mtl (original) (raw)

mtl: Monad classes for transformers, using functional dependencies

MTL is a collection of monad classes, extending the transformerspackage, using functional dependencies for generic lifting of monadic actions.

[Skip to Readme]

Modules

[Index] [Quick Jump]

• Control
  • Monad
    * Control.Monad.Accum
    * Control.Monad.Cont
    * Control.Monad.Cont.Class
    * Error
    * Control.Monad.Error.Class
    * Control.Monad.Except
    * Control.Monad.Identity
    * Control.Monad.RWS
    * Control.Monad.RWS.CPS
    * Control.Monad.RWS.Class
    * Control.Monad.RWS.Lazy
    * Control.Monad.RWS.Strict
    * Control.Monad.Reader
    * Control.Monad.Reader.Class
    * Control.Monad.Select
    * Control.Monad.State
    * Control.Monad.State.Class
    * Control.Monad.State.Lazy
    * Control.Monad.State.Strict
    * Control.Monad.Trans
    * Control.Monad.Writer
    * Control.Monad.Writer.CPS
    * Control.Monad.Writer.Class
    * Control.Monad.Writer.Lazy
    * Control.Monad.Writer.Strict

Readme for mtl-2.3.1

[back to package description]

MTL is a collection of monad classes, extending the transformerspackage, using functional dependencies for generic lifting of monadic actions.

Structure

Transformers in MTL are divided into classes and data types. Classes define the monadic operations of transformers. Data types, generally from the transformers package, implement transformers, and MTL provides instances for all the transformer type classes.

MTL and transformers use a common module, data type, and function naming scheme. As an example, let's imagine we have a transformerFoo.

In the Control.Monad.Foo module, we'd find:

• A type class MonadFoo with the transformer operations.
• A data type FooT with instances for all monad transformer classes.
• Functions to run the transformed computation, e.g. runFooT. For the actual transformers, there are usually a number of useful runner functions.

Lifting

When using monad transformers, you often need to "lift" a monadic action into your transformed monadic action. This is done using thelift function from MonadTrans in the Control.Monad.Trans.Classmodule:

lift :: (Monad m, MonadTrans t) => m a -> t m a

The action m a is lifted into the transformer action t m a.

As an example, here we lift an action of type IO a into an action of type ExceptT MyError IO a:

data MyError = EmptyLine

mightFail :: ExceptT MyError IO ()
mightFail = do
  l <- lift getLine
  when (null l) (throwError EmptyLine)

Transformers

The following outlines the available monad classes and transformers in MTL and transformers. For more details, and the corresponding documentation of the mtl version you are using, see the documentation on Hackage.

• Control.Monad.Cont
  The Continuation monad transformer adds the ability to usecontinuation-passing style (CPS)in a monadic computation. Continuations can be used to manipulate the control flow of a program, e.g. early exit, error handling, or suspending a computation.
  • Class: Control.Monad.Cont.Class.MonadCont
  • Transformer: Control.Monad.Cont.ContT
• Control.Monad.Error (deprecated!)
  The Error monad transformer has been deprecated in favor ofControl.Monad.Except.
• Control.Monad.Except
  The Except monad transformer adds the ability to fail with an error in a monadic computation.
  • Class: Control.Monad.Except.Class.MonadError
  • Transformer: Control.Monad.Except.ExceptT
• Control.Monad.Identity
  The Identity monad transformer does not add any abilities to a monad. It simply applies the bound function to its inner monad without any modification.
  • Transformer: Control.Monad.Trans.Identity.IdentityT (in the transformers package)
  • Identity functor and monad: Data.Functor.Identity.Identity (in the base package)
• Control.Monad.RWS
  A convenient transformer that combines the Reader, Writer, and State monad transformers.
  • Lazy transformer: Control.Monad.RWS.Lazy.RWST (which is the default, exported by Control.Monad.RWS)
  • Strict transformer: Control.Monad.RWS.Strict.RWST
• Control.Monad.Reader
  The Reader monad transformer represents a computation which can read values from an environment.
  • Class: Control.Monad.Reader.Class.MonadReader
  • Transformer: Control.Monad.Reader.ReaderT
• Control.Monad.State
  The State monad transformer represents a computation which can read and write internal state values. If you only need to _read_values, you might want to useReaderinstead.
  • Class: Control.Monad.State.Class.MonadState
  • Lazy transformer: Control.Monad.State.Lazy.StateT (the default, exported by Control.Monad.State)
  • Strict transformer: Control.Monad.State.Strict.StateT
• Control.Monad.Writer
  The Writer monad transformer represents a computation that can produce a stream of data in addition to the computed values. This can be used to collect values in some data structure with aMonoid instance. This can be used for things like logging and accumulating values throughout a computation.
  • Class: Control.Monad.Writer.Class.MonadWriter
  • Lazy transformers: Control.Monad.Writer.Lazy.WriterT
  • Strict transformers: Control.Monad.Writer.Strict.WriterT
• Control.Monad.Accum
  The Accum monad transformer represents a computation which manages append-only state, or a writer that can read all previous inputs. It binds a function to a monadic value by lazily accumulating subcomputations via (<>). For more general access, use State instead.
  • Class: Control.Monad.Accum
  • Transformer: Control.Monad.Trans.Accum.AccumT
• Control.Monad.Select
  The Select monad transformer represents a computation which can do backtracking search using a 'ranked' evaluation strategy. Binding a function to a monad value chains together evaluation strategies in the sense that the results of previous strategies may influence subsequent rank and evaluation strategies in subcomputations.
  • Class: Control.Monad.Select
  • Transformer: Control.Monad.Trans.Select.SelectT

Resources

• mtl on Hackage
• The Monad Transformerschapter in "What I Wish I Knew When Learning Haskell".
• References:
  • This package is inspired by the paper Functional Programming with Overloading and Higher-Order Polymorphism, by Mark P Jones, in Advanced School of Functional Programming, 1995 (http://web.cecs.pdx.edu/~mpj/pubs/springschool.html).