Integer sequence

From Rosetta Code
Task
Integer sequence
You are encouraged to solve this task according to the task description, using any language you may know.
Task

Create a program that, when run, would display all integers from   1   to     (or any relevant implementation limit),   in sequence   (i.e.   1, 2, 3, 4, etc)   if given enough time.


An example may not be able to reach arbitrarily-large numbers based on implementations limits.   For example, if integers are represented as a 32-bit unsigned value with 0 as the smallest representable value, the largest representable value would be 4,294,967,295.   Some languages support arbitrarily-large numbers as a built-in feature, while others make use of a module or library.

If appropriate, provide an example which reflect the language implementation's common built-in limits as well as an example which supports arbitrarily large numbers, and describe the nature of such limitations—or lack thereof.

Contents

360 Assembly[edit]

For maximum compatibility, this program uses only the basic instruction set (S/360).

*        Integer sequence      06/05/2016
INTSEQED CSECT
USING INTSEQED,12
LR 12,15
LA 6,1 i=1
LOOP CVD 6,DW binary to pack decimal
MVC WTOMSG+4(12),EM12 load mask
ED WTOMSG+4(12),DW+2 packed dec to char
WTO MF=(E,WTOMSG) write to console
LA 6,1(6) i=i+1
B LOOP goto loop
WTOMSG DC 0F,H'80',H'0',CL80' '
DW DS 0D,PL8 pack dec 15num
EM12 DC X'402020202020202020202120' mask CL12 11num
END INTSEQED
Output:
...
      314090
      314091
      314092
      314093
      314094
      314095
      314096
      314097
      314098
      314099
...

0815[edit]

}:_:<:1:+%<:a:~$^:_:

8080 Assembly[edit]

Actually printing the numbers out would depend on the hardware and operating system.

        ORG     0100H
MVI A, 0  ; move immediate
LOOP: INR A  ; increment
 ; call 'PRINT' subroutine, if required
JMP LOOP  ; jump unconditionally
 
END

Ada[edit]

with Ada.Text_IO;
procedure Integers is
Value : Integer := 1;
begin
loop
Ada.Text_IO.Put_Line (Integer'Image (Value));
Value := Value + 1; -- raises exception Constraint_Error on overflow
end loop;
end Integers;

Alternative (iterating through all values of Positive (positive part of Integer) without endless loop):

with Ada.Text_IO;
procedure Positives is
begin
for Value in Positive'Range loop
Ada.Text_IO.Put_Line (Positive'Image (Value));
end loop;
end Positives;

ALGOL 68[edit]

Works with: ALGOL 68 version Revision 1 - no extensions to language used.
Works with: ALGOL 68G version Any - tested with release 1.18.0-9h.tiny.

The upper limit of the loop variable i is max int currently +2147483647 for ALGOL 68G.

main:
(
FOR i DO
printf(($g(0)","$,i))
OD
)

Partial output:

1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,...

ALGOL W[edit]

begin
 % print the integers from 1 onwards  %
 % Algol W only has 32-bit integers. When i reaches 2^32,  %
 % an integer overflow event would be raised which by default,  %
 % should terminate the program  %
integer i;
i := 1;
while true do begin
write( i );
i := i + 1
end loop_forever ;
end.

Applesoft BASIC[edit]

Integer variables can be within the range of -32767 to 32767.

 10 I% = 1
20 PRINT I%;
30 I% = I% + 1
40 PRINT ", ";
50 GOTO 20

Last screen of scrolled output:

, 32646, 32647, 32648, 32649, 32650, 326
51, 32652, 32653, 32654, 32655, 32656, 3
2657, 32658, 32659, 32660, 32661, 32662,
32663, 32664, 32665, 32666, 32667, 3266
8, 32669, 32670, 32671, 32672, 32673, 32
674, 32675, 32676, 32677, 32678, 32679,
32680, 32681, 32682, 32683, 32684, 32685
, 32686, 32687, 32688, 32689, 32690, 326
91, 32692, 32693, 32694, 32695, 32696, 3
2697, 32698, 32699, 32700, 32701, 32702,
32703, 32704, 32705, 32706, 32707, 3270
8, 32709, 32710, 32711, 32712, 32713, 32
714, 32715, 32716, 32717, 32718, 32719,
32720, 32721, 32722, 32723, 32724, 32725
, 32726, 32727, 32728, 32729, 32730, 327
31, 32732, 32733, 32734, 32735, 32736, 3
2737, 32738, 32739, 32740, 32741, 32742,
32743, 32744, 32745, 32746, 32747, 3274
8, 32749, 32750, 32751, 32752, 32753, 32
754, 32755, 32756, 32757, 32758, 32759,
32760, 32761, 32762, 32763, 32764, 32765
, 32766, 32767
?ILLEGAL QUANTITY ERROR IN 30
]

ARM Assembly[edit]

.text
.global main
 
@ An ARM program that keeps incrementing R0 forever
@
@ If desired, a call to some 'PRINT' routine --
@ which would depend on the OS -- could be included
 
main:
mov r0, #0 @ start with R0 = 0
 
repeat:
@ call to 'PRINT' routine
add r0, r0, #1 @ increment R0
b repeat @ unconditional branch

AutoHotkey[edit]

This uses traytip to show the results. A msgbox, tooltip, or fileappend could also be used.

x=0
Loop
TrayTip, Count, % ++x

AWK[edit]

BEGIN {
for( i=0; i != i + 1; i++ )
print( i )
}

Awk uses floating-point numbers. This loop terminates when i becomes too large for integer precision. With IEEE doubles, this loop terminates when i reaches 2 ^ 53.

Axe[edit]

Integers in Axe are limited to 16 bits, or a maximum of 65535. This script will run infinitely until either the variable overflows or a key is pressed.

