Difference between revisions of "Keywords"
(Added subsection "rec" and links; minor edits) 
(fix link to user's guide) 

(89 intermediate revisions by 23 users not shown)  
Line 1:  Line 1:  
This page lists all Haskell keywords, feel free to edit. [[Hoogle]] searches will return results from this page. Please respect the Anchor macros. 
This page lists all Haskell keywords, feel free to edit. [[Hoogle]] searches will return results from this page. Please respect the Anchor macros. 

−  For additional information you might want to look at [http://www.haskell.org/onlinereport/ the Haskell 
+  For additional information you might want to look at [http://www.haskell.org/onlinereport/haskell2010/ the Haskell 2010 report]. 
−  == 
+  == ! == 
+  Whenever a data [[constructor]] is applied, each argument to the 

−  The "pipe" is used in several places 

+  constructor is evaluated if and only if the corresponding type in the 

+  algebraic data[[type]] declaration has a strictness flag, denoted by an 

+  exclamation point. For example: 

+  <haskell> 

−  Data type definitions, "or" 

+  data STList a 

+  = STCons a !(STList a)  the second argument to STCons will be 

+   evaluated before STCons is applied 

+   STNil 

+  </haskell> 

+  
+  to illustrate the difference between strict versus lazy constructor 

+  application, consider the following: 

<haskell> 
<haskell> 

+  stList = STCons 1 undefined 

−  data Maybe a = Just a  Nothing 

+  lzList = (:) 1 undefined 

+  stHead (STCons h _) = h  this evaluates to undefined when applied to stList 

+  lzHead (h : _) = h  this evaluates to 1 when applied to lzList 

</haskell> 
</haskell> 

+  ! is also used in the [https://downloads.haskell.org/~ghc/7.8.4/docs/html/users_guide/bangpatterns.html "bang patterns"] (GHC extension), to indicate 

−  List comprehensions, "where" 

+  strictness in patterns: 

<haskell> 
<haskell> 

−  +  f !x !y = x + y 

</haskell> 
</haskell> 

+  == ' == 

−  Guards, "when" 

+  * Character literal: <hask>'a'</hask> 

+  * [[Template Haskell]]: Name of a (value) variable or data constructor: <hask>'length</hask>, <hask>'Left</hask> 

+  * (in types, GHC specific) Promoted data constructor: <hask>'True</hask> 

+  
+  == <nowiki>''</nowiki> == 

+  * [[Template Haskell]]: Name of a type constructor or class: <hask>''Int</hask>, <hask>''Either</hask>, <hask>''Show</hask> 

+  
+  ==  == 

+  This operator token is magic/irregular in the sense that 

+  <haskell>( 1)</haskell> 

+  is parsed as the negative integer 1, rather than as an operator [[Section of an infix operatorsection]], as it would be for any other operator: 

+  <haskell>(* 1) :: Num a => a > a</haskell> 

+  <haskell>(++ "foo") :: String > String</haskell> 

+  
+  It is syntactic sugar for the <hask>negate</hask> function in Prelude. See [[unary operator]]. 

+  If you want the section, you can use the <hask>subtract</hask> function or <hask>(+(1))</hask>. 

+  
+  ==  == 

+  
+  Starts a singleline comment, unless immediately followed by an operator character other than <hask></hask>: 

<haskell> 
<haskell> 

+  main = print "hello world"  this is a comment 

−  safeTail x  null x = [] 

+  this is a comment as well 

−   otherwise = tail x 

+  this too 

+  foobar + this_is_the_second_argument_of_the_dash_dash_plus_operator 

</haskell> 
</haskell> 

+  
+  The multiline variant for comments is <hask>{ comment }</hask>. 

+  
+  == < == 

+  [http://downloads.haskell.org/~ghc/latest/docs/html/users_guide/glasgow_exts.html#arrownotation Arrow notation] 

+  
+  == << == 

+  [http://downloads.haskell.org/~ghc/latest/docs/html/users_guide/glasgow_exts.html#arrownotation Arrow notation] 

== > == 
== > == 

−  The function type constructor: 
+  * The function type constructor: 
<haskell> 
<haskell> 

length :: [a] > Int 
length :: [a] > Int 

</haskell> 
</haskell> 

+  * In lambda functions: 

−  In lambdas: 

<haskell> 
<haskell> 

\x > x + 1 
\x > x + 1 

</haskell> 
</haskell> 

−  To denote alternatives in case statements: 
+  * To denote alternatives in case statements: 
<haskell> 
<haskell> 

case Just 3 of 
case Just 3 of 

Line 45:  Line 90:  
</haskell> 
</haskell> 

+  or with LambdaCase: 

−  And on the kind level (GHC specific): 

<haskell> 
<haskell> 

−  ( 
+  (\ case 1 > 0 
+  ; _ > 1 ) 

</haskell> 
</haskell> 

+  
+  or with MultiWayIf: 

+  <haskell> 

+  if  1 == 0 > 1 

+   1 == 2 > 2 

+   otherwise > 3 

+  </haskell> 

+  
+  * On the kind level (GHC specific): 

+  <haskell> 

+  ghci> :kind (>) 

+  (>) :: * > * > * 

+  </haskell> 

+  
+  * [[Functional dependencies]] 

+  <haskell> 

+   This examples assumes that each type 'c' can "contain" only one type 

+   i.e. type 'c' uniquely determines type 'elt' 

+  class Contains c elt  c > elt where 

+  ... 

