcommand. This is only done during creation, any later setting changes to the parent are not inherited.

For safety, .addCommand() does not automatically copy the inherited settings from the parent command. There is a helper routine .copyInheritedSettings() for copying the settings when they are wanted.

Command-arguments

For subcommands, you can specify the argument syntax in the call to .command() (as shown above). This is the only method usable for subcommands implemented using a stand-alone executable, but for other subcommands you can instead use the following method.

To configure a command, you can use .argument() to specify each expected command-argument. You supply the argument name and an optional description. The argument may be <required> or [optional]. You can specify a default value for an optional command-argument.

Example file: argument.js

1program
2  .version('0.1.0')
3  .argument('<username>', 'user to login')
4  .argument('[password]', 'password for user, if required', 'no password given')
5  .action((username, password) => {
6    console.log('username:', username);
7    console.log('password:', password);
8  });

The last argument of a command can be variadic, and only the last argument. To make an argument variadic you append ... to the argument name. A variadic argument is passed to the action handler as an array. For example:

1program
2  .version('0.1.0')
3  .command('rmdir')
4  .argument('<dirs...>')
5  .action(function (dirs) {
6    dirs.forEach((dir) => {
7      console.log('rmdir %s', dir);
8    });
9  });

There is a convenience method to add multiple arguments at once, but without descriptions:

1program
2  .arguments('<username> <password>');

More configuration

There are some additional features available by constructing an Argument explicitly for less common cases.

Example file: arguments-extra.js

1program
2  .addArgument(new commander.Argument('<drink-size>', 'drink cup size').choices(['small', 'medium', 'large']))
3  .addArgument(new commander.Argument('[timeout]', 'timeout in seconds').default(60, 'one minute'))

Custom argument processing

You may specify a function to do custom processing of command-arguments (like for option-arguments). The callback function receives two parameters, the user specified command-argument and the previous value for the argument. It returns the new value for the argument.

The processed argument values are passed to the action handler, and saved as .processedArgs.

You can optionally specify the default/starting value for the argument after the function parameter.

Example file: arguments-custom-processing.js

1program
2  .command('add')
3  .argument('<first>', 'integer argument', myParseInt)
4  .argument('[second]', 'integer argument', myParseInt, 1000)
5  .action((first, second) => {
6    console.log(`${first} + ${second} = ${first + second}`);
7  })
8;

Action handler

The action handler gets passed a parameter for each command-argument you declared, and two additional parameters which are the parsed options and the command object itself.

Example file: thank.js

1program
2  .argument('<name>')
3  .option('-t, --title <honorific>', 'title to use before name')
4  .option('-d, --debug', 'display some debugging')
5  .action((name, options, command) => {
6    if (options.debug) {
7      console.error('Called %s with options %o', command.name(), options);
8    }
9    const title = options.title ? `${options.title} ` : '';
10    console.log(`Thank-you ${title}${name}`);
11  });

If you prefer, you can work with the command directly and skip declaring the parameters for the action handler. The this keyword is set to the running command and can be used from a function expression (but not from an arrow function).

Example file: action-this.js

1program
2  .command('serve')
3  .argument('<script>')
4  .option('-p, --port <number>', 'port number', 80)
5  .action(function() {
6    console.error('Run script %s on port %s', this.args[0], this.opts().port);
7  });

You may supply an async action handler, in which case you call .parseAsync rather than .parse.

1async function run() { /* code goes here */ }
2
3async function main() {
4  program
5    .command('run')
6    .action(run);
7  await program.parseAsync(process.argv);
8}

A command's options and arguments on the command line are validated when the command is used. Any unknown options or missing arguments or excess arguments will be reported as an error. You can suppress the unknown option check with .allowUnknownOption(). You can suppress the excess arguments check with .allowExcessArguments().

Stand-alone executable (sub)commands

When .command() is invoked with a description argument, this tells Commander that you're going to use stand-alone executables for subcommands. Commander will search the files in the directory of the entry script for a file with the name combination command-subcommand, like pm-install or pm-search in the example below. The search includes trying common file extensions, like .js. You may specify a custom name (and path) with the executableFile configuration option. You may specify a custom search directory for subcommands with .executableDir().