While getKey(0)
End
0→I
Repeat getKey(0)
Disp I▶Dec,i
I++
EndIf I=0

BASIC[edit]

Works with: ZX Spectrum Basic
10 LET A = 0
20 LET A = A + 1
30 PRINT A
40 GO TO 20
Works with: QBasic
A = 0
DO: A = A + 1: PRINT A: LOOP 1

BBC BASIC[edit]

Native version, limited to 53-bit integers (maximum output 9007199254740992):

      *FLOAT 64
REPEAT
i += 1
PRINT TAB(0,0) i;
UNTIL FALSE

Version using Huge Integer Math and Encryption library (up to 2^31 bits, but this program limited to 65535 decimal digits because of maximum string length):

      INSTALL @lib$+"HIMELIB"
PROC_himeinit("")
reg% = 1
 
PROC_hiputdec(reg%, "0")
REPEAT
SYS `hi_Incr`, ^reg%, ^reg%
PRINT TAB(0,0) FN_higetdec(reg%);
UNTIL FALSE

bc[edit]

while (++i) i

beeswax[edit]

Using an ordinary loop structure:

 qNP<
_1>{d

Using a jump instruction:

_1F6~@[email protected]

Numbers in beeswax are unsigned 64-bit integers, so after reaching 2^64-1 the counter wraps around to 0.

Befunge[edit]

The range of a numeric value in Befunge is implementation dependent, but is commonly 32 bit signed integers for the stack, so a maximum value of 2147483647. However, note that some implementations have a smaller range for displayed values, so the sequence may appear to wrap to negative numbers while the internal value is in fact still increasing.

Also note that the range of values written to the code page or 'playfield' is often much smaller - frequently only supporting 8 bits, sometimes signed, sometimes unsigned.

1+:0`!#@_:.55+,

Bracmat[edit]

Translation of: Ruby

Bracmat uses big numbers. Numbers are stored with a radix 10, each decimal digit occupying one byte. When multiplying or dividing, numbers are temporarily converted to radix 10000 (32-bit systems: 1 digit occupies two bytes) or radix 100000000 (64-bit systems: 1 digit occupies four bytes) to speed up the computation.

0:?n&whl'out$(1+!n:?n)

Brainf***[edit]

This program assumes that decrementing past zero wraps around, but it doesn't rely on cell size, other than that a cell can hold at least six bits. It also assumes the ASCII character set. This is an arbitrarily large number implementation.

++++++++++>>>+[[->>+<[+>->+<<---------------------------------------
-------------------[>>-<++++++++++<[+>-<]]>[-<+>]<++++++++++++++++++
++++++++++++++++++++++++++++++>]<[<]>>[-<+++++++++++++++++++++++++++
++++++++++++++++++++++>]>]>[>>>]<<<[.<<<]<.>>>+]

This modification of the previous program will print out 1 to the maximum cell value, still assuming wrapping. On many implementations, this will print out 1-255.

++++++++++>>-[>+[->>+<[+>->+<<--------------------------------------
--------------------[>>-<++++++++++<[+>-<]]>[-<+>]<+++++++++++++++++
+++++++++++++++++++++++++++++++>]<[<]>>[-<++++++++++++++++++++++++++
+++++++++++++++++++++++>]>]>[>>>]<<<[.<<<]<.>>-]

This program can count in any base counting system under 256. Note: Change the characters in quotes equal to the base counting system you want to use.

+[<<+>>[[<<"-----------"["+++++++++++"<]>]>[<<<<+>>+>>[>>]<]<]>>[>>]<<]

Brat[edit]

i = 1
 
loop {
p i
i = i + 1
}

Burlesque[edit]

 
1R@
 

C[edit]

Prints from 1 to max unsigned integer (usually 2**32 -1), then stops.

#include <stdio.h>
 
int main()
{
unsigned int i = 0;
while (++i) printf("%u\n", i);
 
return 0;
}

Library: GMP
[edit]

This one never stops. It's not even likely that you'll run out of memory before you run out of patience.
#include <gmp.h>
 
int main()
{
mpz_t i;
mpz_init(i); /* zero now */
 
while (1) {
mpz_add_ui(i, i, 1); /* i = i + 1 */
gmp_printf("%Zd\n", i);
}
 
return 0;
}

Library: OpenSSL
[edit]

OpenSSL provides arbitrarily large integers.

#include <openssl/bn.h>		/* BN_*() */
#include <openssl/err.h> /* ERR_*() */
#include <stdio.h> /* fprintf(), puts() */
 
void
fail(const char *message)
{
fprintf(stderr, "%s: error 0x%08lx\n", ERR_get_error());
exit(1);
}
 
int
main()
{
BIGNUM i;
char *s;
 
BN_init(&i);
for (;;) {
if (BN_add_word(&i, 1) == 0)
fail("BN_add_word");
s = BN_bn2dec(&i);
if (s == NULL)
fail("BN_bn2dec");
puts(s);
OPENSSL_free(s);
}
/* NOTREACHED */
}

C#[edit]

using System;
using System.Numerics;
 
class Program
{
static void Main()
{
BigInteger i = 1;
while (true)
{
Console.WriteLine(i++);
}
}
}

ChucK[edit]

Math.INT_MAX is a constant value that represents the greater integer, 32 bit , 64 bit systems.

 
for(1 => int i; i < Math.INT_MAX; i ++)
{
<<< i >>>;
}
 

C++[edit]

#include <cstdint>
#include <iostream>
#include <limits>
 
int main()
{
auto i = std::uintmax_t{};
 
while (i < std::numeric_limits<decltype(i)>::max())
std::cout << ++i << '\n';
}

Clean[edit]

In Clean this example has a limit of basically 2147483648.

module IntegerSequence
 
import StdEnv
 
Start = [x \\ x <- [1..]]

Output:

[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,..

Clojure[edit]

(map println (next (range)))

COBOL[edit]

       IDENTIFICATION DIVISION.
PROGRAM-ID. Int-Sequence.
 
DATA DIVISION.
WORKING-STORAGE SECTION.
* *> 36 digits is the largest size a numeric field can have.
01 I PIC 9(36).
 