+  </haskell> 

+  
+  * [[View patterns]] 

+  
+  == :: == 

+  
+  Read as "has type": 

+  
+  <haskell> 

+  length :: [a] > Int 

+  </haskell> 

+  
+  "Length has type listof'a' to Int" 

+  
+  Or "has kind" (GHC specific): 

+  
+  <haskell> 

+  Either :: * > * > * 

+  </haskell> 

+  
+  == ; == 

+  * Statement separator in an explicit block (see [[layout]]) 

== < == 
== < == 

−  In donotation, "draw from": 
+  * In donotation, "draw from": 
<haskell> 
<haskell> 

do x < getChar 
do x < getChar 

Line 58:  Line 146:  
</haskell> 
</haskell> 

−  In list comprehension generators, " 
+  * In list comprehension generators, "in": 
<haskell> 
<haskell> 

[ (x,y)  x < [1..10], y < ['a'..'z'] ] 
[ (x,y)  x < [1..10], y < ['a'..'z'] ] 

</haskell> 
</haskell> 

−  In [[pattern guard 
+  * In [[pattern guard]]s, "matches": 
<haskell> 
<haskell> 

f x y  Just z < g x = True 
f x y  Just z < g x = True 

Line 69:  Line 157:  
</haskell> 
</haskell> 

−  == 
+  == , == 
+  
+  Separator in lists, tuples, records. 

−  Patterns of the form var@pat are called aspatterns, and allow one to 

−  use var as a name for the value being matched by pat. For example: 

<haskell> 
<haskell> 

+  [1,2,3] 

−  case e of { xs@(x:rest) > if x==0 then rest else xs } 

+  (1,2,3) 

+  Point {x = 1, y = 2} 

</haskell> 
</haskell> 

+  In list comprehensions before generators, "and" (the first comma after <code></code>): 

−  is equivalent to: 

+  <haskell> 

+  [ (x,y)  x < [1..10], y < ['a'..'z'], x > 42 ] 

+  </haskell> 

+  In list comprehensions before Boolean tests, "when" (the second comma after <code></code>): 

<haskell> 
<haskell> 

+  [ (x,y)  x < [1..10], y < ['a'..'z'], x > 42 ] 

−  let { xs = e } in 

−  case xs of { (x:rest) > if x==0 then rest else xs } 

</haskell> 
</haskell> 

+  In guards inside case expressions, "and when": 

−  == ! == 

+  <haskell> 

+  case [1,3,9] of xs  (x:ys) < xs, (y:_) < ys, let z=x+1, z /= y > [x,y,z] 

+  </haskell> 

+  In module import and export lists: 

−  Whenever a data [[constructor]] is applied, each argument to the 

−  constructor is evaluated if and only if the corresponding type in the 

−  algebraic data[[type]] declaration has a strictness flag, denoted by an 

−  exclamation point. For example: 

<haskell> 
<haskell> 

+  module MyModule 

−  data STList a 

+  ( MyData (C1,C2) 

−  = STCons a !(STList a)  the second argument to STCons will be 

+  , myFun ) where 

−   evaluated before STCons is applied 

+  
−   STNil 

+  import MyModule (MyData (C1,C2), myFun) 

</haskell> 
</haskell> 

+  == = == 

−  to illustrate the difference between strict versus lazy constructor 

+  Used in definitions. 

−  application, consider the following: 

<haskell> 
<haskell> 

+  x = 4 

−  stList = STCons 1 undefined 

−  lzList = (:) 1 undefined 

−  stHead (STCons h _) = h  this evaluates to undefined when applied to stList 

−  lzHead (h : _) = h  this evaluates to 1 when applied to lzList 

</haskell> 
</haskell> 

+  Also in patternmatching records: 

−  ! is also used in the [http://www.haskell.org/ghc/docs/latest/html/users_guide/bangpatterns.html "bang patterns"] (GHC extension), to indicate 

−  strictness in patterns: 

<haskell> 
<haskell> 

+  case point of 

−  f !x !y = x + y 

+  Point {x = x0, y = y0} > f x0 y0 

</haskell> 
</haskell> 

+  == => == 

−  Finally, it is the array subscript operator: 

+  
+  Used to indicate instance contexts, for example: 

+  
<haskell> 
<haskell> 

−  +  sort :: Ord a => [a] > [a] 

</haskell> 
</haskell> 

−  == 
+  == > == 
+  In a Bird's style [[Literate_programmingLiterate Haskell file]], the > character is used to introduce a code line. 

