Loops/While: Difference between revisions

Content added Content deleted

Inline

Latest revision as of 23:09, 9 March 2024

Task

Start an integer value at 1024.

Loop while it is greater than zero.

Print the value (with a newline) and divide it by two each time through the loop.

Related tasks

0815

<:400:~}:_:%<:a:~$=<:2:=/^:_:

11l

Translation of: Python

V n = 1024
L n > 0
   print(n)
   n I/= 2

6502 Assembly

Code is called as a subroutine (i.e. JSR LoopsWhile). Specific OS/hardware routines for printing are left unimplemented.

LoopsWhile:	PHA			;push accumulator onto stack

		LDA #$00		;the 6502 is an 8-bit processor
		STA Ilow		;and so 1024 ($0400) must be stored in two memory locations
		LDA #$04
		STA Ihigh		
WhileLoop:	LDA Ilow
		BNE NotZero
		LDA Ihigh
		BEQ EndLoop
NotZero:	JSR PrintI		;routine not implemented
		LSR Ihigh		;shift right
		ROR Ilow		;rotate right
		JMP WhileLoop

EndLoop:	PLA			;restore accumulator from stack
		RTS			;return from subroutine

68000 Assembly

Printing routines are left out since this is not the focus of the task.

main:
MOVE.W #1024,D0

WhileLoop:
jsr PrintHexWord
LSR.W #1,D0
BNE WhileLoop

AArch64 Assembly

Works with: as version Raspberry Pi 3B version Buster 64 bits

/* ARM assembly AARCH64 Raspberry PI 3B */
/*  program loopwhile64.s   */
 
/*******************************************/
/* Constantes file                         */
/*******************************************/
/* for this file see task include a file in language AArch64 assembly*/
.include "../includeConstantesARM64.inc"
/*********************************/
/* Initialized data              */
/*********************************/
.data
szMessResult:      .asciz "@"            // message result
szCarriageReturn:  .asciz "\n"
/*********************************/
/* UnInitialized data            */
/*********************************/
.bss 
sZoneConv:           .skip 24
/*********************************/
/*  code section                 */
/*********************************/
.text
.global main 
main:                                    // entry of program 
    mov x20,#1024                        // loop counter
1:                                       // begin loop 
    mov x0,x20
    ldr x1,qAdrsZoneConv                 // display value
    bl conversion10                      // decimal conversion
    ldr x0,qAdrszMessResult
    ldr x1,qAdrsZoneConv
    bl strInsertAtCharInc                // insert result at @ character
    bl affichageMess                     // display message
    ldr x0,qAdrszCarriageReturn
    bl affichageMess                     // display return line
    lsr x20,x20,1                        // division by 2
    cmp x20,0                            // end ?
    bgt 1b                               // no ->begin loop one
 
 
100:                                     // standard end of the program 
    mov x0,0                             // return code
    mov x8,EXIT                          // request to exit program
    svc 0                                // perform the system call
 
qAdrsZoneConv:            .quad sZoneConv
qAdrszMessResult:         .quad szMessResult
qAdrszCarriageReturn:     .quad szCarriageReturn

/********************************************************/
/*        File Include fonctions                        */
/********************************************************/
/* for this file see task include a file in language AArch64 assembly */
.include "../includeARM64.inc"

Output:

Action!

PROC Main()
  CARD i=[1024]

  WHILE i>0
  DO
    PrintCE(i)
    i=i/2
  OD
RETURN

Output:

Screenshot from Atari 8-bit computer

ActionScript

var i:int = 1024;
while (i > 0) {
    trace(i);
    i /= 2;
}

Ada

declare
   I : Integer := 1024;
begin
   while I > 0 loop
      Put_Line(Integer'Image(I));
      I := I / 2;
   end loop;
end;

Agena

Tested with Agena 2.9.5 Win32

scope
    local i := 1024;
    while i > 0 do
        print( i );
        i := i \ 2
    od
epocs