PROCEDURE DIVISION.
* *> Display numbers until I overflows.
PERFORM VARYING I FROM 1 BY 1 UNTIL I = 0
DISPLAY I
END-PERFORM
 
GOBACK
.

CoffeeScript[edit]

Like with most languages, counting is straightforward in CoffeeScript, so the program below tries to handle very large numbers. See the comments for starting the sequence from 1.

 
# This very limited BCD-based collection of functions
# makes it easy to count very large numbers. All arrays
# start off with the ones columns in position zero.
# Using arrays of decimal-based digits to model integers
# doesn't make much sense for most tasks, but if you
# want to keep counting forever, this does the trick.
 
BcdInteger =
from_string: (s) ->
arr = []
for c in s
arr.unshift parseInt(c)
arr
 
render: (arr) ->
s = ''
for elem in arr
s = elem.toString() + s
s
 
succ: (arr) ->
arr = (elem for elem in arr)
i = 0
while arr[i] == 9
arr[i] = 0
i += 1
arr[i] ||= 0
arr[i] += 1
arr
 
# To start counting from 1, change the next line!
big_int = BcdInteger.from_string "199999999999999999999999999999999999999999999999999999"
while true
console.log BcdInteger.render big_int
big_int = BcdInteger.succ big_int
 

output

 
> coffee foo.coffee | head -5
199999999999999999999999999999999999999999999999999999
200000000000000000000000000000000000000000000000000000
200000000000000000000000000000000000000000000000000001
200000000000000000000000000000000000000000000000000002
200000000000000000000000000000000000000000000000000003
 


Common Lisp[edit]

(loop for i from 1 do (print i))

If your compiler does tail call elimination (note: this has absolutely no advantage over normal loops):

(defun pp (x) (pp (1+ (print x))))
(funcall (compile 'pp) 1) ; it's less likely interpreted mode will eliminate tails

Component Pascal[edit]

BlackBox Component Builder

 
MODULE IntegerSequence;
IMPORT StdLog;
 
PROCEDURE Do*;
VAR
i: INTEGER;
BEGIN
FOR i := 0 TO MAX(INTEGER) DO;
StdLog.Int(i)
END;
StdLog.Ln
END Do;
 
END IntegerSequence.
 

Execute: ^Q IntegerSequence.Do
Output:

 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 ...

Computer/zero Assembly[edit]

This program counts up to 255 in the accumulator, after which it starts again from zero.

start:  ADD  one
JMP start
one: 1

D[edit]

import std.stdio, std.bigint;
 
void main() {
BigInt i;
while (true)
writeln(++i);
}

Alternative:

import std.stdio, std.traits, std.bigint, std.string;
 
void integerSequence(T)() if (isIntegral!T || is(T == BigInt)) {
T now = 1;
T max = 0;
static if (!is(T == BigInt))
max = T.max;
 
do
write(now, " ");
while (now++ != max);
 
writeln("\nDone!");
}
 
void main() {
writeln("How much time do you have?");
writeln(" 0. I'm in hurry.");
writeln(" 1. I've some time.");
writeln(" 2. I'm on vacation.");
writeln(" 3. I'm unemployed...");
writeln(" 4. I'm immortal!");
write("Enter 0-4 or nothing to quit: ");
 
string answer;
readf("%s\n", &answer);
 
switch (answer.toLower()) {
case "0": integerSequence!ubyte(); break;
case "1": integerSequence!short(); break;
case "2": integerSequence!uint(); break;
case "3": integerSequence!long(); break;
case "4": integerSequence!BigInt(); break;
default: writeln("\nBye bye!"); break;
}
}

Dc[edit]

1[p1+lpx]dspx

DCL[edit]

$ i = 1
$ loop:
$ write sys$output i
$ i = i + 1
$ goto loop
Output:
1
2
3
...
2147483646
2147483647
-2147483648
-2147483647
...
-1
0
1
...

Delphi[edit]

program IntegerSequence;
 
{$APPTYPE CONSOLE}
 
var
i: Integer;
begin
for i := 1 to High(i) do
WriteLn(i);
end.

Déjà Vu[edit]

1
 
while /= -- dup dup:
!. dup
++
 
drop

This continues to print numbers until double precision IEEE 754 cannot represent adjacent integers any more (9007199254740992, to be exact).

In the future, the implementation may switch to arbitrary precision, so it will keep running until memory fills up.

DWScript[edit]

High(i) returns the maximum supported value, typically, it is the highest signed 64 bit integer.

 
var i: Integer;
 
for i:=1 to High(i) do
PrintLn(i);
 

E[edit]

for i in int > 0 { println(i) }

EchoLisp[edit]

 
(lib 'bigint) ;; arbitrary length integers
(for ((n (in-naturals))) (writeln n))
 

EDSAC order code[edit]

[ Integer sequence
================
 
A program for the EDSAC
 
Displays integers 1,2,3...
in binary form in the first
word of storage tank 2
until stopped
 
Works with Initial Orders 2 ]
 
T56K [ set load point ]
GK [ set base address ]
 
A3@ [ increment accumulator ]
U64F [ copy accumulator to 64 ]
E@ [ jump to base address ]
 
P0D [ constant: 1 ]
 
EZPF [ begin at load point ]

Eiffel[edit]

 
class
APPLICATION
inherit
ARGUMENTS
create
make
feature {NONE} -- Initialization
make
-- Run application.
do
from
number := 0
until
number = number.max_value
loop
print(number)
print(", ")
number := number + 1
end
end
number:INTEGER_64
end
 

Elixir[edit]

Stream.iterate(1, &(&1+1)) |> Enum.each(&(IO.puts &1))

Emacs Lisp[edit]

Displays in the message area interactively, or to standard output under -batch.

(dotimes (i most-positive-fixnum)
(message "%d" (1+ i)))

Erlang[edit]

 F = fun(FF, I) -> io:format("~p~n", [I]), FF(FF, I + 1) end, F(F,0). 

ERRE[edit]

 
.............
A%=0
LOOP
A%=A%+1
PRINT(A%;)
END LOOP
.............
 

