Closures/Value capture

Closures/Value capture
You are encouraged to solve this task according to the task description, using any language you may know.

Task

Create a list of ten functions, in the simplest manner possible   (anonymous functions are encouraged),   such that the function at index   i   (you may choose to start   i   from either   0   or   1),   when run, should return the square of the index,   that is,   i ².

Display the result of running any but the last function, to demonstrate that the function indeed remembers its value.

Goal

Demonstrate how to create a series of independent closures based on the same template but maintain separate copies of the variable closed over.

In imperative languages, one would generally use a loop with a mutable counter variable.

For each function to maintain the correct number, it has to capture the value of the variable at the time it was created, rather than just a reference to the variable, which would have a different value by the time the function was run.

See also: Multiple distinct objects

11l

[(() -> Int)] funcs
L(i) 10
   funcs.append(() -> @=i * @=i)
print(funcs[3]())

Output:

9

Acornsoft Lisp

Since this Lisp is dynamically scoped and does not have any built-in closure mechanism, we have to construct one which we'll call freeze. (The name is inspired by the Pop-2 programming languages's "frozen formals".)

(freeze varlist lambda-expr) finds the current values of the variables in varlist and returns a lambda-expression that is like the original except that, when called, it binds those variables to their captured values. For example, if a's value is 1 and b's is 2,

(freeze '(a b) '(lambda (c) (list a b c)))

would return

(lambda (c)
  ((lambda ((a . 1) (b . 2))
     (list a b c))))

What does that mean? A cons (name . value) in a lambda-expressions's formal parameters is the syntax for a formal with a default value. The value is literally the value; it's not an expression that's evaluated. This

( (lambda ((a . 1) (b . 2))
    (list a b c)) )

calls the function represented by that lambda-expression. Since it does not give the function any arguments, a and b get their default values (which are the values captured by freeze).

(Although code within such a 'closure' can assign new values to the captured variables, it would have only a temporary effect and would not change the values seen in subsequent calls to the same closure. That's one sense in which the variable values are "frozen".)

Here is the definition of freeze:

(defun freeze (_fvars_ _lambda-expr_)
  (freeze-vars
    (mapc cons _fvars_ (mapc eval _fvars_))
    (cadr _lambda-expr_)
    (cddr _lambda-expr_)))

(defun freeze-vars (bindings lvars lbody)
  (list 'lambda lvars
        (list (cons 'lambda (cons bindings lbody)))))

Once we have freeze, we can create a list of square-returning functions and then call them:

(defun range (from to)
  (cond ((greaterp from to) '())
        (t (cons from (range (add1 from) to)))))

(defun example ()
  (mapc '(lambda (f) (f))
        (mapc '(lambda (i)
                 (freeze '(i) '(lambda () (times i i))))
              (range 1 10))))

Output:

(example) returns

(1 4 9 16 25 36 49 64 81 100)

Ada

One way to realize closures in Ada is the usage of protected objects.

with Ada.Text_IO;

procedure Value_Capture is
   
   protected type Fun is -- declaration of the type of a protected object
      entry Init(Index: Natural);
      function Result return Natural;
   private
      N: Natural := 0;
   end Fun;
   
   protected body Fun is -- the implementation of a protected object
      entry Init(Index: Natural) when N=0 is
      begin -- after N has been set to a nonzero value, it cannot be changed any more
         N := Index;
      end Init;
      function Result return Natural is (N*N);
   end Fun;
   
   A: array (1 .. 10) of Fun; -- an array holding 10 protected objects
   
begin
   for I in A'Range loop -- initialize the protected objects
      A(I).Init(I);
   end loop;
   
   for I in A'First .. A'Last-1 loop -- evaluate the functions, except for the last
      Ada.Text_IO.Put(Integer'Image(A(I).Result));
   end loop;
end Value_Capture;

Output:

 1 4 9 16 25 36 49 64 81

ALGOL 68

Works with: ALGOL 68G version 2.8

[1:10]PROC(BOOL)INT squares;

FOR i FROM 1 TO 10 DO
        HEAP INT captured i := i;
        squares[i] := ((REF INT by ref i, INT by val i,BOOL b)INT:(INT i = by ref i; (b|by ref i := 0); by val i*i))
                (captured i, captured i,)
OD;

FOR i FROM 1 TO 8 DO print(squares[i](i MOD 2 = 0)) OD;
print(new line);
FOR i FROM 1 TO 10 DO print(squares[i](FALSE)) OD

Output:

         +1         +4         +9        +16        +25        +36        +49        +64
         +1         +0         +9         +0        +25         +0        +49         +0        +81       +100

Using partial parametrization as proposed in Algol Bulletin by Charles Lindsey. Algol68G does not support binding all actual parameters "partially" without deproceduring, so a PROC(BOOL)INT mode is used instead of a PROC INT. The variable captured i is passed twice, once by reference and once by value, to demonstrate that it is possible to capture both ways, and a little extra code is added to show that the closure can modify the captured variable.

AntLang

fns: {n: x; {n expt 2}} map range[10]
(8 elem fns)[]

AppleScript

Translation of: JavaScript

on run
    set fns to {}
    
    repeat with i from 1 to 10
        set end of fns to closure(i)
    end repeat
    
    |λ|() of item 3 of fns
end run

on closure(x)
    script
        on |λ|()
            x * x
        end |λ|
    end script
end closure

Output:

9

Or, in a more functional pattern of composition:

-- CLOSURE --------------------------------------------------------------------

script closure
    on |λ|(x)
        script
            on |λ|()
                x * x
            end |λ|
        end script
    end |λ|
end script

|λ|() of (item 3 of (map(closure, enumFromTo(1, 10))))


-- GENERIC FUNCTIONS ----------------------------------------------------------

-- enumFromTo :: Int -> Int -> [Int]
on enumFromTo(m, n)
    if n < m then
        set d to -1
    else
        set d to 1
    end if
    set lst to {}
    repeat with i from m to n by d
        set end of lst to i
    end repeat
    return lst
end enumFromTo

-- map :: (a -> b) -> [a] -> [b]
on map(f, xs)
    tell mReturn(f)
        set lng to length of xs
        set lst to {}
        repeat with i from 1 to lng
            set end of lst to |λ|(item i of xs, i, xs)
        end repeat
        return lst
    end tell
end map

-- Lift 2nd class handler function into 1st class script wrapper 
-- mReturn :: Handler -> Script
on mReturn(f)
    if class of f is script then
        f
    else
        script
            property |λ| : f
        end script
    end if
end mReturn

Output:

9

Arturo

funcs: [ø]

loop 1..10 'f ->
    'funcs ++ function [] with 'f [
        f * f
    ]

print call funcs\3 []

Output:

9

Axiom

Using the Spad compiler:

)abbrev package TESTP TestPackage
TestPackage() : with
     test: () -> List((()->Integer))
   == add
     test() == [(() +-> i^2) for i in 1..10]

This can be called from the interpreter using:

[x() for x in test()]

Output:

[1,4,9,16,25,36,49,64,81,100]
                                     Type: List(Integer)

Babel

((main { 
    { iter 
        1 take bons 1 take
        dup cp 
        {*} cp 
        3 take 
        append }
    10 times
    collect !
    {eval %d nl <<} each }))

Output:

100
81
64
49
36
25
16
9
4
1

Essentially, a function has been constructed for each value to be squared (10 down to 1). The cp operator ensures that we generate a fresh copy of the number to be squared, as well as the code for multiplying, {*}. In the final each loop, we eval each of the constructed functions and output the result.