You handle the options for an executable (sub)command in the executable, and don't declare them at the top-level.

Example file: pm

1program
2  .name('pm')
3  .version('0.1.0')
4  .command('install [package-names...]', 'install one or more packages')
5  .command('search [query]', 'search with optional query')
6  .command('update', 'update installed packages', { executableFile: 'myUpdateSubCommand' })
7  .command('list', 'list packages installed', { isDefault: true });
8
9program.parse(process.argv);

If the program is designed to be installed globally, make sure the executables have proper modes, like 755.

Life cycle hooks

You can add callback hooks to a command for life cycle events.

Example file: hook.js

1program
2  .option('-t, --trace', 'display trace statements for commands')
3  .hook('preAction', (thisCommand, actionCommand) => {
4    if (thisCommand.opts().trace) {
5      console.log(`About to call action handler for subcommand: ${actionCommand.name()}`);
6      console.log('arguments: %O', actionCommand.args);
7      console.log('options: %o', actionCommand.opts());
8    }
9  });

The callback hook can be async, in which case you call .parseAsync rather than .parse. You can add multiple hooks per event.

The supported events are:

event name	when hook called	callback parameters
preAction, postAction	before/after action handler for this command and its nested subcommands	(thisCommand, actionCommand)
preSubcommand	before parsing direct subcommand	(thisCommand, subcommand)

For an overview of the life cycle events see parsing life cycle and hooks.

Automated help

The help information is auto-generated based on the information commander already knows about your program. The default help option is -h,--help.

Example file: pizza

1$ node ./examples/pizza --help
2Usage: pizza [options]
3
4An application for pizza ordering
5
6Options:
7  -p, --peppers        Add peppers
8  -c, --cheese <type>  Add the specified type of cheese (default: "marble")
9  -C, --no-cheese      You do not want any cheese
10  -h, --help           display help for command

A help command is added by default if your command has subcommands. It can be used alone, or with a subcommand name to show further help for the subcommand. These are effectively the same if the shell program has implicit help:

1shell help
2shell --help
3
4shell help spawn
5shell spawn --help

Long descriptions are wrapped to fit the available width. (However, a description that includes a line-break followed by whitespace is assumed to be pre-formatted and not wrapped.)

Custom help

You can add extra text to be displayed along with the built-in help.

Example file: custom-help

1program
2  .option('-f, --foo', 'enable some foo');
3
4program.addHelpText('after', `
5
6Example call:
7  $ custom-help --help`);

Yields the following help output:

1Usage: custom-help [options]
2
3Options:
4  -f, --foo   enable some foo
5  -h, --help  display help for command
6
7Example call:
8  $ custom-help --help

The positions in order displayed are:

• beforeAll: add to the program for a global banner or header
• before: display extra information before built-in help
• after: display extra information after built-in help
• afterAll: add to the program for a global footer (epilog)

The positions "beforeAll" and "afterAll" apply to the command and all its subcommands.

The second parameter can be a string, or a function returning a string. The function is passed a context object for your convenience. The properties are:

• error: a boolean for whether the help is being displayed due to a usage error
• command: the Command which is displaying the help

Display help after errors

The default behaviour for usage errors is to just display a short error message. You can change the behaviour to show the full help or a custom help message after an error.

1program.showHelpAfterError();
2// or
3program.showHelpAfterError('(add --help for additional information)');

1$ pizza --unknown
2error: unknown option '--unknown'
3(add --help for additional information)

The default behaviour is to suggest correct spelling after an error for an unknown command or option. You can disable this.

1program.showSuggestionAfterError(false);

1$ pizza --hepl
2error: unknown option '--hepl'
3(Did you mean --help?)

Display help from code

.help(): display help information and exit immediately. You can optionally pass { error: true } to display on stderr and exit with an error status.

.outputHelp(): output help information without exiting. You can optionally pass { error: true } to display on stderr.

.helpInformation(): get the built-in command help information as a string for processing or displaying yourself.

.name

The command name appears in the help, and is also used for locating stand-alone executable subcommands.

You may specify the program name using .name() or in the Command constructor. For the program, Commander will fall back to using the script name from the full arguments passed into .parse(). However, the script name varies depending on how your program is launched, so you may wish to specify it explicitly.