% is integer-type specificator. Integer type works on 16-bit signed numbers (reserved constant MAXINT is 32767). Beyond this limit execution will give Runtime error #6 (overflow).

Euphoria[edit]

integer i
i = 0
while 1 do
? i
i += 1
end while

F#[edit]

// lazy sequence of integers starting with i
let rec integers i =
seq { yield i
yield! integers (i+1) }
 
Seq.iter (printfn "%d") (integers 1)

lazy sequence of int32 starting from 0

let integers = Seq.initInfinite id

lazy sequence of int32 starting from n

let integers n = Seq.initInfinite ((+) n)

lazy sequence (not necessarily of int32) starting from n (using unfold anamorphism)

let inline numbers n =
Seq.unfold (fun n -> Some (n, n + LanguagePrimitives.GenericOne)) n
> numbers 0 |> Seq.take 10;;
val it : seq<int> = seq [0; 1; 2; 3; ...]
> let bignumber = 12345678901234567890123456789012345678901234567890;;
val bignumber : System.Numerics.BigInteger =
  12345678901234567890123456789012345678901234567890
> numbers bignumber |> Seq.take 10;;
val it : seq<System.Numerics.BigInteger> =
 seq
   [12345678901234567890123456789012345678901234567890 {IsEven = true;
                                                        IsOne = false;
                                                        IsPowerOfTwo = false;
                                                        IsZero = false;
                                                        Sign = 1;};
    12345678901234567890123456789012345678901234567891 {IsEven = false;
                                                        IsOne = false;
                                                        IsPowerOfTwo = false;
                                                        IsZero = false;
                                                        Sign = 1;};
    12345678901234567890123456789012345678901234567892 {IsEven = true;
                                                        IsOne = false;
                                                        IsPowerOfTwo = false;
                                                        IsZero = false;
                                                        Sign = 1;};
    12345678901234567890123456789012345678901234567893 {IsEven = false;
                                                        IsOne = false;
                                                        IsPowerOfTwo = false;
                                                        IsZero = false;
                                                        Sign = 1;}; ...]
> numbers 42.42 |> Seq.take 10;;
val it : seq<float> = seq [42.42; 43.42; 44.42; 45.42; ...]

Factor[edit]

USE: lists.lazy
1 lfrom [ . ] leach

Fantom[edit]

 
class Main
{
public static Void main()
{
i := 1
while (true)
{
echo (i)
i += 1
}
}
}
 

Fantom's integers are 64-bit signed, and so the numbers will return to 0 and continue again, if you wait long enough! You can use Java BigInteger via FFI

Fish[edit]

Since there aren't really libraries in Fish and I wouldn't know how to program arbitarily large integers, so here's an example that just goes on until the interpreter's number limit:

0>:n1+v
^o" "<

Forth[edit]

: ints ( -- )
0 begin 1+ dup cr u. dup -1 = until drop ;

Fortran[edit]

Works with: Fortran version 90 and later
program Intseq
implicit none
 
integer, parameter :: i64 = selected_int_kind(18)
integer(i64) :: n = 1
 
! n is declared as a 64 bit signed integer so the program will display up to
! 9223372036854775807 before overflowing to -9223372036854775808
do
print*, n
n = n + 1
end do
end program

FreeBASIC[edit]

' FB 1.05.0 Win64
 
' FB does not natively support arbitrarily large integers though support can be added
' by using an external library such as GMP. For now we will just use an unsigned long integer.
 
Print "Press Ctrl + C to stop the program at any time"
Dim i As Long = 1
 
Do
Print i
i += 1
Loop Until i = 0 ' will wrap back to 0 when it reaches 4,294,967,296
 
Sleep

Frink[edit]

All of Frink's numbers can be arbitrarily-sized:

 
i=0
while true
{
println[i]
i = i + 1
}
 

FunL[edit]

The following has no limit since FunL has arbitrary size integers.

for i <- 1.. do println( i )

Futhark[edit]

Infinite loops cannot produce results in Futhark, so this program accepts an input indicating how many integers to generate. It encodes the size of the returned array in its type.

 
fun main(n: int): [n]int = iota n
 

GAP[edit]

InfiniteLoop := function()
local n;
n := 1;
while true do
Display(n);
n := n + 1;
od;
end;
 
# Prepare some coffee
InfiniteLoop();

Go[edit]

Size of int type is implementation dependent. After the maximum positive value, it rolls over to maximum negative, without error. Type uint will roll over to zero.

package main
 
import "fmt"
 
func main() {
for i := 1;; i++ {
fmt.Println(i)
}
}

The big.Int type does not roll over and is limited only by available memory, or practically, by whatever external factor halts CPU execution: human operator, lightning storm, CPU fan failure, heat death of universe, etc.

package main
 
import (
"big"
"fmt"
)
 
func main() {
one := big.NewInt(1)
for i := big.NewInt(1);; i.Add(i, one) {
fmt.Println(i)
}
}

Gridscript[edit]

 
#INTEGER SEQUENCE.
 
@width
@height 1
 
(1,1):START
(3,1):STORE 1
(5,1):CHECKPOINT 0
(7,1):PRINT
(9,1):INCREMENT
(11,1):GOTO 0
 

GUISS[edit]

Graphical User Interface Support Script makes use of installed programs. There are no variables, no loop structures and no jumps within the language so iteration is achieved by repetative instructions. In this example, we will just use the desktop calculator and keep adding one to get a counter. We stop after counting to ten in this example.

Start,Programs,Accessories,Calculator,
Button:[plus],Button:1,Button:[equals],Button:[plus],Button:1,Button:[equals],
Button:[plus],Button:1,Button:[equals],Button:[plus],Button:1,Button:[equals],
Button:[plus],Button:1,Button:[equals],Button:[plus],Button:1,Button:[equals],
Button:[plus],Button:1,Button:[equals],Button:[plus],Button:1,Button:[equals],
Button:[plus],Button:1,Button:[equals],Button:[plus],Button:1,Button:[equals]

Haskell[edit]

mapM_ print [1..]