−  Read as "has type": 

<haskell> 
<haskell> 

+  comment line 

−  length :: [a] > Int 

+  
+  > main = print "hello world" 

+  </haskell> 

+  
+  == ? == 

+  
+  * {{GHCUsersGuideexts/implicit_parametersa section about Implicit Parameters}} 

+  
+  <haskell> 

+  ghci> :t ?foo ++ "bar" 

+  ?foo ++ "bar" :: (?foo::[Char]) => [Char] 

+  </haskell> 

+  
+  == # == 

+  
+  * [https://downloads.haskell.org/~ghc/7.6.2/docs/html/users_guide/syntaxextns.html MagicHash] 

+  
+  == * == 

+  
+  * Is an ordinary operator name on the value level 

+  
+  * On the [[kind]] level: The kind of boxed types (GHCspecific) 

+  
+  <haskell> 

+  ghci> :kind Int 

+  Int :: * 

+  </haskell> 

+  
+  == @ == 

+  
+  * Patterns of the form <hask>var@pat</hask> are called aspatterns, and allow one to use <hask>var</hask> as a name for the value being matched by <hask>pat</hask>. For example: 

+  <haskell> 

+  case e of { xs@(x:rest) > if x==0 then rest else xs } 

+  
+   is equivalent to: 

+  
+  let { xs = e } in 

+  case xs of { (x:rest) > if x==0 then rest else xs } 

+  </haskell> 

+  
+  * [https://ghc.gitlab.haskell.org/ghc/doc/users_guide/exts/type_applications.html#visibletypeapplication Visible type applications] 

+  
+  == [, ] == 

+  * [[Template Haskell]] 

+  ** Expression quotation: <hask> [ print 1 ] </hask> 

+  ** Declaration quotation: <hask> [d main = print 1 ] </hask> 

+  ** Type quotation: <hask> [t Either Int () ] </hask> 

+  ** Pattern quotation: <hask> [p (x,y) ] </hask> 

+  ** [[Quasiquotation]]: <hask> [nameOfQuasiQuoter ... ] </hask> 

+  
+  == \ == 

+  The backslash "\" is used 

+  
+  * in multiline strings 

+  <haskell> 

+  "foo\ 

+  \bar" 

+  </haskell> 

+  
+  * in lambda functions 

+  <haskell> 

+  \x > x + 1 

</haskell> 
</haskell> 

−  "Length has type listof'a' to Int" 

== _ == 
== _ == 

Line 144:  Line 299:  
<haskell> 
<haskell> 

case e of { [x,y,z] > if x==0 then True else False } 
case e of { [x,y,z] > if x==0 then True else False } 

+  </haskell> 

+  
+  
+  
+  
+  
+  == ` == 

+  
+  A function enclosed in back ticks "`" can be used as an infix operator. 

+  
+  <haskell>2 `subtract` 10</haskell> 

+  is the same as 

+  <haskell>subtract 2 10</haskell> 

+  
+  == {, } == 

+  * Explicit block (disable [[layout]]), possibly with ";" . 

+  
+  * Record update notation 

+  <haskell> 

+  changePrice :: Thing > Price > Thing 

+  changePrice x new = x { price = new } 

+  </haskell> 

+  
+  * Comments (see below) 

+  
+  == {, } == 

+  
+  Everything between "{" followed by a space and "}" is a block comment. 

+  
+  <haskell> 

+  { 

+  hello 

+  world 

+  } 

+  </haskell> 

+  
+  ==  == 

+  
+  The "pipe" is used in several places 

+  
+  * Data type definitions, "or" 

+  <haskell> 

+  data Maybe a = Just a  Nothing 

+  </haskell> 

+  
+  * List comprehensions, "for" (as in, "list of <code>a*a</code> for <code>a</code> in <code>[1..]</code>) 

+  <haskell> 

+  squares = [a*a  a < [1..]] 

+  </haskell> 

+  
+  * Guards, "when" 

+  <haskell> 

+  safeTail x  null x = [] 

+   otherwise = tail x 

+  </haskell> 

+  
+  * [[Functional dependencies]], "where" 

+  <haskell> 

+  class Contains c elt  c > elt where 

+  ... 

</haskell> 
</haskell> 

== ~ == 
== ~ == 

−  Lazy pattern bindings. Matching the pattern ~pat against a value always 
+  * Lazy pattern bindings. Matching the pattern <hask>~pat</hask> against a value always succeeds, and matching will only diverge when one of the variables bound in the pattern is used. 
−  suceeds, and matching will only diverge when one of the variables bound 

−  in the pattern is used. 

<haskell> 
<haskell> 

−  f1,f2 :: Maybe Int > String 
+  f1, f2 :: Maybe Int > String 
+  f1 x = case x of 

−  (+++),(++++) :: (a>b)>(c>d)>(a,c)>(b,d); 

+  Just n > "Got it" 

−  
−  +  f2 x = case x of 

−  +  ~(Just n) > "Got it" 

+  
−  
−  ( 
+  (+++), (++++) :: (a > b) > (c > d) > (a, c) > (b, d) 
−  (f 
+  (f +++ g) ~(x, y) = (f x, g y) 
+  (f ++++ g) (x, y) = (f x, g y) 

</haskell> 
</haskell> 

Line 181:  Line 395:  
For more details see [http://en.wikibooks.org/wiki/Haskell/Laziness#Lazy_pattern_matching the Haskell Wikibook]. 
For more details see [http://en.wikibooks.org/wiki/Haskell/Laziness#Lazy_pattern_matching the Haskell Wikibook]. 

