I'm working on modernizing Rosetta Code's infrastructure. Starting with communications. Please accept this time-limited open invite to RC's Slack.. --Michael Mol (talk) 20:59, 30 May 2020 (UTC)

# Integer sequence

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

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.

## 0815

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

## 360 Assembly

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

`*        Integer sequence      06/05/2016INTSEQED CSECT         USING  INTSEQED,12         LR    12,15         LA    6,1             i=1LOOP     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 loopWTOMSG   DC    0F,H'80',H'0',CL80' 'DW       DS    0D,PL8          pack dec 15numEM12     DC    X'402020202020202020202120'  mask CL12 11num         END   INTSEQED`
Output:
```...
314090
314091
314092
314093
314094
314095
314096
314097
314098
314099
...
```

## 8080 Assembly

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

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

A more complex, arbitrary precision version that can count as high as you have free bytes of memory to use. (This does assemble with CP/M's MAC assembler, but since it doesn't implement PRBUFR, it's only useful for exposition purposes, or for loading into DDT.)

`         ORG     0100HBITS    EQU     128       ; 128 bits of precisionBYTES   EQU     BITS / 8  ; Number of bytes we store those bits in         ; Zero out the storage for our number        LXI     H,BUFR    ; HL points at BUFR. (HL is idiomatically used for pointers)        MVI     C,BYTES   ; C holds the number of bytes we'll use        XRA     A         ; XOR with A is a 1-byte instruction to set A to zeroINIT:   MOV     M,A       ; Store 0 to address pointed to by HL        INX     H         ; Advance HL to the next byte        DCR     C         ; Count down        JNZ     INIT      ; Keep looping if we're not done         ; The "very long integer" is zeroed, so start the loopLOOP:   CALL    PRBUFR    ; Output our number        LXI     H,BUFR    ; HL Points to BUFR        MVI     C,BYTES   ; Count down (assume fewer than 256 bytes in our integer)NEXT:   INR     M         ; Increment the byte pointed to by HL. Sets the zero flag        JNZ     LOOP      ; If the increment didn't overflow A, start the loop over                          ; This byte overflowed, so we need to advance to the next byte in our number        INX     H         ; We store our byes in order of increasing significance        DCR     C         ; Count down to make sure we don't overflow our buffer        JNZ     NEXT      ; jump to process the next, more significant byte         ; If we get here, we have overflowed our integer!        HALT              ; TODO: probably something other than "halt the CPU" PRBUFR: ; TODO, a subroutine that shows all of the digits in BUFR on the console        ; Assume that this code trashes all our registers...        RET BUFR:   ; This space will hold our number        ; We zero this memory before the loop        END`

`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 isbegin   for Value in Positive'Range loop      Ada.Text_IO.Put_Line (Positive'Image (Value));   end loop;end Positives;`

## ALGOL 68

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

`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

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, 32651, 32652, 32653, 32654, 32655, 32656, 32657, 32658, 32659, 32660, 32661, 32662, 32663, 32664, 32665, 32666, 32667, 32668, 32669, 32670, 32671, 32672, 32673, 32674, 32675, 32676, 32677, 32678, 32679, 32680, 32681, 32682, 32683, 32684, 32685, 32686, 32687, 32688, 32689, 32690, 32691, 32692, 32693, 32694, 32695, 32696, 32697, 32698, 32699, 32700, 32701, 32702, 32703, 32704, 32705, 32706, 32707, 32708, 32709, 32710, 32711, 32712, 32713, 32714, 32715, 32716, 32717, 32718, 32719, 32720, 32721, 32722, 32723, 32724, 32725, 32726, 32727, 32728, 32729, 32730, 32731, 32732, 32733, 32734, 32735, 32736, 32737, 32738, 32739, 32740, 32741, 32742, 32743, 32744, 32745, 32746, 32747, 32748, 32749, 32750, 32751, 32752, 32753, 32754, 32755, 32756, 32757, 32758, 32759, 32760, 32761, 32762, 32763, 32764, 32765, 32766, 32767                          ?ILLEGAL QUANTITY ERROR IN 30           ]`

## ARM Assembly

`.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`

## ArnoldC

`IT'S SHOWTIMEHEY CHRISTMAS TREE nYOU SET US UP @NO PROBLEMOSTICK AROUND @NO PROBLEMOTALK TO THE HAND nGET TO THE CHOPPER nHERE IS MY INVITATION nGET UP @NO PROBLEMOENOUGH TALKCHILLYOU HAVE BEEN TERMINATED`