Bracmat

( -1:?i
& :?funcs
&   whl
  ' ( 1+!i:<10:?i
    & !funcs ()'(.$i^2):?funcs
    )
& whl'(!funcs:%?func %?funcs&out$(!func$))
);

Output:

0
1
4
9
16
25
36
49
64

C

Function image copying approach

Non-portable. Copying a function body depends on implementation-specific semantics of volatile, if the replacement target still exists after optimization, if the dest memory is suitably aligned, if the memory is executable, if it makes any function calls to a relative offset, if it refers to any memory location with an absolute address, etc. It only very occasionally works.

#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <sys/mman.h>

typedef int (*f_int)();
 
#define TAG 0xdeadbeef
int _tmpl() { 
	volatile int x = TAG;
	return x * x;
}

#define PROT (PROT_EXEC | PROT_WRITE)
#define FLAGS (MAP_PRIVATE | MAP_ANONYMOUS) 
f_int dupf(int v)
{
	size_t len = (void*)dupf - (void*)_tmpl;
	f_int ret = mmap(NULL, len, PROT, FLAGS, 0, 0);
	char *p;
	if(ret == MAP_FAILED) {
		perror("mmap");
		exit(-1);
	}
	memcpy(ret, _tmpl, len);
	for (p = (char*)ret; p < (char*)ret + len - sizeof(int); p++)
		if (*(int *)p == TAG) *(int *)p = v;
	return ret;
}
 
int main()
{
	f_int funcs[10];
	int i;
	for (i = 0; i < 10; i++) funcs[i] = dupf(i);
 
	for (i = 0; i < 9; i++)
		printf("func[%d]: %d\n", i, funcs[i]());
 
	return 0;
}

Output:

func[0]: 0
func[1]: 1
func[2]: 4
func[3]: 9
func[4]: 16
func[5]: 25
func[6]: 36
func[7]: 49
func[8]: 64

Greenspunned mini Lisp dialect

See Closures/Variable_capture/C for complete code. The relevant excerpt is:

void init(void)
{
  t = intern(lit("t"));
  x = intern(lit("x"));
}

val square(val env)
{
  val xbind = assoc(env, x); /* look up binding of variable x in env */
  val xval = cdr(xbind);     /* value is the cdr of the binding cell */
  return num(cnum(xval) * cnum(xval));
}

int main(void)
{
  int i;
  val funlist = nil, iter;

  init();

  for (i = 0; i < 10; i++) {
    val closure_env = cons(cons(x, num(i)), nil);
    funlist = cons(func_f0(closure_env, square), funlist);
  }

  for (iter = funlist; iter != nil; iter = cdr(iter)) {
    val fun = car(iter);
    val square = funcall(fun, nao);

    printf("%d\n", cnum(square));
  }
  return 0;
}

Here, we create an environment explicitly as an association list which we can search with the assoc function. The environment contains a binding for the symbol x. The square function retrieves the value and returns its square.

Output:

$ ./a.out
81
64
49
36
25
16
9
4
1
0

C#

Using Linq

using System;
using System.Linq;

class Program
{
    static void Main()
    {
        var captor = (Func<int, Func<int>>)(number => () => number * number);
        var functions = Enumerable.Range(0, 10).Select(captor);
        foreach (var function in functions.Take(9))
        {
            Console.WriteLine(function());
        }
    }
}

Output:

0
1
4
9
16
25
36
49
64

Using delegates only

using System;
using System.Collections.Generic;

class Program
{
    static void Main( string[] args )
    {
        List<Func<int>> l = new List<Func<int>>();
        for ( int i = 0; i < 10; ++i )
        {
            // This is key to avoiding the closure trap, because
            // the anonymous delegate captures a reference to 
            // outer variables, not their value.  So we create 10
            // variables, and each created anonymous delegate 
            // has references to that variable, not the loop variable
            var captured_val = i;
            l.Add( delegate() { return captured_val * captured_val; } );
        }

        l.ForEach( delegate( Func<int> f ) { Console.WriteLine( f() ); } );
    }
}

Output:

0
1
4
9
16
25
36
49
64

C++

Works with: C++11

#include <iostream>
#include <functional>
#include <vector>

int main() {
  std::vector<std::function<int()> > funcs;
  for (int i = 0; i < 10; i++)
    funcs.push_back([=]() { return i * i; });
  for ( std::function<int( )> f : funcs ) 
    std::cout << f( ) << std::endl ; 
  return 0;
}

Output:

0
1
4
9
16
25
36
49
64
81

Ceylon

shared void run() {
	
	//create a list of closures with a list comprehension
	value closures = [for(i in 0:10) () => i ^ 2];
	
	for(i->closure in closures.indexed) {
		print("closure number ``i`` returns: ``closure()``");
	}
}

Clojure

(def funcs (map #(fn [] (* % %)) (range 11)))
(printf "%d\n%d\n" ((nth funcs 3)) ((nth funcs 4)))

Output:

9
16

CoffeeScript

# Generate an array of functions.
funcs = ( for i in [ 0...10 ] then do ( i ) -> -> i * i )

# Call each function to demonstrate value capture.
console.log func() for func in funcs

Common Lisp

CL-USER> (defparameter alist
	   (loop for i from 1 to 10
	      collect (cons i (let ((i i))
				(lambda () (* i i))))))
ALIST
CL-USER> (funcall (cdr (assoc 2 alist)))
4
CL-USER> (funcall (cdr (assoc 8 alist)))
64

The loop mutates its binding i. The purpose of (let ((i i)) ...) is to create a different binding i for each lambda to capture. Otherwise, all 10 lambdas would capture the same binding and return 100.

D

Less Functional Version

import std.stdio;

void main() {
    int delegate()[] funcs;

    foreach (i; 0 .. 10)
        funcs ~= (i => () => i ^^ 2)(i);

    writeln(funcs[3]());
}

Output:

9

More Functional Version

void main() {
    import std.stdio, std.range, std.algorithm;

    10.iota.map!(i => () => i ^^ 2).map!q{ a() }.writeln;
}

Output:

[0, 1, 4, 9, 16, 25, 36, 49, 64, 81]

Delphi

Works with: Delphi 2009

program Project1;

type
  TFuncIntResult = reference to function: Integer;

// use function that returns anonymous method to avoid capturing the loop variable
function CreateFunc(i: Integer): TFuncIntResult;
begin
  Result :=
    function: Integer
    begin
      Result := i * i;
    end;
end;

var
  Funcs: array[0..9] of TFuncIntResult;
  i: integer;
begin
  // create 10 anonymous functions
  for i := Low(Funcs) to High(Funcs) do
    Funcs[i] := CreateFunc(i);

  // call all 10 functions
  for i := Low(Funcs) to High(Funcs) do
    Writeln(Funcs[i]());
end.