+  * Equality constraints. Assert that two types in a context must be the same: 

−  ==  == 

−  
−  Line comment character. Everything after <hask></hask> on a line is ignored. 

<haskell> 
<haskell> 

+  example :: F a ~ b => a > b 

−  main = print "hello world"  comment here 

</haskell> 
</haskell> 

+  Here the type "F a" must be the same as the type "b", which allows one to constrain polymorphism (especially where type families are involved), but to a lesser extent than functional dependencies. See [[Type_families#Equality_constraintsType Families]]. 

−  The multiline variant of comments is <hask>{ comment }</hask>. 

−  
−  == > == 

−  
−  In a Bird's style [[Literate_programmingLiterate Haskell file]], the > character is used to introduce a code line. 

−  
−  <haskell> 

−  comment line 

−  
−  > main = print "hello world" 

−  </haskell> 

== as == 
== as == 

Line 219:  Line 421:  
</haskell> 
</haskell> 

−  where each <code>match</code> is of the general form 
+  where each <code>match</code><sub>i</sub> is of the general form 
<haskell> 
<haskell> 

−  +   g1 > e1 

−  +  ... 

−  +   gm > em 

−  +  where decls 

</haskell> 
</haskell> 

Each alternative consists of patterns <code>p</code><sub>i</sub> and their matches, <code>match</code><sub>i</sub>. Each 
Each alternative consists of patterns <code>p</code><sub>i</sub> and their matches, <code>match</code><sub>i</sub>. Each 

−  match in turn consists of a sequence of pairs of guards <code>g</code><sub> 
+  <code>match</code><sub>i</sub> in turn consists of a sequence of pairs of guards <code>g</code><sub>ij</sub> and bodies <code>e</code><sub>ij</sub> 
−  (expressions), followed by optional bindings (<code>decls</code>) that scope over all 
+  (expressions), followed by optional bindings (<code>decls</code><sub>i</sub>) that scope over all 
of the guards and expressions of the alternative. An alternative of the 
of the guards and expressions of the alternative. An alternative of the 

form 
form 

Line 255:  Line 457:  
top to bottom, in the environment of the case expression extended first 
top to bottom, in the environment of the case expression extended first 

by the bindings created during the matching of the pattern, and then by 
by the bindings created during the matching of the pattern, and then by 

−  the <code>decls</code> 
+  the <code>decls</code><sub>i</sub> in the <code>where</code> clause associated with that alternative. If one 
of the guards evaluates to <code>True</code>, the corresponding righthand side is 
of the guards evaluates to <code>True</code>, the corresponding righthand side is 

evaluated in the same environment as the guard. If all the guards 
evaluated in the same environment as the guard. If all the guards 

Line 295:  Line 497:  
See the page on [[type]]s for more information, links and examples. 
See the page on [[type]]s for more information, links and examples. 

+  
+  == data family == 

+  Declares a datatype family (see [[type families]]). GHC language extension. 

+  
+  == data instance == 

+  Declares a datatype family instance (see [[type families]]). GHC language extension. 

+  
== default == 
== default == 

Line 329:  Line 538:  
 C 
 C 

deriving (Eq, Ord, Show) 
deriving (Eq, Ord, Show) 

+  </haskell> 

+  
+  In the case of newtypes, GHC extends this mechanism to [[Cunning Newtype Deriving]]. 

+  
+  == deriving instance == 

+  
+  Standalone deriving (GHC language extension). 

+  
+  <haskell> 

+  {# LANGUAGE StandaloneDeriving #} 

+  data A = A 

+  
+  deriving instance Show A 

</haskell> 
</haskell> 

Line 349:  Line 571:  
== forall == 
== forall == 

−  This is a GHC/Hugs extension, and as such is not portable Haskell 98. 
+  This is a GHC/Hugs extension, and as such is not portable Haskell 98/2010. 
It is only a reserved word within types. 
It is only a reserved word within types. 

Type variables in a Haskell type expression are all assumed to be 
Type variables in a Haskell type expression are all assumed to be 

universally quantified; there is no explicit syntax for universal 
universally quantified; there is no explicit syntax for universal 

−  quantification, in standard Haskell 98. For example, the type expression 
+  quantification, in standard Haskell 98/2010. For example, the type expression 
<hask>a > a</hask> denotes the type <hask>forall a. a >a</hask>. 
<hask>a > a</hask> denotes the type <hask>forall a. a >a</hask>. 

