MyNixOS website logo
Description

Declarative command line option parser.

Declarative and easy to use command line option parser

optparse-declarative

Hackage GitHub Actions: test Join the chat at https://gitter.im/optparse-declarative/community

optparse-declarative is a declarative and easy-to-use command-line option parser.

Install

$ cabal install optparse-declarative

Usage

Writing a simple command

First, you need to enable DataKinds extension. Then import Options.Declarative module.

{-# LANGUAGE DataKinds #-}
import           Options.Declarative

Next, define command line options as a type of the function. For example, this is a simple greeting program with -g option that takes a message of type String and an unnamed command-line argument that specifies a name:

greet :: Flag "g" '["greet"] "STRING" "greeting message" (Def "Hello" String)
      -> Arg "NAME" String
      -> Cmd "Greeting command" ()
greet msg name =
    liftIO $ putStrLn $ get msg ++ ", " ++ get name ++ "!"

There are two types of options, Flag and Arg. Flag represents a named argument (e.g., --greet "Hola"), and Arg an unnamed argument (e.g., John of greet --greet Hola John). The last argument of Flag and Arg is the type of the value of the argument; in this example, they are both String. You can specify any type for the value as long as the type is an instance of ArgRead typeclass, in which the conversion function from String to the specified type is defined. Options.Declarative provides following instances of ArgRead typeclass.

  • Int
  • Integer
  • Bool
  • Double
  • String
  • (ArgRead a) => Maybe a

Users can add a new instance of ArgRead to support any user-defined type. Please see Section "How to add a new instance of ArgRead" for details.

If you wish to specify a default value for allowing users to omit a value, use the modifier Def with the default value as the second type argument (and the third type argument is the type of the value). You need to specify the default value in String instead of the final value of the target type; the string will be converted to the final value via ArgRead typeclass.

In the example above, the variable msg has a very complex type (Flag "g" '["greet"] "STRING" "greeting message" (Def "Hello" String)). In order to get the value of the target type (in this case, that is String), you can use get function.

The whole type of command is Cmd. Cmd is an instance of MonadIO and it has some extra information.

Finally, you can run the whole program by run_.

main :: IO ()
main = run_ greet

Here is an example session with the program shown above.

$ ghc simple.hs

$ ./simple
simple: not enough arguments
Try 'simple --help' for more information.

$ ./simple --help
Usage: simple [OPTION...] NAME
Options:
  -g STRING  --greet=STRING  greeting message
  -?         --help          display this help and exit

$ ./simple World
Hello, World!

$ ./simple --greet=Goodbye World
Goodbye, World!

Note that only the final option is used when multiple options of the same name are given. This behavior emulates the behavior of a naive program that uses GNU Getopt.

$ ./simple --greet=Hello --greet=Goodbye World
Goodbye, World!

There is another way of interpreting multiple options of the same name. Suppose if you need to get multiple values from the same option. Say, you wish to get ["Hello", "Goodbye"] from the command-line option --greet=Hello --greeet=Goodbye. Then, you can use the type [] to indicate that it accepts multiple values. The first line of the function greet in the example above would be changed as this:

greet :: Flag "g" '["greet"] "STRING" "greeting message" [String]

The value returned by get will be a value of type [String]. See the complete working example at example/list.hs for details.

Writing multiple subcommands

You can write (nested) subcommands. You don't know what subcommands are? Imagine git command. git has subcommands such as git add, git commit, git log, etc. git has nested subcommands such as git remote add, git remote rm, etc. optparse-declarative provides an easy way to provide such possibly nested subcommands.

Just group subcommands by Group, then you get a subcommand parser. Here is an example with two subcommands greet and connect:

{-# LANGUAGE DataKinds #-}

import           Options.Declarative

main :: IO ()
main = run_ $
    Group "Test program for library"
    [ subCmd "greet"   greet
    , subCmd "connect" connect
    ]

greet :: Flag "g" '["greet"] "STRING" "greeting message" (Def "Hello" String)
      -> Flag "" '["decolate"] "" "decolate message" Bool
      -> Arg "NAME" String
      -> Cmd "Greeting command" ()
greet msg deco name = do
    let f x | get deco = "*** " ++ x ++ " ***"
            | otherwise = x
    liftIO $ putStrLn $ f $ get msg ++ ", " ++ get name ++ "!"

connect :: Flag "h" '["host"] "HOST" "host name"   (Def "localhost" String)
        -> Flag "p" '["port"] "PORT" "port number" (Def "8080"      Int   )
        -> Cmd "Connect command" ()
connect host port = do
    let addr = get host ++ ":" ++ show (get port)
    liftIO $ putStrLn $ "connect to " ++ addr

This is a sample session for the program above:

$ ./subcmd --help
Usage: subcmd [OPTION...] <COMMAND> [ARGS...]
Options:
  -?  --help  display this help and exit

Commands:
  greet       Greeting command
  port        Server command

$ ./subcmd connect --port=1234
connect to localhost:1234

If you wish to specify the program name or the version number, use run instead of run_. The first argument of run is a program name (of type String). The second argument is a version number (of type Maybe String).

main :: IO ()
main = run "program_name" (Just "1.3.2") $
    Group "Test program for library"
    [ subCmd "greet"   greet
    , subCmd "connect" connect
    ]

For more examples, please see example directory.