Output:

0
1
4
9
16
25
36
49
64
81

Dyalect

Dyalect captures variables by reference, therefore a way to achieve this is to capture a variable through a closure which in its turn returns a anonymous function like so:

var xs = []
let num = 10

for n in 0..<num {
    xs.Add((n => () => n * n)(n))
}

for x in xs {
    print(x())
}

Output:

0
1
4
9
16
25
36
49
64
81

This is similar to a JavaScript (ES6) solution.

EchoLisp

(define (fgen i) (lambda () (* i i)))
(define fs (for/vector ((i 10)) (fgen i))) ;; vector of 10 anonymous functions
((vector-ref fs 5)) ;; calls fs[5]
    → 25

Elena

ELENA 6.x :

import system'routines;
import extensions;
 
public program()
{
    var functions := Array.allocate(10).populate::(int i => { ^ i * i} );
 
    functions.forEach::(func) { console.printLine(func()) }
}

Output:

0
1
4
9
16
25
36
49
64
81

Elixir

funs = for i <- 0..9, do: (fn -> i*i end)
Enum.each(funs, &IO.puts &1.())

Output:

0
1
4
9
16
25
36
49
64
81

Emacs Lisp

As of Emacs 24.3, lexical closures are supported, therefore alleviating hacks such as lexical-let.

;;  -*- lexical-binding: t; -*-
(mapcar #'funcall
        (mapcar (lambda (x)
                  (lambda ()
                    (* x x)))
                '(1 2 3 4 5 6 7 8 9 10)))
;; => (1 4 9 16 25 36 49 64 81 100)

Erlang

Erlang uses lexical scoping and has anonymous functions.

-module(capture_demo).
-export([demo/0]).

demo() ->
    Funs = lists:map(fun (X) ->
                             fun () ->
                                     X * X
                             end
                     end,
                     lists:seq(1,10)),
    lists:foreach(fun (F) ->
                    io:fwrite("~B~n",[F()])
            end, Funs).

1> capture_demo:demo().
1
4
9
16
25
36
49
64
81
100
ok

F#

Nearly identical to OCaml

[<EntryPoint>]
let main argv = 
    let fs = List.init 10 (fun i -> fun () -> i*i)
    do List.iter (fun f -> printfn "%d" <| f()) fs
    0

With List.map

[<EntryPoint>]
let main argv = 
    let fs = List.map (fun i -> fun () -> i*i) [0..9]
    do List.iter (fun f -> printfn "%d" <| f()) fs
    0

With List.mapi

[<EntryPoint>]
let main argv = 
    let fs = List.mapi (fun i x -> fun () -> i*i) (List.replicate 10 None) 
    do List.iter (fun f -> printfn "%d" <| f()) fs
    0

With an infinite sequence

[<EntryPoint>]
let main argv = 
    let fs = Seq.initInfinite (fun i -> fun () -> i*i)
    do Seq.iter (fun f -> printfn "%d" <| f()) (Seq.take 10 fs)
    0

Output:

0
1
4
9
16
25
36
49
64
81

Factor

Using lexical variables

USING: io kernel locals math prettyprint sequences ;

[let
    ! Create a sequence of 10 quotations
    10 iota [
        :> i            ! Bind lexical variable i
        [ i i * ]       ! Push a quotation to calculate i squared
    ] map :> seq

    { 3 8 } [
        dup pprint " squared is " write
        seq nth call .
    ] each
]

$ ./factor script.factor
3 squared is 9
8 squared is 64

The code :> i always binds a new variable. This happens inside a loop, so this program creates 10 different bindings. Each closure [ i i * ] captures a different binding, and remembers a different value.

The wrong way would use f :> i! 10 iota [ i! [ i i * ] ] map :> seq to mutate a single binding. Then the program would print, "3 squared is 81", "8 squared is 81".

Using fried quotations

Forget the variable! Each fried quotation captures some values by pulling them from the stack.

USING: fry io kernel math prettyprint sequences ;

! Push a sequence of 10 quotations
10 iota [
    '[ _ dup * ]        ! Push a quotation ( i -- i*i )
] map

{ 3 8 } [
    dup pprint " squared is " write
    over nth call .
] each
drop

Fantom

class Closures
{
  Void main ()
  {
    // define a list of functions, which take no arguments and return an Int
    |->Int|[] functions := [,]

    // create and store a function which returns i*i for i in 0 to 10
    (0..10).each |Int i|
    {
      functions.add (|->Int| { i*i })
    }

    // show result of calling function at index position 7
    echo ("Function at index: " + 7 + " outputs " + functions[7].call)
  }
}

Output:

Function at index: 7 outputs 49

Forth

: xt-array here { a }
    10 cells allot 10 0 do
	:noname i ]] literal dup * ; [[ a i cells + !
    loop a ;

xt-array 5 cells + @ execute .

Output:

25

FreeBASIC

FreeBASIC doesn't support closures or anonymous methods, as such. However, what we can do is to create an array of objects to capture their index and then call a method on those objects which squares the index. This approach is similar to how some other object oriented languages implement closures 'under the hood'.

' FB 1.05.0 Win64

Type Closure
  Private:
    index As Integer
  Public:
    Declare Constructor(index As Integer = 0)
    Declare Function Square As Integer 
End Type

Constructor Closure(index As Integer = 0)
   This.index = index
End Constructor

Function Closure.Square As Integer
   Return index * index
End Function

Dim a(1 To 10) As Closure

' create Closure objects which capture their index
For i As Integer = 1 To 10
  a(i) = Closure(i)
Next

' call the Square method on all but the last object
For i As Integer = 1 to 9
  Print a(i).Square
Next

Print
Print "Press any key to quit"
Sleep

Output:

 1
 4
 9
 16
 25
 36
 49
 64
 81

Go

package main

import "fmt"

func main() {
    fs := make([]func() int, 10)
    for i := range fs {
        i := i
        fs[i] = func() int {
            return i * i
        }
    }
    fmt.Println("func #0:", fs[0]())
    fmt.Println("func #3:", fs[3]())
}

Output:

func #0: 0
func #3: 9

Groovy

Solution:

def closures = (0..9).collect{ i -> { -> i*i } }

Test:

assert closures instanceof List
assert closures.size() == 10
closures.each { assert it instanceof Closure }
println closures[7]()

Output:

49

Haskell

Using map:

fs = map (\i _ -> i * i) [1 .. 10]

Using list comprehensions:

fs = [const $ i * i | i <- [1 .. 10]]

Using infinite lists:

fs = take 10 coFs where coFs = [const $ i * i | i <- [1 ..]]