## AutoHotkey

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

`x=0Loop    TrayTip, Count, % ++x`

## AWK

`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

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)End0→IRepeat getKey(0) Disp I▶Dec,i I++EndIf I=0`

## BASIC

Works with: ZX Spectrum Basic
`10 LET A = 020 LET A = A + 130 PRINT A40 GO TO 20`
Works with: QBasic
`A = 0DO: A = A + 1: PRINT A: LOOP 1`

## Batch File

Variables are limited to 32bit integer, capable of a maximum value of `2,147,483,647`

` @echo offset number=0:loopset /a number+=1echo %number%goto loop `
Output:
```...
2147483644
2147483645
2147483646
2147483647
-2147483648
-2147483647
-2147483646
-2147483645
...
```

## BBC BASIC

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

`while (++i) i`

## beeswax

Using an ordinary loop structure:

` qNP<_1>{d`

Using a jump instruction:

`[email protected]{[email protected]`

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

## Befunge

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

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***

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

`i = 1 loop {  p i  i = i + 1}`

## Burlesque

` [email protected] `

## C

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

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

OpenSSL provides arbitrarily large integers.

`#include <openssl/bn.h>		/* BN_*() */#include <openssl/err.h>	/* ERR_*() */#include <stdio.h>		/* fprintf(), puts() */ voidfail(const char *message){	fprintf(stderr, "%s: error 0x%08lx\n", ERR_get_error());	exit(1);} intmain(){	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#

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

## C++

`#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';}`

## ChucK

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 >>>;} `

## Clean

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

`(map println (next (range)))`

## COBOL

`       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

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 -5199999999999999999999999999999999999999999999999999999200000000000000000000000000000000000000000000000000000200000000000000000000000000000000000000000000000000001200000000000000000000000000000000000000000000000000002200000000000000000000000000000000000000000000000000003 `

## Common Lisp

`(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

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.LnEND 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

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

`start:  ADD  one        JMP  startone:         1`

## Crystal

Will run as long as enough memory to represent numbers.

`require "big" (1.to_big_i ..).each { |i| puts i } `

## D

`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

`1[p1+lpx]dspx`

## DCL

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

## Delphi

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

## DWScript

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); `

## Dyalect

`var n = 0while true {    n += 1    print(n)}`

## Déjà Vu

`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.

## E

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

## EchoLisp

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

## EDSAC order code

`[ 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       ] [email protected]   [ increment accumulator  ]U64F  [ copy accumulator to 64 ][email protected]    [ jump to base address   ] P0D   [ constant: 1            ] EZPF  [ begin at load point    ]`

## Eiffel

` class	APPLICATIONinherit	ARGUMENTScreate	makefeature {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_64end `

## Elena

ELENA 4.x :

`import extensions; public program(){    var i := 0u;    while (true)    {        console.printLine(i);         i += 1u    }}`

## Elixir

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

## Emacs Lisp

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

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

## Erlang

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

## ERRE

` .............A%=0LOOP  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

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

## F#

`// lazy sequence of integers starting with ilet 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

`USE: lists.lazy1 lfrom [ . ] leach`

## Fantom

` 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

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

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

## Fortran

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 doend program`

## FreeBASIC

`' 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 integer (32bit). Print "Press Ctrl + C to stop the program at any time"Dim i As UInteger = 1 Do  Print i  i += 1Loop Until i = 0 ' will wrap back to 0 when it reaches 4,294,967,296 Sleep`

## Frink

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

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

## FunL

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

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

## Futhark

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

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

## Go

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

` #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 `

## Groovy

`// 32-bit 2's-complement signed integer (int/Integer)for (def i = 1; i > 0; i++) { println i } // 64-bit 2's-complement signed integer (long/Long)for (def i = 1L; i > 0; i+=1L) { println i } // Arbitrarily-long binary signed integer (BigInteger)for (def i = 1g; ; i+=1g) { println i }`

## GUISS

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]`

`mapM_ print [1..]`

Or less imperatively:

`putStr \$ unlines \$ map show [1..]`

## HolyC

Prints from 1 to max unsigned 64 bit integer (2**64 -1), then stops.