Default options

optparse-declarative provides a few default options. For example, --help is defined automatically so users do not have to write it by their own. If run with run and the version number, --version is defined automatically. Also, --verbosity option (-v for short) is defined by default. getVerbosity returns the verbosity level in Int. -v gives 1, -vv gives 2, -vvv gives 3. Alternatively, --verbose=3 would yield 3.

How to add a new instance of ArgRead

Users need to create an instance of ArgRead for supporting a new type for the command line argument. Here is the definition of class ArgRead.

class ArgRead a where
    -- | Type of the argument
    type Unwrap a :: *
    type Unwrap a = a

    -- | Get the argument's value
    unwrap :: a -> Unwrap a
    default unwrap :: a ~ Unwrap a => a -> Unwrap a
    unwrap = id

    -- | Argument parser
    argRead :: [String] -> Maybe a
    default argRead :: Read a => [String] -> Maybe a
    argRead ss = getLast $ mconcat $ Last . readMaybe <$> ss

    -- | Indicate this argument is mandatory
    needArg :: Proxy a -> Bool
    needArg _ = True

Suppose you are adding a support for your type T. We explain which function to define explicitly, depending on the property of T.

If T is the type of the final value you take out of a command line, you do not have to define Unwrap. If T is a wrapper like Def, define type Unwrap T = <unwrapped type>. For Def x y, type Unwrap (Def x y) = y. If you defined Unwrap, define unwrap that takes an actual value out of the wrapped value.

argRead is the main function that converts String into a value. If the type is an instance of Read and you are satisfied with how read converts a String into value, there is no need to define your own argRead. Otherwise, you define a function that converts a String into a value of the target type. When parsing is successful, return Just. When it fails, return Nothing. If the input is [], it indicates the option does not have an argument; otherwise the input is a list of a single String. Last but not least, define needArg _ = False when the option allows us to omit the associated value; consider a boolean option like --help.

Metadata

Version

0.4.2

License

Platforms (75)

    Darwin
    FreeBSD
    Genode
    GHCJS
    Linux
    MMIXware
    NetBSD
    none
    OpenBSD
    Redox
    Solaris
    WASI
    Windows
Show all
  • aarch64-darwin
  • aarch64-genode
  • aarch64-linux
  • aarch64-netbsd
  • aarch64-none
  • aarch64_be-none
  • arm-none
  • armv5tel-linux
  • armv6l-linux
  • armv6l-netbsd
  • armv6l-none
  • armv7a-darwin
  • armv7a-linux
  • armv7a-netbsd
  • armv7l-linux
  • armv7l-netbsd
  • avr-none
  • i686-cygwin
  • i686-darwin
  • i686-freebsd
  • i686-genode
  • i686-linux
  • i686-netbsd
  • i686-none
  • i686-openbsd
  • i686-windows
  • javascript-ghcjs
  • loongarch64-linux
  • m68k-linux
  • m68k-netbsd
  • m68k-none
  • microblaze-linux
  • microblaze-none
  • microblazeel-linux
  • microblazeel-none
  • mips-linux
  • mips-none
  • mips64-linux
  • mips64-none
  • mips64el-linux
  • mipsel-linux
  • mipsel-netbsd
  • mmix-mmixware
  • msp430-none
  • or1k-none
  • powerpc-netbsd
  • powerpc-none
  • powerpc64-linux
  • powerpc64le-linux
  • powerpcle-none
  • riscv32-linux
  • riscv32-netbsd
  • riscv32-none
  • riscv64-linux
  • riscv64-netbsd
  • riscv64-none
  • rx-none
  • s390-linux
  • s390-none
  • s390x-linux
  • s390x-none
  • vc4-none
  • wasm32-wasi
  • wasm64-wasi
  • x86_64-cygwin
  • x86_64-darwin
  • x86_64-freebsd
  • x86_64-genode
  • x86_64-linux
  • x86_64-netbsd
  • x86_64-none
  • x86_64-openbsd
  • x86_64-redox
  • x86_64-solaris
  • x86_64-windows