Difference between revisions of "Functor"

From HaskellWiki
Jump to: navigation, search
m (I corrected a grammatical error)
m (I corrected minor grammatical mistakes)
Line 24: Line 24:
 
== Description ==
 
== Description ==
   
An abstract datatype <hask>f a</hask>, which has the ability for its value(s) to be mapped over can become an instance of the <hask>Functor</hask> typeclass. That is to say, a new <hask>Functor</hask>, <hask>f b</hask> can be made from <hask>f a</hask> by transforming all of it's value(s), whilst leaving the structure of <hask>f</hask> itself unmodified.
+
An abstract datatype <hask>f a</hask>, which has the ability for its value(s) to be mapped over can become an instance of the <hask>Functor</hask> typeclass. That is to say, a new <hask>Functor</hask>, <hask>f b</hask> can be made from <hask>f a</hask> by transforming all of its value(s), whilst leaving the structure of <hask>f</hask> itself unmodified.
   
 
Declaring <hask>f</hask> an instance of <hask>Functor</hask> allows functions relating to mapping to be used on structures of type <hask>f a</hask> for all <hask>a</hask>.
 
Declaring <hask>f</hask> an instance of <hask>Functor</hask> allows functions relating to mapping to be used on structures of type <hask>f a</hask> for all <hask>a</hask>.
Line 39: Line 39:
 
:If two sequential mapping operations are performed one after the other using two functions, the result should be the same as a single mapping operation with one function that is equivalent to applying the first function to the result of the second.
 
:If two sequential mapping operations are performed one after the other using two functions, the result should be the same as a single mapping operation with one function that is equivalent to applying the first function to the result of the second.
   
These two laws ensure that functors behave the way they were intended. The values of the functor are only modified by the function provided to the mapping operation. The mapping operation by itself does not modify the values in the functor, only the function. The structure of the functor remains unchanged and only the values are modified. <hask>fmap</hask> returns an identical functor as the original, with it's values swapped to the result of calling a given function with the original value as an argument.
+
These two laws ensure that functors behave the way they were intended. The values of the functor are only modified by the function provided to the mapping operation. The mapping operation by itself does not modify the values in the functor, only the function. The structure of the functor remains unchanged and only the values are modified. <hask>fmap</hask> returns an identical functor as the original, with its values swapped to the result of calling a given function with the original value as an argument.
   
 
== Methods ==
 
== Methods ==

Revision as of 20:49, 10 November 2019

The Functor typeclass represents the mathematical functor: a mapping between categories in the context of category theory. In practice a functor represents a type that can be mapped over.

See also Applicative functor which is a special case of Functor

Packages

  • (base) Prelude
  • (base) Data.Functor
  • (base) Control.Monad

Syntax

class Functor f where
    fmap :: (a -> b) -> f a -> f b
    (<$) :: a -> f b -> f a

Minimal Complete Definition

fmap

Description

An abstract datatype f a, which has the ability for its value(s) to be mapped over can become an instance of the Functor typeclass. That is to say, a new Functor, f b can be made from f a by transforming all of its value(s), whilst leaving the structure of f itself unmodified.

Declaring f an instance of Functor allows functions relating to mapping to be used on structures of type f a for all a.

Functors are required to obey certain laws in regards to their mapping. Ensuring instances of Functor obey these laws means the behaviour of fmap remains predictable.

Functor Laws

Functors must preserve identity morphisms
fmap id = id
When performing the mapping operation, if the values in the functor are mapped to themselves, the result will be an unmodified functor.
Functors preserve composition of morphisms
fmap (f . g)  ==  fmap f . fmap g
If two sequential mapping operations are performed one after the other using two functions, the result should be the same as a single mapping operation with one function that is equivalent to applying the first function to the result of the second.

These two laws ensure that functors behave the way they were intended. The values of the functor are only modified by the function provided to the mapping operation. The mapping operation by itself does not modify the values in the functor, only the function. The structure of the functor remains unchanged and only the values are modified. fmap returns an identical functor as the original, with its values swapped to the result of calling a given function with the original value as an argument.

Methods

fmap :: (a -> b) -> f a -> f b

Create a new f b, from an f a using the results of calling a function on every value in the f a.
(<$) :: a -> f b -> f a
Create a new f a, from an f b by replacing all of the values in the f b by a given value of type a.

Related Functions

($>) :: f a -> b -> f b
Create a new f b, from an f a by replacing all of the values in the f a by a given value of type b.
(<$>) :: (a -> b) -> f a -> f b
An infix synonym for Data.Functor.fmap

void :: Functor f => f a -> f ()

create a new f () from an f a by replacing all of the values in the f a by ()

Further reading