Or less imperatively:

putStr $ unlines $ map show [1..]

Icon and Unicon[edit]

Icon and Unicon support large integers by default. The built-in generator seq(i,j) yields the infinite sequence i, i+j, i+2*j, etc. Converting the results to strings for display will likely eat your lunch before the sequence will take its toll.

procedure main()
every write(seq(1)) # the most concise way
end

J[edit]

The following will count indefinitely but once the 32-bit (or 64-bit depending on J engine version) limit is reached, the results will be reported as floating point values (which would immediately halt on 64 bit J and halt with the 53 bit precision limit is exceeded on 32 bit J). Since that could take many, many centuries, even on a 32 bit machine, more likely problems include the user dying of old age and failing to pay the electric bill resulting in the machine being powered off.

 count=: (smoutput ] >:)^:_

The above works with both fixed sized integers and floating point numbers (fixed sized integers are automatically promoted to floating point, if they overflow), but also works with extended precision integers (which will not overflow, unless they get so large that they cannot be represented in memory, but that should exceed lifetime of the universe, let alone lifetime of the computer).

This adds support for extended precision (in that it converts non-extended precision arguments to extended precision arguments) and will display integers to ∞ (or at least until the machine is turned off or interrupted or crashes).

 count=: (smoutput ] >:)@x:^:_

Java[edit]

Long limit:

public class Count{
public static void main(String[] args){
for(long i = 1; ;i++) System.out.println(i);
}
}

"Forever":

import java.math.BigInteger;
 
public class Count{
public static void main(String[] args){
for(BigInteger i = BigInteger.ONE; ;i = i.add(BigInteger.ONE)) System.out.println(i);
}
}

JavaScript[edit]

var i = 0;
 
while (true)
document.write(++i + ' ');

Joy[edit]

 
1 [0 >] [dup put succ] while pop.

Counting stops at maxint, which is 2147483647

jq[edit]

Currently, julia does not support infinite-precision arithmetic, but very large integers are converted to floating-point numbers, so the following will continue to generate integers (beginning with 0) indefinitely in recent versions of jq that have tail recursion optimization:

def iota: ., (. + 1 | iota);
0 | iota
In versions of jq which have while, one could also write:
0 | while(true;. + 1)
This idiom is likely to be more useful as while supports break.

Another technique would be to use recurse:

0 | recurse(. + 1)
For generating integers, the generator, range(m;n), is more likely to be useful in practice; if m and n are integers, it generates integers from m to n-1, inclusive.

Julia[edit]

i = zero(BigInt)    # or i = big(0)
while true
println(i += 1)
end

The built-in BigInt type is an arbitrary precision integer (based on the GMP library), so the value of i is limited only by available memory. To use (much faster) hardware fixed-width integer types, use e.g. zero(Int32) or zero(Int64). (Initializing i = 0 will use fixed-width integers that are the same size as the hardware address width, e.g. 64-bit on a 64-bit machine.)

K[edit]

  {`0:"\n",$x+:1;x}/1

Using a while loop:

  i:0; while[1;`0:"\n",$i+:1]

Lang5[edit]

0 do dup . 1 + loop

Lasso[edit]

local(number = 1)
while(#number > 0) => {^
#number++
' '
//#number > 100 ? #number = -2 // uncomment this row if you want to halt the run after proving concept
^}

This will run until you exhaust the system resources it's run under.

Liberty BASIC[edit]

Liberty BASIC handles extremely large integers. The following code was halted by user at 10,000,000 in test run.

 while 1
i=i+1
locate 1,1
print i
scan
wend
 

Limbo[edit]

The int (32 bits) and big (64 bits) types are both signed, so they wrap around. This version uses the infinite precision integer library:

implement CountToInfinity;
 
include "sys.m"; sys: Sys;
include "draw.m";
include "ipints.m"; ipints: IPints;
IPint: import ipints;
 
CountToInfinity: module {
init: fn(nil: ref Draw->Context, nil: list of string);
};
 
init(nil: ref Draw->Context, nil: list of string)
{
sys = load Sys Sys->PATH;
ipints = load IPints IPints->PATH;
 
i := IPint.inttoip(0);
one := IPint.inttoip(1);
for(;;) {
sys->print("%s\n", i.iptostr(10));
i = i.add(one);
}
}
 

Lingo[edit]

i = 1
repeat while i>0
put i
i = i+1
end repeat

Lingo uses signed 32 bit integers, so max. supported integer value is 2147483647:

put the maxInteger
-- 2147483647

Beyond this limit values behave like negative numbers:

put the maxInteger+1
-- -2147483648
put the maxInteger+2
-- -2147483647

Up to the (quite high) number where floats (double-precission) start rounding, floats can be used to exceed the integer limit:

the floatPrecision = 0 -- forces floats to be printed without fractional digits
 
put float(the maxInteger)+1
-- 2147483648
 
-- max. whole value that can be stored as 8-byte-float precisely
maxFloat = power(2,53) -- 9007199254740992.0
 
i = 1.0
repeat while i<=maxFloat
put i
i = i+1
end repeat
-- 1
-- 2
-- 3
-- ...

Lua[edit]

 
i = 1
 
-- in the event that the number inadvertently wraps around,
-- stop looping - this is unlikely with Lua's default underlying
-- number type (double), but on platform without double
-- the C type used for numbers can be changed
while i > 0 do
print( i )
i = i + 1
end
 

Maple[edit]

Maple has arbitrary-precision integers so there are no built-in limits on the size of the integers represented.

for n do
print(n)
end do;

Mathematica / Wolfram Language[edit]

Built in arbitrary precision support meanst the following will not overflow.

 
x = 1;
Monitor[While[True, x++], x]
 

MATLAB / Octave[edit]

 a = 1; while (1) printf('%i\n',a); a=a+1; end; 

Typically, numbers are stored as double precision floating point numbers, giving accurate integer values up to about 2^53=bitmax('double')=9.0072e+15. Above this limit, round off errors occur. This limitation can be overcome by defining the numeric value as type int64 or uint64

 a = uint64(1); while (1) printf('%i\n',a); a=a+1; end; 

This will run up to 2^64 and then stop increasing, there will be no overflow.