`U64 i = 0;while (++i) Print("%d\n", i); `

## Icon and Unicon

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 wayend`

## IS-BASIC

`100 FOR I=1 TO INF110   PRINT I;120 NEXT`

INF = 9.999999999E62

## J

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

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

This code is accurate up to 2^53 where it will be stuck an 2^53 because a IEEE 64-bit double can not represent 2^53 + 1.

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

This example uses a BigInt literal to support arbitrary large integers.

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

## Joy

` 1 [0 >] [dup put succ] while pop.`

Counting stops at `maxint`, which is 2147483647

## jq

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

`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

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

Using a `while` loop:

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

## Kotlin

`import java.math.BigInteger // version 1.0.5-2 fun main(args: Array<String>) {    // print until 2147483647    (0..Int.MAX_VALUE).forEach { println(it) }     // print forever    var n = BigInteger.ZERO    while (true) {        println(n)        n += BigInteger.ONE    }}`

## Lambdatalk

The long_add primitive allow counting beyond the javascript numbers limits, depending on the system memory.

` {def infinite_set {lambda {:i}  {if true                                 // will never change   then :i {infinite_set {long_add :i 1}}  // extends {+ :i 1}   else You have reached infinity! }}}     // probably never.-> infinite_set {infinite_set 0}-> 0 1 2 3 ... forever `

## Lang5

`0 do dup . 1 + loop`

## Lasso

`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

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    scanwend `

## Limbo

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

`i = 1repeat while i>0  put i  i = i+1end 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-- -2147483648put 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 preciselymaxFloat = power(2,53) -- 9007199254740992.0 i = 1.0repeat while i<=maxFloat  put i  i = i+1end repeat-- 1-- 2-- 3-- ...`

## LLVM

Translation of: C
`; This is not strictly LLVM, as it uses the C library function "printf".; LLVM does not provide a way to print values, so the alternative would be; to just load the string into memory, and that would be boring. ; Additional comments have been inserted, as well as changes made from the output produced by clang such as putting more meaningful labels for the jumps ;--- The declarations for the external C functionsdeclare i32 @printf(i8*, ...) \$"FORMAT_STR" = comdat any@"FORMAT_STR" = linkonce_odr unnamed_addr constant [4 x i8] c"%u\0A\00", comdat, align 1 ; Function Attrs: noinline nounwind optnone uwtabledefine i32 @main() #0 {  %1 = alloca i32, align 4          ;-- allocate i  store i32 0, i32* %1, align 4     ;-- store i as 0  br label %loop loop:  %2 = load i32, i32* %1, align 4   ;-- load i  %3 = add i32 %2, 1                ;-- increment i  store i32 %3, i32* %1, align 4    ;-- store i  %4 = icmp ne i32 %3, 0            ;-- i != 0  br i1 %4, label %loop_body, label %exit loop_body:  %5 = load i32, i32* %1, align 4   ;-- load i  %6 = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @"FORMAT_STR", i32 0, i32 0), i32 %5)  br label %loop exit:  ret i32 0} attributes #0 = { noinline nounwind optnone uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" }`

## Lua

` 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 changedwhile i > 0 do    print( i )    i = i + 1end `

## M2000 Interpreter

` \\ easy way[email protected]\\ Def statement defines one time (second pass produce error)Rem : Def Decimal a=1Rem : Def a as decimal=1\\ Global shadow any global with same name, but not local\\ globals can change type, local can't change \\ to assign value to global need <=\\ Symbol = always make local variables (and shadows globals)Rem : Global a as decimal =1\\Local make a new local and shadow one with same nameRem : Local a as decimal=1\\ we can create an "auto rounding" variable\\ an integer with any type (double, single, decimal, currency, long, integer)\\ rounding to .5 : up for positive numbers and down to negative\\ 1.5 round to 2 and -1.5 round to -2a%[email protected] \\ variables a, a%, a\$, arrays/functions a(), a\$(), sub a() and the module a can exist together\\ A block may act as loop structure using an internal flag\\ A Loop statement mark a flag in the block, so can be anywhere inside,\\ this flag reset to false before restart.{loop : Print a : a++} `

## Maple

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

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

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

## MATLAB / Octave

` 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

`for i do disp(i);`

## min

Works with: min version 0.19.3

min's integers are 64-bit signed. This will eventually overflow.