Aime

integer i;

i = 1024;
while (i) {
    o_plan(i, "\n");
    i /= 2;
}

ALGOL 60

The Loops/While structure was in the Algol 60 report of January 1963.

begin
	comment Loops/While - algol60 - 21/10/2014;
	integer i;
	for i:=1024,i div 2 while i>0 do outinteger(1,i)
end

Output:

1024  512  256  128  64  32  16  8  4  2  1

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

Works with: ELLA ALGOL 68 version Any (with appropriate job cards) - tested with release 1.8-8d

INT i := 1024;
WHILE i > 0 DO
   print(i);
   i := i OVER 2
OD

Output:

      +1024       +512       +256       +128        +64        +32        +16         +8         +4         +2         +1

ALGOL W

begin
    integer i;
    i := 1024;
    while i > 0 do
    begin
        write( i );
        i := i div 2
    end
end.

ALGOL-M

begin
    integer i;
    i := 1024;
    while i > 0 do begin
        write( i );
        i := i / 2;
    end;
end

Amazing Hopper

#include <jambo.h>

Main
   i=1024
   Loop
       Printnl 'i'
       i \= 2
   Back if 'i' is positive
End

Assembler Hopper code:

main:
i=1024
____CODE_JUMP____883612951:,
{i};{"\n"}print;
i\=2
{i},jpos(____CODE_JUMP____883612951),____CODE_JUMP____854411479:,
emptystack?do{{0}};return

Output:

AmbientTalk

Note: in AmbientTalk, while:do: is a keyworded message (as in Smalltalk). Both arguments to this message must be blocks (aka anonymous functions or thunks).

// print 1024 512 etc
def i := 1024;
while: { i > 0 } do: {
  system.print(" "+i);
  i := i/2;
}

AmigaE

PROC main()
  DEF i = 1024
  WHILE i > 0
    WriteF('\d\n', i)
    i := i / 2
  ENDWHILE
ENDPROC

AppleScript

AppleScript does not natively support a standard out. Use the Script Editor's Event Log as the output.

set i to 1024
repeat while i > 0
	log i
	set i to i / 2
end repeat

ARM Assembly

Works with: as version Raspberry Pi

/* ARM assembly Raspberry PI  */
/*  program loopwhile.s   */

/* Constantes    */
.equ STDOUT, 1     @ Linux output console
.equ EXIT,   1     @ Linux syscall
.equ WRITE,  4     @ Linux syscall

/*********************************/
/* Initialized data              */
/*********************************/
.data
szMessResult:      .ascii ""                    @ message result
sMessValeur:       .fill 11, 1, ' '
szCarriageReturn:  .asciz "\n"
/*********************************/
/* UnInitialized data            */
/*********************************/
.bss 
/*********************************/
/*  code section                 */
/*********************************/
.text
.global main 
main:                                       @ entry of program 
    mov r4,#1024                            @ loop counter
1:                                          @ begin loop 
    mov r0,r4
    ldr r1,iAdrsMessValeur                  @ display value
    bl conversion10                         @ decimal conversion
    ldr r0,iAdrszMessResult
    bl affichageMess                        @ display message
    ldr r0,iAdrszCarriageReturn
    bl affichageMess                        @ display return line
    lsr r4,#1                               @ division by 2
    cmp r4,#0                               @ end ?
    bgt 1b                                  @ no ->begin loop one


100:                                        @ standard end of the program 
    mov r0, #0                              @ return code
    mov r7, #EXIT                           @ request to exit program
    svc #0                                  @ perform the system call