>> a=uint64(10e16+1)    % 10e16 is first converted into a double precision number causing some round-off error. 
a = 100000000000000000
>> a=uint64(10e16)+1
a = 100000000000000001

The above limitations can be overcome with additional toolboxes for symbolic computation or multiprecision computing.

Matlab and Octave recommend vectorizing the code, one might pre-allocate the sequence up to a specific N.

  N = 2^30; printf('%d\n', 1:N);    

The main limitation is the available memory on your machine. The standard version of Octave has a limit that a single data structure can hold at most 2^31 elements. In order to overcome this limit, Octave must be compiled with "./configure --enable-64", but this is currently not well supported.

Maxima[edit]

for i do disp(i);

МК-61/52[edit]

1	П4	ИП4	С/П	КИП4	БП	02

ML/I[edit]

MCSKIP "WITH" NL
"" Integer sequence
"" Will overflow when it reaches implementation-defined signed integer limit
MCSKIP MT,<>
MCINS %.
MCDEF DEMO WITHS NL AS <MCSET T1=1
%L1.%T1.
MCSET T1=T1+1
MCGO L1
>
DEMO

NetRexx[edit]

Rexx Built In[edit]

NetRexx provides built-in support for very large precision arithmetic via the Rexx class.

/* NetRexx */
options replace format comments java crossref symbols binary
 
k_ = Rexx
bigDigits = 999999999 -- Maximum setting for digits allowed by NetRexx
numeric digits bigDigits
 
loop k_ = 1
say k_
end k_
 

Using BigInteger[edit]

Java's BigInteger class is also available for very large precision arithmetic.

/* NetRexx */
options replace format comments java crossref symbols binary
 
import java.math.BigInteger
 
-- allow an option to change the output radix.
parse arg radix .
if radix.length() == 0 then radix = 10 -- default to decimal
k_ = BigInteger
k_ = BigInteger.ZERO
 
loop forever
k_ = k_.add(BigInteger.ONE)
say k_.toString(int radix)
end
 

NewLISP[edit]

(while (println (++ i)))

Nim[edit]

var i:int64 = 0
while true:
inc i
echo i

Using BigInts:

import bigints
 
var i = 0.initBigInt
while true:
i += 1
echo i

Oberon-2[edit]

Works with oo2c Version 2

 
MODULE IntegerSeq;
IMPORT
Out,
Object:BigInt;
 
PROCEDURE IntegerSequence*;
VAR
i: LONGINT;
BEGIN
FOR i := 0 TO MAX(LONGINT) DO
Out.LongInt(i,0);Out.String(", ")
END;
Out.Ln
END IntegerSequence;
 
PROCEDURE BigIntSequence*;
VAR
i: BigInt.BigInt;
BEGIN
i := BigInt.zero;
LOOP
Out.Object(i.ToString() + ", ");
i := i.Add(BigInt.one);
END
END BigIntSequence;
 
END IntegerSeq.
 

Objeck[edit]

 
bundle Default {
class Count {
function : Main(args : String[]) ~ Nil {
i := 0;
do {
i->PrintLine();
i += 1;
} while(i <> 0);
}
}
}
 

OCaml[edit]

with an imperative style:

let () =
let i = ref 0 in
while true do
print_int !i;
print_newline ();
incr i;
done

with a functional style:

let () =
let rec aux i =
print_int i;
print_newline ();
aux (succ i)
in
aux 0

Oforth[edit]

Oforth handles arbitrary integer precision.

The loop will stop when out of memory

: integers  1 while( true ) [ dup . 1+ ] ;

OpenEdge/Progress[edit]

OpenEdge has three data types that can be used for this task:

  1. INTEGER (32-bit signed integer)
    DEF VAR ii AS INTEGER FORMAT "->>>>>>>>9" NO-UNDO.
     
    DO WHILE TRUE:
    ii = ii + 1.
    DISPLAY ii.
    END.

    When an integer rolls over its maximum of 2147483647 error 15747 is raised (Value # too large to fit in INTEGER.).

  2. INT64 (64-bit signed integer)
    DEF VAR ii AS INT64 FORMAT "->>>>>>>>>>>>>>>>>>9" NO-UNDO.
     
    DO WHILE TRUE:
    ii = ii + 1.
    DISPLAY ii.
    END.

    When a 64-bit integer overflows no error is raised and the signed integer becomes negative.

  3. DECIMAL (50 digits)
    DEF VAR de AS DECIMAL FORMAT "->>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>9" NO-UNDO.
     
    DO WHILE TRUE:
    de = de + 1.
    DISPLAY de.
    END.

    When a decimal requires mores than 50 digits error 536 is raised (Decimal number is too large.).

Order[edit]

Order supports arbitrarily-large positive integers natively. However, the simple version:

#include <order/interpreter.h>
 
#define ORDER_PP_DEF_8printloop ORDER_PP_FN( \
8fn(8N, \
8do(8print(8to_lit(8N) 8comma 8space), \
8printloop(8inc(8N)))) )

 
ORDER_PP( 8printloop(1) )

... while technically fulfilling the task, will probably never display anything, as most C Preprocessor implementations won't print their output until the file is done processing. Since the C Preprocessor is not technically Turing-complete, the Order interpreter has a maximum number of steps it can execute - but this number is very, very large (from the documentation: "the Order interpreter could easily be extended with a couple of hundred macros to prolong the wait well beyond the estimated lifetime of the sun"), so the compiler is rather more likely to simply run out of memory.