For clarity, however, we often write quantification explicitly when 
For clarity, however, we often write quantification explicitly when 

Line 368:  Line 590:  
</haskell> 
</haskell> 

−  GHC [ 
+  GHC [https://downloads.haskell.org/~ghc/latest/docs/html/users_guide/exts/poly_kinds.html#inferringdependencyinuserwrittenforalls introduces] a <hask>forall</hask> keyword, allowing explicit quantification, for example, to encode 
−  [http:// 
+  [http://downloads.haskell.org/~ghc/latest/docs/html/users_guide/exts/existential_quantification.html existential types]: 
<haskell> 
<haskell> 

data Foo = forall a. MkFoo a (a > Bool) 
data Foo = forall a. MkFoo a (a > Bool) 

Line 382:  Line 604:  
== foreign == 
== foreign == 

+  A keyword for the [[Foreign Function Interface]] (commonly called the FFI) that introduces either a <hask>foreign import</hask> declaration, which makes a function from a nonHaskell library available in a Haskell program, or a <hask>foreign export</hask> declaration, which allows a function from a Haskell module to be called in nonHaskell contexts. 

−  A keyword for the foreign function interface that is enabled by ffi, fffi 

−  or implied by fglasgowexts 

== hiding == 
== hiding == 

Line 440:  Line 661:  
the fixity and binding precedence of one or more operators. The integer 
the fixity and binding precedence of one or more operators. The integer 

in a fixity declaration must be in the range 0 to 9. A fixity 
in a fixity declaration must be in the range 0 to 9. A fixity 

−  declaration may appear anywhere that a type signature appears and, like 
+  declaration may appear anywhere that a [[type signature]] appears and, like 
a type signature, declares a property of a particular operator. 
a type signature, declares a property of a particular operator. 

Line 472:  Line 693:  
<haskell>let { d1 ; ... ; dn } in e</haskell> 
<haskell>let { d1 ; ... ; dn } in e</haskell> 

−  They introduce a nested, lexicallyscoped, mutuallyrecursive list of declarations (let is often called 
+  They introduce a nested, lexicallyscoped, mutuallyrecursive list of declarations (let is often called [https://en.wikipedia.org/wiki/Let_expression let rec] in other languages). The scope of the declarations is the expression e and the right hand side of the declarations. 
−  Within <hask>do</hask>blocks or list comprehensions <hask>let { d1 ; ... ; dn }</hask> without <hask>in</hask> serves to 
+  Within <hask>do</hask>blocks or list comprehensions <hask>let { d1 ; ... ; dn }</hask> without <hask>in</hask> serves to introduce local bindings. 
−  == [ 
+  == [https://downloads.haskell.org/ghc/9.4.4/docs/users_guide/exts/recursive_do.html#themdonotation mdo] == 
−  The recursive <hask>do</hask> keyword enabled by fglasgowexts 
+  The recursive <hask>do</hask> keyword enabled by fglasgowexts. 
== module == 
== module == 

Line 504:  Line 725:  
newtype FunctionIdentifier = FunctionIdentifier Identifier 
newtype FunctionIdentifier = FunctionIdentifier Identifier 

−  Most often, one supplies [[smart constructors]] and [[ 
+  Most often, one supplies [[smart constructors]] and [[smart destructorsdestructors]] for these to ease working with them. 
See the page on [[type]]s for more information, links and examples. 
See the page on [[type]]s for more information, links and examples. 

For the differences between <code>newtype</code> and <code>data</code>, see [[Newtype]]. 
For the differences between <code>newtype</code> and <code>data</code>, see [[Newtype]]. 

+  
+  == proc == 

+  proc (arrow abstraction) 

+  is a kind of lambda, except that it constructs an arrow instead of a function. 

+  
+  [[Arrow notation]] 

== qualified == 
== qualified == 

Line 531:  Line 758:  
== rec == 
== rec == 

−  The [ 
+  The [https://downloads.haskell.org/ghc/9.4.4/docs/users_guide/exts/recursive_do.html#themdonotation rec] keyword can be used when the <code>XDoRec</code> flag is given; it allows recursive bindings in a doblock. 
<haskell> 
<haskell> 

Line 562:  Line 789:  
type ShowS = String > String 
type ShowS = String > String 

</haskell> 
</haskell> 

+  
+  == type family == 

+  Declares a type synonym family (see [[type families]]). GHC language extension. 

+  
+  == type instance == 

+  Declares a type synonym family instance (see [[type families]]). GHC language extension. 

+  
+  
== where == 
== where == 

−  Used to introduce a module, instance or 
+  Used to introduce a module, instance, class or [[GADT]]: 
<haskell> 
<haskell> 

module Main where 
module Main where 

Line 574:  Line 809:  
instance Num Int where 
instance Num Int where 

... 
... 

+  
+  data Something a where 

+  ... 