Testing:

> :t fs
fs :: [b -> Integer]
> map ($ ()) fs
[1,4,9,16,25,36,49,64,81,100]
> fs !! 9 $ ()
100
> fs !! 8 $ undefined
81

Icon and Unicon

This uses Unicon specific calling sequences for co-expressions. It can be made to run under Icon by modifying the calling syntax.

procedure main(args)                                      # Closure/Variable Capture
    every put(L := [], vcapture(1 to 10))                 # build list of index closures
    write("Randomly selecting L[",i := ?*L,"] = ",L[i]()) # L[i]() calls the closure
end
    
# The anonymous 'function', as a co-expression.  Most of the code is standard 
# boilerplate needed to use a co-expression as an anonymous function.

procedure vcapture(x)             # vcapture closes over its argument 
   return makeProc { repeat { (x[1]^2) @ &source } }  
end

procedure makeProc(A)             # the makeProc PDCO from the UniLib Utils package
    return (@A[1], A[1])
end

Library: Unicon Code Library

package Utils provides makeProc Summary of Anonymous Functions in Unicon

Output:

Randomly selecting L[8] = 64

Insitux

(var funcs (for x (range 11) #(* x x)))

[(0 funcs) ((3 funcs)) ((4 funcs))]

Output:

[#(* x x) 9 16]

Io

blist := list(0,1,2,3,4,5,6,7,8,9) map(i,block(i,block(i*i)) call(i))
writeln(blist at(3) call)  // prints 9

J

Explicit version

The natural way of implementing this in J is to define a function which produces a gerund of a constant function.

constF=:3 :0
  {.''`(y "_)
)

Thus, a list of 10 functions each producing a value in 0..9, and another with their squares:

flist=: constF"0 i.10
slist=: constF"0 *:i.10

Referencing a function by its index (its position in that list):

   flist @.3
3"_
   slist @.3
9"_

Using a function, given its index:

   flist @.4''
4
   slist @.4''
16

Running a randomly picked function which is not the last one:

   flist@.(?9) ''
7
   slist@.(?9) ''
25

Using temporary locales

The problem statement "Demonstrate how to create a series of independent closures based on the same template but maintain separate copies of the variable closed over" conflicts with the problem title "Value capture" in languages have sufficient abstraction to distinguish between value capture and variable and variable capture. This conflict even appears in J, and in general cases can require treatment of issues well outside the scope of this task.

Still, to address the task description, we should include a "variable capture" implementation, which in J could imply the use of "temporary locales" despite the fact that this approach would not satisfy the "simplest fashion possible" requirement.

For example, we could define an adverb 'geni' which takes a base function (which in this case will be *: -- a function which squares an argument) and a value (which in this case will be an index), creates a locale where that value will be stored in a variable named i and then returns an anonymous function which takes a reference to the locale (rather than the value) and extracts the value from the locale to generate the result.

We'll also use J's nuvoc {{ ... }} nesting definitional mechanism which implicitly determines the type of a definition instead of explicitly representing the definition types (1 :, 2 :, 3 :, ...) which discourages nesting blocks.

geni=: {{
  N=. cocreate''
  i__N=. y
  N
}}
task=: {{ u {{ u {{ u i__n [ y }} (geni y)`'' }}"0 i. y }}

This would be really bad form if we were intending to be useful, but - as described above - this approach is somewhat relevant to the task requirements.

Example use:

   fns=: *: task 10
   fns@.3 ''
9
   fns@.5 ''
25
   fns@.7 ''
49

Tacit (unorthodox) version

In J only adverbs and conjunctions (functionals) can produce verbs (functions)... Unless they are forced to cloak as verbs; in this instance, the rank conjunction (“) cloaks as a dyadic verb. (This does not work in recent versions of J as this takes advantage of a bug/feature where the interpreter does not produce a result with the correct shape):

   ( VL=. (<@:((<'"')(0:`)(,^:)&_))"0@:(^&2)@:i. 10 ) NB. Producing a list of boxed anonymous verbs (functions)
┌───┬───┬───┬───┬────┬────┬────┬────┬────┬────┐
│0"_│1"_│4"_│9"_│16"_│25"_│36"_│49"_│64"_│81"_│
└───┴───┴───┴───┴────┴────┴────┴────┴────┴────┘
   
   {::&VL 5                                           NB. Evoking the 6th verb (function)
25"_
   {::&VL 5 ''                                        NB. Invoking the 6th verb with a dummy argument ('')
25

Java

Works with: Java version 8+

import java.util.function.Supplier;
import java.util.ArrayList;

public class ValueCapture {
    public static void main(String[] args) {
	ArrayList<Supplier<Integer>> funcs = new ArrayList<>();
	for (int i = 0; i < 10; i++) {
	    int j = i;
	    funcs.add(() -> j * j);
	}

	Supplier<Integer> foo = funcs.get(3);
	System.out.println(foo.get()); // prints "9"
    }
}

Alternative implementation that also

Works with: Java version 8+

import java.util.List;
import java.util.function.IntSupplier;
import java.util.stream.IntStream;

import static java.util.stream.Collectors.toList;

public interface ValueCapture {
  public static void main(String... arguments) {
    List<IntSupplier> closures = IntStream.rangeClosed(0, 10)
      .<IntSupplier>mapToObj(i -> () -> i * i)
      .collect(toList())
    ;

    IntSupplier closure = closures.get(3);
    System.out.println(closure.getAsInt()); // prints "9"
  }
}

JavaScript

Imperative

var funcs = [];
for (var i = 0; i < 10; i++) {
    funcs.push( (function(i) {
                     return function() { return i * i; }
                })(i) );
}
window.alert(funcs[3]()); // alerts "9"

Works with: JavaScript version 1.7+

(Firefox 2+)

<script type="application/javascript;version=1.7">
var funcs = [];
for (var i = 0; i < 10; i++) {
    let (i = i) {
        funcs.push( function() { return i * i; } );
    }
}
window.alert(funcs[3]()); // alerts "9"
</script>

Works with: JavaScript version ES6

"use strict";
let funcs = [];
for (let i = 0; i < 10; ++i) {
    funcs.push((i => () => i*i)(i));
}
console.log(funcs[3]());

Functional

Works with: JavaScript version ES5

(function () {
    'use strict';

    // Int -> Int -> [Int]
    function range(m, n) {
        return Array.apply(null, Array(n - m + 1))
            .map(function (x, i) {
                return m + i;
            });
    }

    var lstFns = range(0, 10)
        .map(function (i) {
            return function () {
                return i * i;
            };
        })
        
    return lstFns[3]();

})();

Output:

9

Works with: JavaScript version ES6

let funcs = [...Array(10).keys()].map(i => () => i*i);

Output:

console.log(funcs[3]());
9

Julia

funcs = [ () -> i^2 for i = 1:10 ]

Output:

julia> funcs[7]()
49

Kotlin

// version 1.0.6

fun main(args: Array<String>) {
    // create an array of 10 anonymous functions which return the square of their index
    val funcs = Array(10){ fun(): Int = it * it }
    // call all but the last
    (0 .. 8).forEach { println(funcs[it]()) } 
}

Output:

0
1
4
9
16
25
36
49
64

Lambdatalk

A translation from Javascript

{def A
 {A.new
  {S.map {lambda {:x} {* :x :x}}
         {S.serie 0 10}
}}}

{A.get 3 {A}}    // equivalent to A[3]
-> 9
{A.get 4 {A}}
-> 16

Latitude

Latitude is particularly well suited to this challenge, as the various iteration constructs actually take method arguments and call them multiple times. Thus, the loop variable is in fact an argument which is already closed over and distinct at each iteration.

functions := 10 times to (Array) map {
  takes '[i].
  proc { (i) * (i). }.
}.

functions visit { println: $1 call. }.