To actually see anything with GCC, add a maximum limit so that the task can complete:

#include <order/interpreter.h>
 
#define ORDER_PP_DEF_8printloop ORDER_PP_FN( \
8fn(8N, \
8do(8print(8to_lit(8N) 8comma 8space), \
8when(8less(8N, 99), 8printloop(8inc(8N))))) )

 
ORDER_PP( 8printloop(1) ) // 1, ..., 99,

PARI/GP[edit]

n=0; while(1,print(++n))

Pascal[edit]

See also Delphi

Works with: Free_Pascal

Quad word has the largest positive range of all ordinal types

Program IntegerSequenceLimited;
var
Number: QWord = 0; // 8 bytes, unsigned: 0 .. 18446744073709551615
begin
repeat
writeln(Number);
inc(Number);
until false;
end.
Library: GMP

With the gmp library your patience is probably the limit :-)

Program IntegerSequenceUnlimited;
 
uses
gmp;
 
var
Number: mpz_t;
 
begin
mpz_init(Number); //* zero now *//
repeat
mp_printf('%Zd' + chr(13) + chr(10), @Number);
mpz_add_ui(Number, Number, 1); //* increase Number *//
until false;
end.

Perl[edit]

my $i = 0;
print ++$i, "\n" while 1;

On 64-bit Perls this will get to 2^64-1 then print 1.84467440737096e+19 forever. On 32-bit Perls using standard doubles this will get to 999999999999999 then start incrementing and printing floats until they lose precision. This behavior can be changed by adding something like:

use bigint;
my $i = 0; print ++$i, "\n" while 1;

which makes almost all integers large (ranges are excluded). Faster alternatives exist with non-core modules, e.g.

  • use bigint lib=>"GMP";
  • use Math::Pari qw/:int/;
  • use Math::GMP qw/:constant/;

Perl 6[edit]

.say for 1..*

Phix[edit]

This will crash at 1,073,741,824 on 32 bit, 4,611,686,018,427,387,904 on 64-bit:

integer i = 0
while 1 do
 ?i
i += 1
end while

This will stall at 9,007,199,254,740,992 on 32-bit, and about twice the above on 64-bit. (after ~15 or 19 digits of precision, adding 1 will simply cease to have any effect)

atom a = 0
while 1 do
 ?a
a += 1
end while

This will probably carry on until the number has over 300 million digits (32-bit, you can square that on 64-bit) which would probably take zillions of times longer than the universe has already existed, if your hardware/OS/power grid kept going that long.

include bigatom.e
bigatom b = ba_new(0)
while 1 do
ba_printf(1,"%B\n",b)
b = ba_add(b,1)
end while

PicoLisp[edit]

(for (I 1 T (inc I))
(printsp I) )

Piet[edit]

Rendered as a Wiki table because uploading images is not possible.

ww ww ww ww ww ww ww
ww ww ww ww ww ww ww
ww ww ww ww ww ww ww

Program explanation on my user page: [1]

PL/I[edit]

 
infinity: procedure options (main);
declare k fixed decimal (30);
put skip edit
((k do k = 1 to 999999999999999999999999999998))(f(31));
end infinity;
 

PostScript[edit]

Library: initlib
 
1 {succ dup =} loop
 

Prolog[edit]

loop(I) :-
writeln(I),
I1 is I+1,
loop(I1).
 

Constraint Handling Rules[edit]

Works with SWI-Prolog and library CHR written by Tom Schrijvers and Jan Wielemaker

:- use_module(library(chr)).
 
:- chr_constraint loop/1.
 
loop(N) <=> writeln(N), N1 is N+1, loop(N1).
 

PowerShell[edit]

try
{
for ([int]$i = 0;;$i++)
{
$i
}
}
catch {break}

PureBasic[edit]

OpenConsole()
Repeat
a.q+1
PrintN(Str(a))
ForEver

Python[edit]

i=1
while i:
print(i)
i += 1

Or, alternatively:

from itertools import count
 
for i in count():
print(i)

Pythons integers are of arbitrary large precision and so programs would probably keep going until OS or hardware system failure.

R[edit]

z <- 0
repeat {
print(z)
z <- z + 1
}

Racket[edit]

Racket uses bignums, so counting should continue up to very large numbers. Naturally, printing these numbers will consume quite a bit of power.

#lang racket
(for ([i (in-naturals)]) (displayln i))
 

Raven[edit]

Raven uses signed 32 bit integer values.

1 as $i
repeat TRUE while
$i "%d\n" print $i 1000 + as $i

Retro[edit]

Retro uses signed integer values.

0 [ [ putn space ] sip 1+ dup 0 <> ] while drop

REXX[edit]

/*count all the protons, electrons, & whatnot in the universe, and then */
/*keep counting. According to some pundits in-the-know, one version of */
/*the big-bang theory is that the universe will collapse back to where */
/*it started, and this computer program will be still counting. */
/*┌────────────────────────────────────────────────────────────────────┐
│ Count all the protons (and electrons!) in the universe, and then │
│ keep counting. According to some pundits in-the-know, one version │
│ of the big-bang theory is that the universe will collapse back to │
│ where it started, and this computer program will still be counting.│
│ │
│ │
│ According to Sir Arthur Eddington in 1938 at his Tamer Lecture at │
│ Trinity College (Cambridge), he postulated that there are exactly │
│ │
│ 136 ∙ 2^256 │
│ │
│ protons in the universe and the same number of electrons, which is │
│ equal to around 1.57477e+79. │
│ │
│ Although, a modern estimate is around 10^80. │
│ │
│ │
│ One estimate of the age of the universe is 13.7 billion years, │
│ or 4.32e+17 seconds. This'll be a piece of cake. │
└────────────────────────────────────────────────────────────────────┘*/

numeric digits 1000000000 /*just in case the universe slows down. */
 
/*this version of a DO loop increments J*/
do j=1 /*Sir Eddington's number, then a googol.*/
say j /*first, destroy some electrons. */
end
say 42 /*(see below for explanation of 42.) */
exit
 