</haskell> 
</haskell> 

Latest revision as of 17:19, 4 February 2023
This page lists all Haskell keywords, feel free to edit. Hoogle searches will return results from this page. Please respect the Anchor macros.
For additional information you might want to look at the Haskell 2010 report.
!
Whenever a data constructor is applied, each argument to the constructor is evaluated if and only if the corresponding type in the algebraic datatype declaration has a strictness flag, denoted by an exclamation point. For example:
data STList a
= STCons a !(STList a)  the second argument to STCons will be
 evaluated before STCons is applied
 STNil
to illustrate the difference between strict versus lazy constructor application, consider the following:
stList = STCons 1 undefined
lzList = (:) 1 undefined
stHead (STCons h _) = h  this evaluates to undefined when applied to stList
lzHead (h : _) = h  this evaluates to 1 when applied to lzList
! is also used in the "bang patterns" (GHC extension), to indicate strictness in patterns:
f !x !y = x + y
'
 Character literal:
'a'
 Template Haskell: Name of a (value) variable or data constructor:
'length
,'Left
 (in types, GHC specific) Promoted data constructor:
'True
''
 Template Haskell: Name of a type constructor or class:
''Int
,''Either
,''Show

This operator token is magic/irregular in the sense that
( 1)
is parsed as the negative integer 1, rather than as an operator section, as it would be for any other operator:
(* 1) :: Num a => a > a
(++ "foo") :: String > String
It is syntactic sugar for the negate
function in Prelude. See unary operator.
If you want the section, you can use the subtract
function or (+(1))
.

Starts a singleline comment, unless immediately followed by an operator character other than 
:
main = print "hello world"  this is a comment
this is a comment as well
this too
foobar + this_is_the_second_argument_of_the_dash_dash_plus_operator
The multiline variant for comments is { comment }
.
<
<<
>
 The function type constructor:
length :: [a] > Int
 In lambda functions:
\x > x + 1
 To denote alternatives in case statements:
case Just 3 of
Nothing > False
Just x > True
or with LambdaCase:
(\ case 1 > 0
; _ > 1 )
or with MultiWayIf:
if  1 == 0 > 1
 1 == 2 > 2
 otherwise > 3
 On the kind level (GHC specific):
ghci> :kind (>)
(>) :: * > * > *
 This examples assumes that each type 'c' can "contain" only one type
 i.e. type 'c' uniquely determines type 'elt'
class Contains c elt  c > elt where
...
::
Read as "has type":
length :: [a] > Int
"Length has type listof'a' to Int"
Or "has kind" (GHC specific):
Either :: * > * > *
;
 Statement separator in an explicit block (see layout)
<
 In donotation, "draw from":
do x < getChar
putChar x
 In list comprehension generators, "in":
[ (x,y)  x < [1..10], y < ['a'..'z'] ]
 In pattern guards, "matches":
f x y  Just z < g x = True
 otherwise = False
,
Separator in lists, tuples, records.
[1,2,3]
(1,2,3)
Point {x = 1, y = 2}
In list comprehensions before generators, "and" (the first comma after 
):
[ (x,y)  x < [1..10], y < ['a'..'z'], x > 42 ]
In list comprehensions before Boolean tests, "when" (the second comma after 
):
[ (x,y)  x < [1..10], y < ['a'..'z'], x > 42 ]
In guards inside case expressions, "and when":
case [1,3,9] of xs  (x:ys) < xs, (y:_) < ys, let z=x+1, z /= y > [x,y,z]
In module import and export lists:
module MyModule
( MyData (C1,C2)
, myFun ) where
import MyModule (MyData (C1,C2), myFun)
=
Used in definitions.
x = 4
Also in patternmatching records:
case point of
Point {x = x0, y = y0} > f x0 y0
=>
Used to indicate instance contexts, for example:
sort :: Ord a => [a] > [a]
>
In a Bird's style Literate Haskell file, the > character is used to introduce a code line.
comment line
> main = print "hello world"
?
ghci> :t ?foo ++ "bar"
?foo ++ "bar" :: (?foo::[Char]) => [Char]
#
*
 Is an ordinary operator name on the value level
 On the kind level: The kind of boxed types (GHCspecific)
ghci> :kind Int
Int :: *
@
 Patterns of the form
var@pat
are called aspatterns, and allow one to usevar
as a name for the value being matched bypat
. For example:
case e of { xs@(x:rest) > if x==0 then rest else xs }
 is equivalent to:
let { xs = e } in
case xs of { (x:rest) > if x==0 then rest else xs }
[, ]
 Template Haskell
 Expression quotation:
[ print 1 ]
 Declaration quotation:
[d main = print 1 ]
 Type quotation:
[t Either Int () ]
 Pattern quotation:
[p (x,y) ]
 Quasiquotation:
[nameOfQuasiQuoter ... ]
 Expression quotation:
\
The backslash "\" is used
 in multiline strings
"foo\
\bar"
 in lambda functions
\x > x + 1
_
Patterns of the form _ are wildcards and are useful when some part of a pattern is not referenced on the righthandside. It is as if an identifier not used elsewhere were put in its place. For example,
case e of { [x,_,_] > if x==0 then True else False }
is equivalent to:
case e of { [x,y,z] > if x==0 then True else False }
`
A function enclosed in back ticks "`" can be used as an infix operator.
2 `subtract` 10
is the same as
subtract 2 10
{, }
 Explicit block (disable layout), possibly with ";" .
 Record update notation
changePrice :: Thing > Price > Thing
changePrice x new = x { price = new }
 Comments (see below)
{, }
Everything between "{" followed by a space and "}" is a block comment.
{
hello
world
}