Output:

0
1
4
9
16
25
36
49
64
81

LFE

Input at the REPL:

> (set funcs (list-comp ((<- m (lists:seq 1 10)))
                      (lambda () (math:pow m 2))))

Output:

(#Fun<lfe_eval.23.101079464> #Fun<lfe_eval.23.101079464>
 #Fun<lfe_eval.23.101079464> #Fun<lfe_eval.23.101079464>
 #Fun<lfe_eval.23.101079464> #Fun<lfe_eval.23.101079464>
 #Fun<lfe_eval.23.101079464> #Fun<lfe_eval.23.101079464>
 #Fun<lfe_eval.23.101079464> #Fun<lfe_eval.23.101079464>)

Calling the functions:

> (funcall (car funcs))
1.0
> (funcall (cadr funcs))
4.0
> (funcall (cadddr funcs))
16.0
> (funcall (lists:nth 8 funcs))
64.0

Lingo

Lingo doesn't really support closures. But with the limitations described at Function composition and based on the fact that Lingo allows to create arbitrary code at runtime, the task can be solved like this:

-- parent script "CallFunction"

property _code

-- if the function is supposed to return something, the code must contain a line that starts with "res="
on new (me, code)
  me._code = code
  return me
end

on call (me)
  ----------------------------------------  
  -- If custom arguments were passed, evaluate them in the current context.
  -- Note: in the code passed to the constructor they have to be referenced
  -- as arg[1], arg[2], ...
  arg = []
  repeat with i = 3 to the paramCount
    arg[i-2] = param(i)
  end repeat
  ----------------------------------------
  res = VOID
  do(me._code)
  return res
end

funcs = []
repeat with i = 1 to 10
  code = "res="&i&"*"&i
  funcs[i] = script("CallFunction").new(code)
end repeat

put call(funcs[3], _movie)
-- 9

Since the original task is a little trivial in terms of not depending on runtime arguments, here also a solution for an extended task: let each function[i] return the square of i plus the sum of all arguments passed to it at runtime:

funcs = []
repeat with i = 1 to 10
  code = ""
  put "res = "&i&"*"&i &RETURN after code
  put "repeat with i = 1 to arg.count" &RETURN after code
  put "  res = res + arg[i]" &RETURN after code
  put "end repeat" after code
  funcs[i] = script("CallFunction").new(code)
end repeat

put call(funcs[3], _movie, 23)
-- 32

put call(funcs[7], _movie, 4, 5, 6)
-- 64

Logtalk

The example that follow uses Logtalk's native support for lambda expressions.

:- object(value_capture).

    :- public(show/0).
    show :-
        integer::sequence(1, 10, List),
        meta::map(create_closure, List, Closures),
        meta::map(call_closure, List, Closures).

    create_closure(Index, [Double]>>(Double is Index*Index)).

    call_closure(Index, Closure) :-
        call(Closure, Result),
        write('Closure '), write(Index), write(' : '), write(Result), nl.

:- end_object.

Output:

| ?- value_capture::show.
Closure 1 : 1
Closure 2 : 4
Closure 3 : 9
Closure 4 : 16
Closure 5 : 25
Closure 6 : 36
Closure 7 : 49
Closure 8 : 64
Closure 9 : 81
Closure 10 : 100
yes

Lua

funcs={}
for i=1,10 do
    table.insert(funcs, function() return i*i end)
end
funcs[2]()
funcs[3]()

Output:

4
9

M2000 Interpreter

Dim Base 0, A(10)
For i=0 to 9 {
      a(i)=lambda i -> i**2
}
For i=0 to 9 {
      Print a(i)()
}

Print

                  0
                  1
                  4
                  9
                 16
                 25
                 36
                 49
                 64
                 81

Export list to clipboard

document a$
For i=0 to 9 {
      a$=format$("{0:0:-20}",a(i)())+{
      }
}
Clipboard a$

Using Inventory, and a stack object (reading from position, and another way, we pop functions, using Read)

Inventory Alfa
For i=0 to 9 {
     Append Alfa, i:=lambda i -> i**2
}
For i=0 to 9 {
      Print Alfa(i)()
}

Beta=Stack
Stack Beta {
      For i=0 to 9 {
           Data lambda i -> i**2
      }
}
Def Fun(X)=X()
\\ reading functions from position 1 to 10
For i=0 to 9 {
     Print fun(stackitem(Beta,i+1))
}
\\ pop functions form stack Beta
Stack Beta {
      While not empty {
            Read M
            Print M()
      }
}

Maple

> L := map( i -> (() -> i^2), [seq](1..10) ):
> seq( L[i](),i=1..10);                      
                  1, 4, 9, 16, 25, 36, 49, 64, 81, 100
> L[4]();
                                   16

Mathematica / Wolfram Language

Function[i, i^2 &] /@ Range@10
->{1^2 &, 2^2 &, 3^2 &, 4^2 &, 5^2 &, 6^2 &, 7^2 &, 8^2 &, 9^2 &, 10^2 &}

%[[2]][]
->4

Nemerle

using System.Console;

module Closures
{
    Main() : void
    { 
        def f(x) { fun() { x ** 2 } }
        def funcs = $[f(x) | x in $[0 .. 10]].ToArray(); // using array for easy indexing
        
        WriteLine($"$(funcs[4]())");
        WriteLine($"$(funcs[2]())");
    }
}

Output:

16
4

Nim

var funcs: seq[proc(): int] = @[]

for i in 0..9:
  (proc =
    let x = i
    funcs.add(proc (): int = x * x))()

for i in 0..8:
  echo "func[", i, "]: ", funcs[i]()

Objeck

use Collection.Generic;

class Capture {
  function : Main(args : String[]) ~ Nil {
     funcs := Vector->New()<FuncHolder<IntHolder> >;
     
     for(i := 0; i < 10; i += 1;) {
       funcs->AddBack(FuncHolder->New(\() ~ IntHolder : () => i * i)<IntHolder>);
     };

     each(i : funcs) {
       func := funcs->Get(i)->Get()<IntHolder>;
       func()->Get()->PrintLine();
     };
  }
}

Output:

0
1
4
9
16
25
36
49
64
81

Objective-C

Works with: Cocoa version Mac OS X 10.6+

with ARC

NSMutableArray *funcs = [[NSMutableArray alloc] init];
for (int i = 0; i < 10; i++) {
  [funcs addObject:[^ { return i * i; } copy]];
}

int (^foo)(void) = funcs[3];
NSLog(@"%d", foo()); // logs "9"

OCaml

All functions in OCaml are closures.

let () =
  let cls = Array.init 10 (fun i -> (function () -> i * i)) in
  Random.self_init ();
  for i = 1 to 6 do
    let x = Random.int 9 in
    Printf.printf " fun.(%d) = %d\n" x (cls.(x) ());
  done

Output:

 fun.(4) = 16
 fun.(1) = 1
 fun.(4) = 16
 fun.(7) = 49
 fun.(3) = 9
 fun.(6) = 36

Oforth

: newClosure(i)  #[ i sq ] ;
10 seq map(#newClosure) at(7) perform .