iAdrsMessValeur:          .int sMessValeur
iAdrszMessResult:         .int szMessResult
iAdrszCarriageReturn:     .int szCarriageReturn
/******************************************************************/
/*     display text with size calculation                         */ 
/******************************************************************/
/* r0 contains the address of the message */
affichageMess:
    push {r0,r1,r2,r7,lr}                   @ save  registres
    mov r2,#0                               @ counter length 
1:                                          @ loop length calculation 
    ldrb r1,[r0,r2]                         @ read octet start position + index 
    cmp r1,#0                               @ if 0 its over 
    addne r2,r2,#1                          @ else add 1 in the length 
    bne 1b                                  @ and loop 
                                            @ so here r2 contains the length of the message 
    mov r1,r0                               @ address message in r1 
    mov r0,#STDOUT                          @ code to write to the standard output Linux 
    mov r7, #WRITE                          @ code call system "write" 
    svc #0                                  @ call systeme 
    pop {r0,r1,r2,r7,lr}                    @ restaur registers */ 
    bx lr                                   @ return  
/******************************************************************/
/*     Converting a register to a decimal                                 */ 
/******************************************************************/
/* r0 contains value and r1 address area   */
.equ LGZONECAL,   10
conversion10:
    push {r1-r4,lr}                         @ save registers 
    mov r3,r1
    mov r2,#LGZONECAL
1:                                          @ start loop
    bl divisionpar10                        @ r0 <- dividende. quotient ->r0 reste -> r1
    add r1,#48                              @ digit
    strb r1,[r3,r2]                         @ store digit on area
    cmp r0,#0                               @ stop if quotient = 0 
    subne r2,#1                               @ previous position    
    bne 1b                                  @ else loop
                                            @ end replaces digit in front of area
    mov r4,#0
2:
    ldrb r1,[r3,r2] 
    strb r1,[r3,r4]                         @ store in area begin
    add r4,#1
    add r2,#1                               @ previous position
    cmp r2,#LGZONECAL                       @ end
    ble 2b                                  @ loop
    mov r1,#0                               @ final zero 
    strb r1,[r3,r4]
100:
    pop {r1-r4,lr}                          @ restaur registres 
    bx lr                                   @return
/***************************************************/
/*   division par 10   signé                       */
/* Thanks to http://thinkingeek.com/arm-assembler-raspberry-pi/*  
/* and   http://www.hackersdelight.org/            */
/***************************************************/
/* r0 dividende   */
/* r0 quotient */	
/* r1 remainder  */
divisionpar10:	
  /* r0 contains the argument to be divided by 10 */
    push {r2-r4}                           @ save registers  */
    mov r4,r0  
    mov r3,#0x6667                         @ r3 <- magic_number  lower
    movt r3,#0x6666                        @ r3 <- magic_number  upper
    smull r1, r2, r3, r0                   @ r1 <- Lower32Bits(r1*r0). r2 <- Upper32Bits(r1*r0) 
    mov r2, r2, ASR #2                     @ r2 <- r2 >> 2
    mov r1, r0, LSR #31                    @ r1 <- r0 >> 31
    add r0, r2, r1                         @ r0 <- r2 + r1 
    add r2,r0,r0, lsl #2                   @ r2 <- r0 * 5 
    sub r1,r4,r2, lsl #1                   @ r1 <- r4 - (r2 * 2)  = r4 - (r0 * 10)
    pop {r2-r4}
    bx lr                                  @ return

ArnoldC

IT'S SHOWTIME
HEY CHRISTMAS TREE n
YOU SET US UP 1024
STICK AROUND n
TALK TO THE HAND n
GET TO THE CHOPPER n
HERE IS MY INVITATION n
HE HAD TO SPLIT 2
ENOUGH TALK
CHILL
YOU HAVE BEEN TERMINATED

Arturo

i: 1024
 
while [i>0] [
	print i
	i: i/2
]

Output:

Asymptote

int i = 1024;

while(i > 0) {
  write(i);
  i = i # 2;       //or also i = quotient(i, 2);
}

//# Integer division; equivalent to quotient(x,y).
//Noting that the Python3 community adopted the comment symbol (//) for integer division, the
//Asymptote community decided to reciprocate and use their comment symbol for integer division!

ATS

With compile-time proof of termination of the loop (which is really a tail recursion).