The "pipe" is used in several places
 Data type definitions, "or"
data Maybe a = Just a  Nothing
 List comprehensions, "for" (as in, "list of
a*a
fora
in[1..]
)
squares = [a*a  a < [1..]]
 Guards, "when"
safeTail x  null x = []
 otherwise = tail x
 Functional dependencies, "where"
class Contains c elt  c > elt where
...
~
 Lazy pattern bindings. Matching the pattern
~pat
against a value always succeeds, and matching will only diverge when one of the variables bound in the pattern is used.
f1, f2 :: Maybe Int > String
f1 x = case x of
Just n > "Got it"
f2 x = case x of
~(Just n) > "Got it"
(+++), (++++) :: (a > b) > (c > d) > (a, c) > (b, d)
(f +++ g) ~(x, y) = (f x, g y)
(f ++++ g) (x, y) = (f x, g y)
Then we have:
f1 Nothing
Exception: Nonexhaustive patterns in case
f2 Nothing
"Got it"
(const 1 +++ const 2) undefined
(1,2)
(const 1 ++++ const 2) undefined
Exception: Prelude.undefined
For more details see the Haskell Wikibook.
 Equality constraints. Assert that two types in a context must be the same:
example :: F a ~ b => a > b
Here the type "F a" must be the same as the type "b", which allows one to constrain polymorphism (especially where type families are involved), but to a lesser extent than functional dependencies. See Type Families.
as
Renaming module imports. Like qualified
and hiding
, as
is not a reserved word but may be used as function or variable name.
import qualified Data.Map as M
main = print (M.empty :: M.Map Int ())
case, of
A case expression has the general form
case e of { p1 match1 ; ... ; pn matchn }
where each match
_{i} is of the general form
 g1 > e1
...
 gm > em
where decls
Each alternative consists of patterns p
_{i} and their matches, match
_{i}. Each
match
_{i} in turn consists of a sequence of pairs of guards g
_{ij} and bodies e
_{ij}
(expressions), followed by optional bindings (decls
_{i}) that scope over all
of the guards and expressions of the alternative. An alternative of the
form
pat > exp where decls
is treated as shorthand for:
pat  True > exp
where decls
A case expression must have at least one alternative and each alternative must have at least one body. Each body must have the same type, and the type of the whole expression is that type.
A case expression is evaluated by pattern matching the expression e
against the individual alternatives. The alternatives are tried
sequentially, from top to bottom. If e
matches the pattern in the
alternative, the guards for that alternative are tried sequentially from
top to bottom, in the environment of the case expression extended first
by the bindings created during the matching of the pattern, and then by
the decls
_{i} in the where
clause associated with that alternative. If one
of the guards evaluates to True
, the corresponding righthand side is
evaluated in the same environment as the guard. If all the guards
evaluate to False
, matching continues with the next alternative. If no
match succeeds, the result is __.
class
A class declaration introduces a new type class and the overloaded operations that must be supported by any type that is an instance of that class.
class Num a where
(+) :: a > a > a
negate :: a > a
data
The data declaration is how one introduces new algebraic data types into Haskell. For example:
data Set a = NilSet
 ConsSet a (Set a)
Another example, to create a datatype to hold an abstract syntax tree for an expression, one could use:
data Exp = Ebin Operator Exp Exp
 Eunary Operator Exp
 Efun FunctionIdentifier [Exp]
 Eid SimpleIdentifier
where the types Operator, FunctionIdentifier
and SimpleIdentifier
are defined elsewhere.
See the page on types for more information, links and examples.
data family
Declares a datatype family (see type families). GHC language extension.
data instance
Declares a datatype family instance (see type families). GHC language extension.
default
Ambiguities in the class Num are most common, so Haskell provides a way to resolve themwith a default declaration:
default (Int)
Only one default declaration is permitted per module, and its effect is limited to that module. If no default declaration is given in a module then it assumed to be:
default (Integer, Double)
deriving
data and newtype declarations contain an optional deriving form. If the form is included, then derived instance declarations are automatically generated for the datatype in each of the named classes.
Derived instances provide convenient commonlyused operations for userdefined datatypes. For example, derived instances for datatypes in the class Eq define the operations == and /=, freeing the programmer from the need to define them.
data T = A
 B
 C
deriving (Eq, Ord, Show)
In the case of newtypes, GHC extends this mechanism to Cunning Newtype Deriving.
deriving instance
Standalone deriving (GHC language extension).
{# LANGUAGE StandaloneDeriving #}
data A = A
deriving instance Show A
do
Syntactic sugar for use with monadic expressions. For example:
do { x ; result < y ; foo result }
is shorthand for:
x >>
y >>= \result >
foo result
forall
This is a GHC/Hugs extension, and as such is not portable Haskell 98/2010. It is only a reserved word within types.
Type variables in a Haskell type expression are all assumed to be
universally quantified; there is no explicit syntax for universal
quantification, in standard Haskell 98/2010. For example, the type expression
a > a
denotes the type forall a. a >a
.
For clarity, however, we often write quantification explicitly when
discussing the types of Haskell programs. When we write an explicitly
quantified type, the scope of the forall extends as far to the right as
possible; for example,
forall a. a > a