Output:

49

PARI/GP

Works with: PARI/GP version 2.4.2 and above

vector(10,i,()->i^2)[5]()

Output:

%1 = 25

Perl

my @f = map(sub { $_ * $_ }, 0 .. 9);   # @f is an array of subs
print $f[$_](), "\n" for (0 .. 8); # call and print all but last

Output:

0
1
4
9
16
25
36
49
64

Phix

Phix does not support closures, but they seem easy enough to emulate

with javascript_semantics
-- First some generic handling stuff, handles partial_args
-- of any mixture of any length and element types.
sequence closures = {}
function add_closure(integer rid, sequence partial_args)
    closures = append(closures,{rid,partial_args})
    return length(closures) -- (return an integer id)
end function
 
function call_closure(integer id, sequence args)
    {integer rid, sequence partial_args} = closures[id]
    return call_func(rid,partial_args&args)
end function
 
-- The test routine to be made into a closure, or ten
-- Note that all external references/captured variables must
-- be passed as arguments, and grouped together on the lhs
function square(integer i)
    return i*i
end function
 
-- Create the ten closures as asked for.
-- Here, cids is just {1,2,3,4,5,6,7,8,9,10}, however ids would be more
-- useful for a mixed bag of closures, possibly stored all over the shop.
-- Likewise add_closure could have been a procedure for this demo, but
-- you would probably want the function in a real-world application.
sequence cids = {}
for i=1 to 10 do
--for i=11 to 20 do -- alternative test
    cids &= add_closure(routine_id("square"),{i})
end for
-- And finally call em (this loop is blissfully unaware what function 
-- it is actually calling, and what partial_arguments it is passing)
for i=1 to 10 do
    printf(1," %d",call_closure(cids[i],{}))
end for

Output:

 1 4 9 16 25 36 49 64 81 100

output if that 11 to 20 add_closure loop is used instead:

 121 144 169 196 225 256 289 324 361 400

Note however that any captured values are effectively immutable, unless you also pass the id to the closure, and that in turn does rude things to closures[id][2].

A dictionary based approach may prove somewhat easier:

with javascript_semantics
function square(integer tid)
    integer i = getd("i",tid)   -- (setd valid here too)
    return i*i
end function

sequence tids = {}
for i=1 to 10 do
--for i=11 to 20 do
    tids &= new_dict({{"i",i}})
end for
for i=1 to 10 do
    printf(1," %d",square(tids[i]))
end for

same output, for both tests

Phixmonti

def power2
    dup *
enddef

getid power2 10 repeat

len for
    dup rot swap get rot swap exec print " " print
endfor

nl

/# Another mode #/
len for
    var i
    i get i swap exec print " " print
endfor

PHP

Works with: PHP version 5.3+

<?php
$funcs = array();
for ($i = 0; $i < 10; $i++) {
    $funcs[] = function () use ($i) { return $i * $i; };
}
echo $funcs[3](), "\n"; // prints 9
?>

Works with: PHP version pre-5.3

This method can capture value types like numbers, strings, arrays, etc., but not objects.

<?php
$funcs = array();
for ($i = 0; $i < 10; $i++) {
    $funcs[] = create_function('', '$i = ' . var_export($i, true) . '; return $i * $i;');
}
echo $funcs[3](), "\n"; // prints 9
?>

PicoLisp

(setq FunList
   (make
      (for @N 10
         (link (curry (@N) () (* @N @N))) ) ) )

Test:

: ((get FunList 2))
-> 4

: ((get FunList 8))
-> 64

Pike

array funcs = ({});
foreach(enumerate(10);; int i)
{ 
  funcs+= ({ 
              lambda(int j)
              {
                  return lambda()
                         { 
                             return j*j; 
                         }; 
              }(i) 
          }); 
}

PowerShell

I'm not sure that I understood the question/task. This task seems to be the same as the 'Accumulator Factory' task.

function Get-Closure ([double]$Number)
{
    {param([double]$Sum) return $script:Number *= $Sum}.GetNewClosure()
}

for ($i = 1; $i -lt 11; $i++)
{ 
    $total = Get-Closure -Number $i

    [PSCustomObject]@{
        Function = $i
        Sum      = & $total -Sum $i
    }
}

Output:

Function Sum
-------- ---
       1   1
       2   4
       3   9
       4  16
       5  25
       6  36
       7  49
       8  64
       9  81
      10 100

$numbers = 1..20 | Get-Random -Count 10

foreach ($number in $numbers)
{
    $total = Get-Closure -Number $number

    [PSCustomObject]@{
        Function = $number
        Sum      = & $total -Sum $number
    }
}

Output:

Function Sum
-------- ---
       4  16
      16 256
       3   9
      17 289
       9  81
      15 225
       7  49
       6  36
       1   1
      20 400

Prolog

Works with SWI-Prolog and module lambda.pl from Ulrich Neumerkel.
lambda.pl can be found there : http://www.complang.tuwien.ac.at/ulrich/Prolog-inedit/lambda.pl

:-use_module(library(lambda)).


closure :-
	numlist(1,10, Lnum),
	maplist(make_func, Lnum, Lfunc),
	maplist(call_func, Lnum, Lfunc).


make_func(I, \X^(X is I*I)).

call_func(N, F) :-
	call(F, R),
	format('Func ~w : ~w~n', [N, R]).

Output:

 ?- closure.
Func 1 : 1
Func 2 : 4
Func 3 : 9
Func 4 : 16
Func 5 : 25
Func 6 : 36
Func 7 : 49
Func 8 : 64
Func 9 : 81
Func 10 : 100
true.

Python

The naive way does not work:

funcs = []
for i in range(10):
    funcs.append(lambda: i * i)
print funcs[3]() # prints 81

The simplest solution is to add optional parameters with default arguments at the end of the parameter list, to create a local copy of the variable, and evaluate the variable at the time the function is created. (The optional parameter is not expected to ever be passed.) Often, the optional parameter will be named the same as the variable to be closed over (leading to odd-looking code of the form foo=foo in the arguments), so that the code inside the function need not be changed, but this might lead to confusion. This technique does not work for functions with a variable number of arguments.

funcs = []
for i in range(10):
    funcs.append(lambda i=i: i * i)
print funcs[3]() # prints 9

or equivalently the list comprehension:

funcs = [lambda i=i: i * i for i in range(10)]
print funcs[3]() # prints 9

Another solution is to wrap an immediately-executed function around our function. The wrapping function creates a new scope, and its execution forces the evaluation of the variable to be closed over.

funcs = []
for i in range(10):
    funcs.append((lambda i: lambda: i * i)(i))
print funcs[3]() # prints 9

or equivalently the list comprehension:

funcs = [(lambda i: lambda: i)(i * i) for i in range(10)]
print funcs[3]() # prints 9

In this case it is also possible to use map() since the function passed to it creates a new scope

funcs = map(lambda i: lambda: i * i, range(10))
print funcs[3]() # prints 9

It is also possible to use eval.

funcs=[eval("lambda:%s"%i**2)for i in range(10)]
print funcs[3]() # prints 9

Quackery

Strictly speaking, we could get away with [ table 0 1 4 9 16 25 36 49 64 81 ] is functions ( n --> n ) for this task, as numbers in Quackery are functions that return their own value when executed, e.g 5 do returns 5, but it feels like cheating.