`0 (dup) () (puts succ) () linrec`

## МК-61/52

`1	П4	ИП4	С/П	КИП4	БП	02`

## ML/I

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

## Modula-2

`MODULE Sequence;FROM FormatString IMPORT FormatString;FROM Terminal IMPORT WriteString,ReadChar; VAR    buf : ARRAY[0..63] OF CHAR;    i : CARDINAL;BEGIN    i := 1;    WHILE i>0 DO        FormatString("%c ", buf, i);        WriteString(buf);        INC(i)    END;    ReadCharEND Sequence.`

## Nanoquery

All native integers in Nanoquery can become arbitrarily large by default, so this program would run until it ran out of memory.

`i = 1while true	println i	i += 1end`

## Necromantus

In Necromantus integer size is limited by the java's int.

` let i = 0;while true{    write(i);    i = i + 1;} `

## NetRexx

### Rexx Built In

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

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

### Using BigInteger

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 decimalk_ = BigIntegerk_ = BigInteger.ZERO loop forever  k_ = k_.add(BigInteger.ONE)  say k_.toString(int radix)  end `

## NewLISP

`(while (println (++ i)))`

## Nim

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

Using BigInts:

`import bigints var i = 0.initBigIntwhile true:  i += 1  echo i`

## Oberon-2

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

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

## OCaml

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

Oforth handles arbitrary integer precision.

The loop will stop when out of memory

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

## Ol

Ol does not limit the size of numbers. So maximal number depends only on available system memory.

` (let loop ((n 1))   (print n)   (loop (+ 1 n))) `

Sample sequence with break for large numbers:

` (let loop ((n 2))   (print n)   (unless (> n 100000000000000000000000000000000)      (loop (* n n)))) `

Output:

```2
4
16
256
65536
4294967296
18446744073709551616
340282366920938463463374607431768211456
```

## OpenEdge/Progress

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

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

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

## Pascal

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 .. 18446744073709551615begin  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

`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/;

## Phix

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

`integer i = 0while 1 do    ?i    i += 1end 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 = 0while 1 do    ?a    a += 1end while`
Library: Phix/mpfr

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 mpfr.empz b = mpz_init(0)while true do    mpz_add_ui(b,b,1)     mpfr_printf(1,"%Zd\n",b)end while`

## PicoLisp

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

## Piet

 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: 

## Pike

`int i=1;while(true)    write("%d\n", i++);`

## PILOT

`C  :n = 1*InfiniteLoopT  :#nC  :n = n + 1J  :*InfiniteLoop`

## PL/I

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

## Plain English

Numbers are signed 32-bit values, so this will overflow somewhere in the neighborhood of 2.1 billion.

`To run:Start up.Put 1 into a number.Loop.Convert the number to a string.Write the string to the console.Bump the number.Repeat.Shut down.`

## PostScript

Library: initlib
` 1 {succ dup =} loop `

## PowerShell

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

## Prolog

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

### Constraint Handling Rules

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). `

## PureBasic

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

## Python

`i=1while 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.

## Q

Translation of: K

`({-1 string x; x+1}\) 1`

Using while:

`i:0; while[1;-1 string (i+:1)]`

## R

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

## Racket

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)) `

## Raku

(formerly Perl 6)

`.say for 1..*`

## Raven

Raven uses signed 32 bit integer values.

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

## Retro

Retro uses signed integer values.

`#0 [ [ n:put spa ] sip n:inc dup n:-zero? ] while drop`

## REXX

`/*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.        */         endsay 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

` size = 10 for n = 1 to size    see n + nlnextsee nl for n in [1:size]    see n + nlnextsee nl i = nwhile n <= size      see n + nl      n = n + 1end `

## Ruby

`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.

Ruby 2.6 introduced open-ended ranges:

`(1..).each{|n| puts n}`

## Run BASIC

`while 1i = i + 1print iwend`

Eventually as it gets larger it becomes a floating point.

## Rust

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

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 $2^{80}$ digits).

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

or

`for (i; 0; true)    i!;`

or

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

## Scala

`Stream from 1 foreach println`

## Scheme

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

Scheme does not limit the size of numbers.

## Seed7

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

No limit:

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

## Smalltalk

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

will run forever.