1program.name('pizza');
2const pm = new Command('pm');

Subcommands get a name when specified using .command(). If you create the subcommand yourself to use with .addCommand(), then set the name using .name() or in the Command constructor.

.usage

This allows you to customise the usage description in the first line of the help. Given:

1program
2  .name("my-command")
3  .usage("[global options] command")

The help will start with:

1Usage: my-command [global options] command

.description and .summary

The description appears in the help for the command. You can optionally supply a shorter summary to use when listed as a subcommand of the program.

1program
2  .command("duplicate")
3  .summary("make a copy")
4  .description(`Make a copy of the current project.
5This may require additional disk space.
6  `);

.helpOption(flags, description)

By default, every command has a help option. You may change the default help flags and description. Pass false to disable the built-in help option.

1program
2  .helpOption('-e, --HELP', 'read more information');

(Or use .addHelpOption() to add an option you construct yourself.)

.helpCommand()

A help command is added by default if your command has subcommands. You can explicitly turn on or off the implicit help command with .helpCommand(true) and .helpCommand(false).

You can both turn on and customise the help command by supplying the name and description:

1program.helpCommand('assist [command]', 'show assistance');

(Or use .addHelpCommand() to add a command you construct yourself.)

Help Groups

The help by default lists options under the the heading Options: and commands under Commands:. You can create your own groups with different headings. The high-level way is to set the desired group heading while adding the options and commands, using .optionsGroup() and .commandsGroup(). The low-level way is using .helpGroup() on an individual Option or Command

Example file: help-groups.js

More configuration

The built-in help is formatted using the Help class. You can configure the help by modifying data properties and methods using .configureHelp(), or by subclassing Help using .createHelp() .

Simple properties include sortSubcommands, sortOptions, and showGlobalOptions. You can add color using the style methods like styleTitle().

For more detail and examples of changing the displayed text, color, and layout see (./docs/help-in-depth.md).

Custom event listeners

You can execute custom actions by listening to command and option events.

1program.on('option:verbose', function () {
2  process.env.VERBOSE = this.opts().verbose;
3});

Bits and pieces

.parse() and .parseAsync()

Call with no parameters to parse process.argv. Detects Electron and special node options like node --eval. Easy mode!

Or call with an array of strings to parse, and optionally where the user arguments start by specifying where the arguments are from:

• 'node': default, argv[0] is the application and argv[1] is the script being run, with user arguments after that
• 'electron': argv[0] is the application and argv[1] varies depending on whether the electron application is packaged
• 'user': just user arguments

For example:

1program.parse(); // parse process.argv and auto-detect electron and special node flags
2program.parse(process.argv); // assume argv[0] is app and argv[1] is script
3program.parse(['--port', '80'], { from: 'user' }); // just user supplied arguments, nothing special about argv[0]

Use parseAsync instead of parse if any of your action handlers are async.

Parsing Configuration

If the default parsing does not suit your needs, there are some behaviours to support other usage patterns.

By default, program options are recognised before and after subcommands. To only look for program options before subcommands, use .enablePositionalOptions(). This lets you use an option for a different purpose in subcommands.

Example file: positional-options.js

With positional options, the -b is a program option in the first line and a subcommand option in the second line:

1program -b subcommand
2program subcommand -b

By default, options are recognised before and after command-arguments. To only process options that come before the command-arguments, use .passThroughOptions(). This lets you pass the arguments and following options through to another program without needing to use -- to end the option processing. To use pass through options in a subcommand, the program needs to enable positional options.

Example file: pass-through-options.js

With pass through options, the --port=80 is a program option in the first line and passed through as a command-argument in the second line:

1program --port=80 arg
2program arg --port=80

By default, the option processing shows an error for an unknown option. To have an unknown option treated as an ordinary command-argument and continue looking for options, use .allowUnknownOption(). This lets you mix known and unknown options.

By default, the argument processing displays an error for more command-arguments than expected. To suppress the error for excess arguments, use.allowExcessArguments().

Legacy options as properties

Before Commander 7, the option values were stored as properties on the command. This was convenient to code, but the downside was possible clashes with existing properties of Command. You can revert to the old behaviour to run unmodified legacy code by using .storeOptionsAsProperties().