means
forall a. (a > a)
GHC introduces a forall
keyword, allowing explicit quantification, for example, to encode
existential types:
data Foo = forall a. MkFoo a (a > Bool)
 Nil
MkFoo :: forall a. a > (a > Bool) > Foo
Nil :: Foo
[MkFoo 3 even, MkFoo 'c' isUpper] :: [Foo]
foreign
A keyword for the Foreign Function Interface (commonly called the FFI) that introduces either a foreign import
declaration, which makes a function from a nonHaskell library available in a Haskell program, or a foreign export
declaration, which allows a function from a Haskell module to be called in nonHaskell contexts.
hiding
When importing modules, without introducing a name into scope, entities can be excluded by using the form
hiding (import1 , ... , importn )
which specifies that all entities exported by the named module should be imported except for those named in the list.
For example:
import Prelude hiding (lookup,filter,foldr,foldl,null,map)
if, then, else
A conditional expression has the form:
if e1 then e2 else e3
and returns the value of e2 if the value of e1 is True, e3 if e1 is False, and __ otherwise.
max a b = if a > b then a else b
import
Modules may reference other modules via explicit import declarations, each giving the name of a module to be imported and specifying its entities to be imported.
For example:
module Main where
import A
import B
main = A.f >> B.f
module A where
f = ...
module B where
f = ...
See also as, hiding , qualified and the page Import
infix, infixl, infixr
A fixity declaration gives the fixity and binding precedence of one or more operators. The integer in a fixity declaration must be in the range 0 to 9. A fixity declaration may appear anywhere that a type signature appears and, like a type signature, declares a property of a particular operator.
There are three kinds of fixity, non, left and rightassociativity (infix, infixl, and infixr, respectively), and ten precedence levels, 0 to 9 inclusive (level 0 binds least tightly, and level 9 binds most tightly).
module Bar where
infixr 7 `op`
op = ...
instance
An instance declaration declares that a type is an instance of a class and includes the definitions of the overloaded operations  called class methods  instantiated on the named type.
instance Num Int where
x + y = addInt x y
negate x = negateInt x
let, in
Let expressions have the general form:
let { d1 ; ... ; dn } in e
They introduce a nested, lexicallyscoped, mutuallyrecursive list of declarations (let is often called let rec in other languages). The scope of the declarations is the expression e and the right hand side of the declarations.
Within do
blocks or list comprehensions let { d1 ; ... ; dn }
without in
serves to introduce local bindings.
mdo
The recursive do
keyword enabled by fglasgowexts.
module
Taken from: A Gentle Introduction to Haskell, Version 98
Technically speaking, a module is really just one big declaration which begins with the keyword module; here's an example for a module whose name is Tree:
module Tree ( Tree(Leaf,Branch), fringe ) where
data Tree a = Leaf a  Branch (Tree a) (Tree a)
fringe :: Tree a > [a]
fringe (Leaf x) = [x]
fringe (Branch left right) = fringe left ++ fringe right
newtype
The newtype
declaration is how one introduces a renaming for an algebraic data type into Haskell. This is different from type
below, as a newtype
requires a new constructor as well. As an example, when writing a compiler
one sometimes further qualifies Identifier
s to assist in type safety checks:
newtype SimpleIdentifier = SimpleIdentifier Identifier newtype FunctionIdentifier = FunctionIdentifier Identifier
Most often, one supplies smart constructors and destructors for these to ease working with them.
See the page on types for more information, links and examples.
For the differences between newtype
and data
, see Newtype.
proc
proc (arrow abstraction) is a kind of lambda, except that it constructs an arrow instead of a function.
qualified
Used to import a module, but not introduce a name into scope. For example, Data.Map exports lookup, which would clash with the Prelude version of lookup, to fix this:
import qualified Data.Map
f x = lookup x  use the Prelude version
g x = Data.Map.lookup x  use the Data.Map version
Of course, Data.Map is a bit of a mouthful, so qualified also allows the use of as.
import qualified Data.Map as M
f x = lookup x  use Prelude version
g x = M.lookup x  use Data.Map version
rec
The rec keyword can be used when the XDoRec
flag is given; it allows recursive bindings in a doblock.
{# LANGUAGE DoRec #}
justOnes = do { rec { xs < Just (1:xs) }
; return (map negate xs) }
type
The type
declaration is how one introduces an alias for an algebraic data type into Haskell. As an example, when writing a compiler
one often creates an alias for identifiers:
type Identifier = String
This allows you to use Identifer
wherever you had used String
and if something is of type Identifier
it
may be used wherever a String
is expected.
See the page on types for more information, links and examples.
Some common type
declarations in the Prelude include:
type FilePath = String
type String = [Char]
type Rational = Ratio Integer
type ReadS a = String > [(a,String)]
type ShowS = String > String
type family
Declares a type synonym family (see type families). GHC language extension.
type instance
Declares a type synonym family instance (see type families). GHC language extension.
where
Used to introduce a module, instance, class or GADT:
module Main where
class Num a where
...
instance Num Int where
...
data Something a where
...
And to bind local variables:
f x = y
where y = x * 2
g z  z > 2 = y
where y = x * 2