## SSEM

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 -= -101000000000000000000000000000000   1. 2 to CI    goto -1 + 111111111111111111111111111111111   2. -1`

## Standard ML

This will print up to Int.maxInt and then raise an Overflow exception. 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;`
Output:
```1
2
3
...
1073741821
1073741822
1073741823

uncaught exception Overflow [overflow]
raised at: <file intSeq.sml>```

## SuperCollider

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 incrementallyj.play; `

## Swift

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

## Symsyn

` | The following code will run forever| Symsyn uses a 64 bit signed integer| The largest positive integer is 9223372036854775807| lpi + 1 = -9223372036854775808 lp x  [] + x go lp `

## Tcl

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

## Tiny BASIC

`     REM will overflow after 32767    LET N = 010  PRINT N    LET N = N + 1    GOTO 10 `

## TUSCRIPT

`\$\$ MODE TUSCRIPTLOOP n=0,999999999n=n+1ENDLOOP`

## UNIX Shell

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

## Ursa

`## integer sequence# # declare an int and loop until it overflowsdecl int iset i 1while true        out i endl console        inc iend while`

## Vala

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

## Visual Basic .NET

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`

### Arbitrarily large numbers

One could use the System.Numerics library as the C# example did, or one can do the following.
A list of Long Integers is maintained as the incremented number. As the incremented value approaches the maximum allowed (base) in the first element of ar, a new item is inserted at the beginning of the list to extend the incremented number. The process has the limitation of when the ar array is enlarged to the point where the program exhausts the available memory, it ought to indicate failure and terminate. It is my understanding that a List count is backed by an Integer.MaxValue limitation and there may also be a 2 GB per object limitation involved. Since writing to the Console is such a slow process, I lack the patience to wait for the program (as written) to fail. If the program is tweaked to fail early, the practical limit seems to be a number 2,415,919,086 digits in length.

`Imports System.Console Module Module1     Dim base, b1 As Long, digits As Integer, sf As String, st As DateTime,        ar As List(Of Long) = {0L}.ToList, c As Integer = ar.Count - 1     Sub Increment(n As Integer)        If ar(n) < b1 Then            ar(n) += 1        Else            ar(n) = 0 : If n > 0 Then                Increment(n - 1)            Else                Try                    ar.Insert(0, 1L) : c += 1                Catch ex As Exception                    WriteLine("Failure when trying to increase beyond {0} digits", CDbl(c) * digits)                    TimeStamp("error")                    Stop                End Try            End If        End If    End Sub     Sub TimeStamp(cause As String)        With DateTime.Now - st            WriteLine("Terminated by {5} at {0} days, {1} hours, {2} minutes, {3}.{4} seconds",                      .Days, .Hours, .Minutes, .Seconds, .Milliseconds, cause)        End With    End Sub     Sub Main(args As String())        digits = Long.MaxValue.ToString.Length - 1        base = CLng(Math.Pow(10, digits)) : b1 = base - 1        base = 10 : b1 = 9        sf = "{" & base.ToString.Replace("1", "0:") & "}"        st = DateTime.Now        While Not KeyAvailable            Increment(c) : Write(ar.First)            For Each item In ar.Skip(1) : Write(sf, item) : Next : WriteLine()        End While        TimeStamp("keypress")    End SubEnd Module`
Output:
```1
2
3
...
10267873
10267874
10267875
Terminated by keypress at 0 days, 0 hours, 30 minutes, 12.980 seconds```

## WDTE

`let s => import 'stream'; s.new 0 (+ 1)-> s.map (io.writeln io.stdout)-> s.drain;`

WDTE's number type is, at the time of writing, backed by Go's `float64` type, so all of the same limitations that apply there apply here. Also, this should not be run in the WDTE playground, as it will run with no output until the browser crashes or is killed.

## Wren

In Wren all numbers are stored in 64-bit floating point form. This means that precise integer calculations are only possible within a maximum absolute magnitude of 2^53-1. There is no 'big integer' support.

However, if the following script were run until the above limit were reached, it would be found that after 99999999999999 the next number would be displayed as 1e14. This is because the System.print method will only display a maximum of 14 digits before switching to scientific notation. As there are no built-in formatting options, there is no way to change this behavior to my knowledge.

One would have to represent numbers as strings and perform arithmetic operations on those strings to display larger numbers in integer form.

`var max = 2.pow(53) // 9007199254740992 (16 digits)for (i in 1...max) System.print(i)}`

## XLISP

`(defun integer-sequence-from (x)	(print x)	(integer-sequence-from (+ x 1)) ) (integer-sequence-from 1)`

## XPL0

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

## zkl

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