1program
2  .storeOptionsAsProperties()
3  .option('-d, --debug')
4  .action((commandAndOptions) => {
5    if (commandAndOptions.debug) {
6      console.error(`Called ${commandAndOptions.name()}`);
7    }
8  });

TypeScript

extra-typings: There is an optional project to infer extra type information from the option and argument definitions. This adds strong typing to the options returned by .opts() and the parameters to .action(). See commander-js/extra-typings for more.

import { Command } from '@commander-js/extra-typings';

ts-node: If you use ts-node and stand-alone executable subcommands written as .ts files, you need to call your program through node to get the subcommands called correctly. e.g.

1node -r ts-node/register pm.ts

createCommand()

This factory function creates a new command. It is exported and may be used instead of using new, like:

1const { createCommand } = require('commander');
2const program = createCommand();

createCommand is also a method of the Command object, and creates a new command rather than a subcommand. This gets used internally when creating subcommands using .command(), and you may override it to customise the new subcommand (example file custom-command-class.js).

Node options such as --harmony

You can enable --harmony option in two ways:

• Use #! /usr/bin/env node --harmony in the subcommands scripts. (Note Windows does not support this pattern.)
• Use the --harmony option when call the command, like node --harmony examples/pm publish. The --harmony option will be preserved when spawning subcommand process.

Debugging stand-alone executable subcommands

An executable subcommand is launched as a separate child process.

If you are using the node inspector for debugging executable subcommands using node --inspect et al., the inspector port is incremented by 1 for the spawned subcommand.

If you are using VSCode to debug executable subcommands you need to set the "autoAttachChildProcesses": true flag in your launch.json configuration.

npm run-script

By default, when you call your program using run-script, npm will parse any options on the command-line and they will not reach your program. Use -- to stop the npm option parsing and pass through all the arguments.

The synopsis for npm run-script explicitly shows the -- for this reason:

1npm run-script <command> [-- <args>]

Display error

This routine is available to invoke the Commander error handling for your own error conditions. (See also the next section about exit handling.)

As well as the error message, you can optionally specify the exitCode (used with process.exit) and code (used with CommanderError).

1program.error('Password must be longer than four characters');
2program.error('Custom processing has failed', { exitCode: 2, code: 'my.custom.error' });

Override exit and output handling

By default, Commander calls process.exit when it detects errors, or after displaying the help or version. You can override this behaviour and optionally supply a callback. The default override throws a CommanderError.

The override callback is passed a CommanderError with properties exitCode number, code string, and message. Commander expects the callback to terminate the normal program flow, and will call process.exit if the callback returns. The normal display of error messages or version or help is not affected by the override which is called after the display.

1program.exitOverride();
2
3try {
4  program.parse(process.argv);
5} catch (err) {
6  // custom processing...
7}

By default, Commander is configured for a command-line application and writes to stdout and stderr. You can modify this behaviour for custom applications. In addition, you can modify the display of error messages.

Example file: configure-output.js

1function errorColor(str) {
2  // Add ANSI escape codes to display text in red.
3  return `\x1b[31m${str}\x1b[0m`;
4}
5
6program
7  .configureOutput({
8    // Visibly override write routines as example!
9    writeOut: (str) => process.stdout.write(`[OUT] ${str}`),
10    writeErr: (str) => process.stdout.write(`[ERR] ${str}`),
11    // Highlight errors in color.
12    outputError: (str, write) => write(errorColor(str))
13  });

Additional documentation

There is more information available about:

• deprecated features still supported for backwards compatibility
• options taking varying arguments
• parsing life cycle and hooks

Support

The current version of Commander is fully supported on Long Term Support versions of Node.js, and requires at least v20. (For older versions of Node.js, use an older version of Commander.)

The main forum for free and community support is the project Issues on GitHub.

Commander for enterprise

Available as part of the Tidelift Subscription

The maintainers of Commander and thousands of other packages are working with Tidelift to deliver commercial support and maintenance for the open source dependencies you use to build your applications. Save time, reduce risk, and improve code health, while paying the maintainers of the exact dependencies you use. Learn more.