  [ table ] is functions ( n --> [ )

  10 times 
    [ i^ ' [ dup * ] join 
      ' functions put ]

  5 functions do echo

Output:

25

R

R is a natural language for this task, but you need to understand the nuances of delayed evaluation. Arguments in R are referred to as promises because they aren't evaluated until first use. If you're not careful, you can bind to a promise that hasn't yet been evaluated, and you won't get what you expect.

# assign 's' a list of ten functions 
s <- sapply (1:10,  # integers 1..10 become argument 'x' below 
    function (x) {
        x  # force evaluation of promise x
	function (i=x) i*i   # this *function* is the return value
    })

s[[5]]()  # call the fifth function in the list of returned functions 
[1] 25    # returns vector of length 1 with the value 25

Note that I bound the captured variable as the default argument on a unary function. If you supply your own argument, as below, it squares the supplied argument and ignores the default argument.

s[[5]](10) 
[1] 100

As a further technicality, note that you need some extra voodoo to modify the bound argument with persistence across calls. This example increments the bound variable after each call.

s <- sapply (1:10,  
    function (x) {
        x  # force evaluation of promise x
	function () {   
            R <- x*x 
            # evaluate the language expression "x <- x + 1" in the persistent parent environment 
            evalq (x <- x + 1, parent.env(environment()))
            R  # return squared value 
    }})

s[[5]]() 
[1] 25     # 5^2
s[[5]]() 
[1] 36     # now 6^2
s[[1]]()
[1] 1      # 1^2
s[[1]]()
[1] 4      # now 2^2

As shown, each instance increments separately.

--- Edit ---

I think that modifying the bound variable can be done in a simpler way. Instead of:

    evalq (x <- x + 1, parent.env(environment()))

substitute:

    x <<- x + 1

Testing:

> s[[5]]()
[1] 25
> s[[5]]()
[1] 36
> s[[5]]()
[1] 49
> s[[2]]()
[1] 4
> s[[2]]()
[1] 9
> s[[2]]()
[1] 16

Racket

#lang racket
(define functions (for/list ([i 10]) (λ() (* i i))))
(map (λ(f) (f)) functions)

Output:

'(0 1 4 9 16 25 36 49 64 81)

Raku

(formerly Perl 6)

Works with: Rakudo version 2015.12

All blocks are anonymous closures in Raku, and parameters are lexicals, so it's easy to generate a list of them. We'll use a gather/take generator loop, and call the closures in random order, just to keep things interesting.

my @c = gather for ^10 -> $i {
    take { $i * $i }
}

.().say for @c.pick(*);  # call them in random order

Output:

36
64
25
1
16
0
4
9
81
49

Or equivalently, using a more functional notation:

say .() for pick *, map -> $i { -> {$i * $i} }, ^10

Red

funs: collect [repeat i 10 [keep func [] reduce [i ** 2]]]

>> funs/7
== 49

REXX

This REXX version supports both a one─ and zero─based list   (it can be specified at the command line.)

The default is to use a zero─based list.

The list can also be specified at the command line.   The default is an ordered list based on an obscure sequence   (but puzzle enthusiasts can figure it out).

No error checking is performed on the user input(s).

/*REXX program has a list of ten functions, each returns its invocation (index) squared.*/
parse arg seed base $                            /*obtain optional arguments from the CL*/
if datatype(seed, 'W')  then call random ,,seed  /*Not given?  Use random  start seed.  */
if base=='' | base=","  then base=0              /* "    "     Use a zero─based list.   */
if $=''  then $= 8 5 4 9 1 3 2 7 6 0             /* "    "     Use ordered function list*/
                                                 /*the $ list must contain 10 functions.*/
say 'the' word("zero one", base+1)'─based list is: '   $       /*show list of functions.*/
                                                 /*BASED  must be either   1   or   0.  */
?='.'random(0, 9)                                /*get a random name of a function.     */
interpret  'CALL'  ?                             /*invoke a randomly selected function. */
say 'function '    ?     " returned "    result  /*display the value of random function.*/
exit                                             /*stick a fork in it,  we're all done. */
/*────────────────────────[Below are the closest things to anonymous functions in REXX].*/
.0: return  .(0)                                 /*function  .0   ─── bump its counter. */
.1: return  .(1)                                 /*    '     .1    "    "   "     "     */
.2: return  .(2)                                 /*    '     .2    "    "   "     "     */
.3: return  .(3)                                 /*    '     .3    "    "   "     "     */
.4: return  .(4)                                 /*    '     .4    "    "   "     "     */
.5: return  .(5)                                 /*    '     .5    "    "   "     "     */
.6: return  .(6)                                 /*    '     .6    "    "   "     "     */
.7: return  .(7)                                 /*    '     .7    "    "   "     "     */
.8: return  .(8)                                 /*    '     .8    "    "   "     "     */
.9: return  .(9)                                 /*    '     .9    "    "   "     "     */
/*──────────────────────────────────────────────────────────────────────────────────────*/
.:  arg #;  _=wordpos(#,$); if _==0  then return 'not in the list.'; return (_-(\base))**2

output   when using the default input   which assume a zero─based list):

the zero─based list is:  8 5 4 9 1 3 2 7 6 0
function  .0  returned  81

output   when using the input of:     ,   1

the one─based list is:  8 5 4 9 1 3 2 7 6 0
function  .0  returned  100

Ring

x = funcs(7)
see x + nl

func funcs n
     fn = list(n)
     for i = 1 to n    
         fn[i] =i*i
     next 
     return fn

Output:

1
4
9
16
25
36
49

Ruby

procs = Array.new(10){|i| ->{i*i} } # -> creates a lambda
p procs[7].call # => 49

In Ruby, lambdas (and procs) are closures.

Rust

One note here about referencing values and capturing values:
Rust employs strong ownership rules that do not allow mutating a value that is referenced (pointed to without allowing mutation) from elsewhere. It also doesn't allow referencing a value that may be dropped before the reference is released. The proof that we really did capture the value is therefore unnecessary. Either we did or it wouldn't have compiled.

fn main() {
    let fs: Vec<_> = (0..10).map(|i| {move || i*i} ).collect();
    println!("7th val: {}", fs[7]());
}

Output:

7th val: 49

Scala

val closures=for(i <- 0 to 9) yield (()=>i*i)
0 to 8 foreach (i=> println(closures(i)()))
println("---\n"+closures(7)())

Output:

0
1
4
9
16
25
36
49
64
---
49

Scheme

;;; Collecting lambdas in a tail-recursive function.
(define (build-list-of-functions n i list)
  (if (< i n)
      (build-list-of-functions n (+ i 1) (cons (lambda () (* (- n i) (- n i))) list))
      list))

(define list-of-functions (build-list-of-functions 10 1 '()))

(map (lambda (f) (f)) list-of-functions)

((list-ref list-of-functions 8))

Output:

'(1 4 9 16 25 36 49 64 81)
81

---

Using Scheme SRFI 1 iota procedure can be simplified to:

(define list-of-functions (map (lambda (x) (lambda () (* x x))) (iota 0 1 10)))

; print the result
(display
  (map (lambda (n) (n)) list-of-functions)
(newline)

Sidef

var f = (
    10.of {|i| func(j){i * j} }
)

9.times { |j|
    say f[j](j)
}

Output:

0
1
4
9
16
25
36
49
64

Starting from i=1:

var f = (1..10).map { |i|
    func(j){i * j}
}

for j (1..9) {
    say f[j-1](j)
}

Output:

1
4
9
16
25
36
49
64
81

Smalltalk

funcs := (1 to: 10) collect: [ :i | [ i * i ] ] .
(funcs at: 3) value displayNl .