/*This REXX program (as it will be limited to the NUMERIC DIGITS above, */
/*will only count up to 1000000000000000000000000000000000000000000... */
/*000000000000000000000000000000000000000000000000000000000000000000000 */
/* ... for another (almost) one billion more zeroes (then subtract 1).*/
 
/*if we can count 1,000 times faster than the fastest PeeCee, and we */
/*started at the moment of the big-bang, we'd be at only 1.72e+28, so */
/*we still have a little ways to go, eh? */
 
/*To clarify, we'd be 28 zeroes into a million zeroes. If PC's get */
/*1,000 times faster again, that would be 31 zeroes into a million. */
 
/*It only took Deep Thought 7.5 million years to come up with the */
/*answer to everything (and it double-checked the answer). It was 42.*/

Ring[edit]

 
size = 10
 
for n = 1 to size
see n + nl
next
see nl
 
for n in [1:size]
see n + nl
next
see nl
 
i = n
while n <= size
see n + nl
n = n + 1
end
 

Ruby[edit]

1.step{|n| puts n}

The step method of Numeric takes two optional arguments. The limit defaults to infinity, the step size to 1.

Ruby does not limit the size of integers.

Run BASIC[edit]

while 1
i = i + 1
print i
wend

Eventually as it gets larger it becomes a floating point.

Rust[edit]

Works with: Rust 1.2
fn main() {
for i in 0.. {
println!("{}", i);
}
}


Looping endlessly:

extern crate num;
 
use num::bigint::BigUint;
use num::traits::{One,Zero};
 
fn main() {
let mut i: BigUint = BigUint::one();
loop {
println!("{}", i);
i = i + BigUint::one();
}
}

Salmon[edit]

Salmon has built-in unlimited-precision integer arithmetic, so these examples will all continue printing decimal values indefinitely, limited only by the amount of memory available (it requires O(log(n)) bits to store an integer n, so if your computer has 1 GB of memory, it will count to a number with on the order of digits).

iterate (i; [0...+oo])
i!;

or

for (i; 0; true)
i!;

or

variable i := 0;
while (true)
{
i!;
++i;
};

Scala[edit]

Stream from 1 foreach println

Scheme[edit]

 
(let loop ((i 1))
(display i) (newline)
(loop (+ 1 i)))
 

Scheme does not limit the size of numbers.

Seed7[edit]

Limit 2147483647:

$ include "seed7_05.s7i";
 
const proc: main is func
local
var integer: number is 0;
begin
repeat
incr(number);
writeln(number);
until number = 2147483647;
end func;

"Forever":

$ include "seed7_05.s7i";
include "bigint.s7i";
 
const proc: main is func
local
var bigInteger: number is 1_;
begin
repeat
writeln(number);
incr(number);
until FALSE;
end func;

Sidef[edit]

No limit:

{|i| say i } * Math.inf;

Smalltalk[edit]

i := 0.
[
Stdout print:i; cr.
i := i + 1
] loop

will run forever.

SSEM[edit]

Since we have no Add instruction, we subtract -1 on each iteration instead of adding 1. The same -1 also serves as a jump target, taking advantage of a quirk of the SSEM architecture (the Current Instruction counter is incremented after the instruction has been executed, not before—so GOTO address has to be coded as GOTO address - 1).

01000000000000010000000000000000   0. Sub. 2     acc -= -1
01000000000000000000000000000000 1. 2 to CI goto -1 + 1
11111111111111111111111111111111 2. -1

SuperCollider[edit]

The SuperCollider language has a 32-bit signed int, and a 64 bit signed float. Instead of locking the interpreter with an infinite loop, we post the values over time.

 
i = Routine { inf.do { |i| i.yield } }; // return all integers, represented by a 64 bit signed float.
j = { inf.do { i.next.postln; 0.01.wait } }; // this prints them incrementally
j.play;
 


Swift[edit]

var i = 0
while true {
println(i++)
}

Standard ML[edit]

This will print up to Int.maxInt and then crash. On a 32 bit machine the max is 1073741823. Alternatively you could use Int64.int (64 bit) or IntInf.int (arbitrary precision).

let
fun printInts(n) =
(
print(Int.toString(n) ^ "\n");
printInts(n+1)
)
in
printInts(1)
end;

Tcl[edit]

package require Tcl 8.5
while true {puts [incr i]}

TUSCRIPT[edit]

$$ MODE TUSCRIPT
LOOP n=0,999999999
n=n+1
ENDLOOP

UNIX Shell[edit]

#!/bin/sh
num=0
while true; do
echo $num
num=`expr $num + 1`
done

Ursa[edit]

#
# integer sequence
#
 
# declare an int and loop until it overflows
decl int i
set i 1
while true
out i endl console
inc i
end while

Vala[edit]

 
uint i = 0;
while (++i < uint.MAX)
stdout.printf("%u\n", i);
 

Visual Basic .NET[edit]

Visual Basic .NET supports an unsigned, 64 bit Integer (maxing out at a whopping 9 223 372 036 854 775 807), however, this is not an intrinsic type, it is a structure that is not supported by the CLS (Common Language Specification).

The CLS supported type (also a structure) is Decimal (an even more impressive range from positive 79 228 162 514 264 337 593 543 950 335 to negative 79 228 162 514 264 337 593 543 950 335), I have used a standard CLS Integer intrinsic type (from -2 147 483 648 through 2 147 483 647).

Note that attempting to store any value larger than the maximum value of any given type (say 2 147 483 648 for an Integer) will result in an OverflowException being thrown ("Arithmetic operation resulted in an overflow.")

    For i As Integer = 0 To Integer.MaxValue
Console.WriteLine(i)
Next

XPL0[edit]

\Displays integers up to 2^31-1 = 2,147,483,647
code CrLf=9, IntOut=11;
int N;
[N:= 1;
repeat IntOut(0, N); CrLf(0);
N:= N+1;
until N<0;
]

zkl[edit]

[1..].pump(Console.println)  // eager
m:=(1).MAX; [1..m].pump(Console.println) // (1).MAX is 9223372036854775807
[1..].pump(100,Console.println) // lazy