Output:

9

Sparkling

In Sparkling, upvalues (variables in the closure) are captured by value.

var fnlist = {};
for var i = 0; i < 10; i++ {
	fnlist[i] = function() {
		return i * i;
	};
}

print(fnlist[3]()); // prints 9
print(fnlist[5]()); // prints 25

Alternately:

var fnlist = map(range(10), function(k, v) {
	return function() {
		return v * v;
	};
});

print(fnlist[3]()); // prints 9
print(fnlist[5]()); // prints 25

Standard ML

List.map (fn x => x () )  ( List.tabulate (10,(fn i => (fn  ()=> i*i)) ) ) ;

Output:

val it = [0,1,4,9,16,25,36,49,64,81] : int list

Swift

By default, Swift captures variables by reference. A naive implementation like the following C-style for loop does not work:

var funcs: [() -> Int] = []
for var i = 0; i < 10; i++ {
  funcs.append({ i * i })
}
println(funcs[3]()) // prints 100

However, using a for-in loop over a range does work, since you get a new constant at every iteration:

var funcs: [() -> Int] = []
for i in 0..<10 {
  funcs.append({ i * i })
}
println(funcs[3]()) // prints 9

The C-style for loop can also work if we explicitly capture the loop counter:

var funcs: [() -> Int] = []
for var i = 0; i < 10; i++ {
  funcs.append({ [i] in i * i })
}
println(funcs[3]()) // prints 9

Alternately, we can also use map() to map over a range, and create the squaring closure inside the mapping closure which has the integer as a parameter:

let funcs = [] + map(0..<10) {i in { i * i }}
println(funcs[3]()) // prints 9

Tcl

Tcl does not support closures (either value-capturing or variable-capturing) by default, but value-capturing closures are easy to emulate.

package require Tcl 8.6; # Just for tailcall command
# Builds a value-capturing closure; does NOT couple variables
proc closure {script} {
    set valuemap {}
    foreach v [uplevel 1 {info vars}] {
	lappend valuemap [list $v [uplevel 1 [list set $v]]]
    }
    set body [list $valuemap $script [uplevel 1 {namespace current}]]
    # Wrap, to stop untoward argument passing
    return [list apply [list {} [list tailcall apply $body]]]
    # A version of the previous line compatible with Tcl 8.5 would be this
    # code, but the closure generated is more fragile:
    ### return [list apply $body]
}

# Simple helper, to avoid capturing unwanted variable
proc collectFor {var from to body} {
    upvar 1 $var v
    set result {}
    for {set v $from} {$v < $to} {incr v} {lappend result [uplevel 1 $body]}
    return $result
}
# Build a list of closures
proc buildList {} {
    collectFor i 0 10 {
	closure {
	    # This is the body of the closure
	    return [expr $i*$i]
	}
    }
}
set theClosures [buildList]
foreach i {a b c d e} {# Do 5 times; demonstrates no variable leakage
    set idx [expr {int(rand()*9)}]; # pick random int from [0..9)
    puts $idx=>[{*}[lindex $theClosures $idx]]
}

Output:

5=>25
0=>0
8=>64
1=>1
8=>64

TXR

Sugared

(let ((funs (mapcar (ret (op * @@1 @@1)) (range 1 10))))
  [mapcar call [funs 0..-1]])

Output:

(1 4 9 16 25 36 49 64 81)

Desugared

Translation of: Emacs Lisp

The explicit lambda structure here is much like the implicit ones in the "Sugared" example:

;; Dropping distracting "skip last" requirement
;; (not implemented in original Elisp either).
(mapcar 'call
	(mapcar (lambda ()
		  (lambda () (* x x))) '(1 2 3 4 5 6 7 8 9 10)))

Delimited Continuations

In this interactive example, we capture delimited continuations inside a simple for loop. Because the variable binding environment is not necessarily in the stack which is captured, we rebind the loop variable.

This is the TXR Lisp interactive listener of TXR 124.
Use the :quit command or type Ctrl-D on empty line to exit.
1> (let ((conts))
      (for ((i 0)) ((< i 10) (nreverse conts)) ((inc i))
        (let ((cap i))
           (push (block sqr
                    (suspend sqr f (op f nil))
                    (* cap cap))
                 conts))))
(#<interpreted fun: lambda #:rest-0112> #<interpreted fun: lambda #:rest-0112>
 #<interpreted fun: lambda #:rest-0112> #<interpreted fun: lambda #:rest-0112>
 #<interpreted fun: lambda #:rest-0112> #<interpreted fun: lambda #:rest-0112>
 #<interpreted fun: lambda #:rest-0112> #<interpreted fun: lambda #:rest-0112>
 #<interpreted fun: lambda #:rest-0112> #<interpreted fun: lambda #:rest-0112>)
2> (call (first *1))
0
3> (call (second *1))
1
4> (call (fifth *1))
16
5> (call [*1 4])
16
6> (call [*1 7])
49

The suspend operator suspends the execution of the sqr block, causing it to return the function (op f nil). The variable f represents the captured continuation as a function. Continuation functions take one mandatory argument. We don't need that here, hence the (op f nil) expression is returned: it curries the one arg continuation function f to a function with no arguments.

The loop pushes these suspended continuations into a list, and then nreverse-s it.

We then interactively call the continuations in the list.

Whenever we call a continuation, the (block sqr ...) environment is restored. and the suspended computation inside the block resumes by returning out of the (suspend ...) form normally. The block then executes to completion, returning the (* cap cap) form's value. At that point, our call to the continuation terminates, yielding that value.

Wren

var fs = List.filled(10, null)
for (i in 0...fs.count) {
    fs[i] = Fn.new { i * i }
}

for (i in 0...fs.count-1) System.print("Function #%(i):  %(fs[i].call())")

Output:

Function #0:  0
Function #1:  1
Function #2:  4
Function #3:  9
Function #4:  16
Function #5:  25
Function #6:  36
Function #7:  49
Function #8:  64

Yabasic

dim funcs$(10)

sub power2(i)
    return i * i
end sub

for i = 1 to 10
    funcs$(i) = "power2"
next

for i = 1 to 10
    print execute(funcs$(i), i)
next

zkl

Create a closure of the index over a square function

(0).pump(10,List,fcn(i){i*i}.fp)[8]() //-->64
list:=(0).pump(10,List,fcn(i){i*i}.fp);
foreach n in (list.len()-1) { list[n]().println() }
list.run(True).println()

Output:

0
1
4
9
16
25
36
49
64
L(0,1,4,9,16,25,36,49,64,81)