#include "share/atspre_staload.hats"

fn
loop_while () : void =
  let
    fun
    loop {n : int | 0 <= n} .<n>.
         (n : uint n) : void =
      if n <> 0U then
        begin
          println! (n);
          loop (n / 2U)
        end
  in
    loop 1024U
  end

implement
main0 () =
  loop_while ()

Output:

It is also possible to write an actual ‘while’ loop, although this method requires a variable (the ‘var’ declaration below) and has other problems, and should be used sparingly.

#include "share/atspre_staload.hats"

fn
loop_while () : void =
  let
    var n : uint = 1024U
  in
    while (0U < n)
      begin
        println! (n);
        n := n / 2U
      end
  end

implement
main0 () =
  loop_while ()

AutoHotkey

i = 1024
While (i > 0)
{
  output = %output%`n%i%
  i := Floor(i / 2)
}
MsgBox % output

AWK

BEGIN {
  v = 1024
  while(v > 0) {
    print v
    v = int(v/2)
  }
}

Axe

1024→A
While A>0
 Disp A▶Dec,i
 A/2→A
End

Bait

fun main() {
	mut i := 1024
	for i > 0 {
		println(i)
		i = i / 2
	}
}

BASIC

In general, the WHILE-WEND (or DO WHILE-LOOP) statement is used or it is simulated with a construct with conditional jump.

ANSI BASIC

Works with: Decimal BASIC

100 LET I = 1024
110 DO WHILE I > 0
120    PRINT I
130    LET I = INT(I / 2)
140 LOOP
150 END

Output:

Applesoft BASIC

 10 I% = 1024
 20  IF I% > 0 THEN  PRINT I%:I% = I% / 2: GOTO 20

ASIC

REM Loops/While

I = 1024
WHILE I > 0
  PRINT I
  I = I / 2
WEND

END

Output:

BaCon

i = 1024
WHILE i > 0
   PRINT i
   i = i / 2
WEND

Ballerina

int i = 1024;
while i > 0 {
    io:println(i);
    i = i / 2;
}

BASIC256

i = 1024

while i > 0
	print i
	i = i \ 2
end while

end

BBC BASIC

Works with: BBC BASIC for Windows

      i% = 1024
      WHILE i%
        PRINT i%
        i% DIV= 2
      ENDWHILE

Chipmunk Basic

Works with: Chipmunk Basic version 3.6.4

Both loops are equivalent

100 i = 1024
110 do while i > 0
120   print i
130   i = int(i/2)
140 loop
150 print
160 i = 1024
170 while i > 0
180   print i
190   i = int(i/2)
200 wend

Commodore BASIC

There is no WHILE construct in Commodore BASIC. A GOTO construct is used instead. Also, an integer variable name has a % sign as its suffix.

10 N% = 1024
20 IF N% = 0 THEN 60
30 PRINT N%
40 N% = N%/2
50 GOTO 20
60 END

Output:

Creative Basic

DEF X:INT

X=1024

OPENCONSOLE

WHILE X>0

   PRINT X
   X=X/2
  
ENDWHILE
'Output starts with 1024 and ends with 1.

'Putting the following in the loop will produce output starting with 512 and ending with 0:
'X=X/2
'PRINT X

PRINT:PRINT"Press any key to end."

'Keep console from closing right away so the figures can be read.
WHILE INKEY$="":ENDWHILE

CLOSECONSOLE

'Since this is, in fact, a Creative Basic console program.
END

Note: Spacing is not an issue. I just find the code to be more readable with spaces.

FreeBASIC

' FB 1.05.0 Win64

Dim i As Integer = 1024

While i > 0
  Print i
  i Shr= 1
Wend

Sleep

Output:

FutureBasic

window 1

long i = 1024

while i > 0
  print i
  i = int( i / 2 )
wend

HandleEvents

Output:

Gambas

Click this link to run this code

Public Sub Main()
Dim siCount As Short = 1024

While siCount > 0 
  Print siCount;;
  siCount /= 2
Wend

End

Output:

1024 512 256 128 64 32 16 8 4 2 1

GW-BASIC

Works with: BASICA

Works with: Chipmunk Basic

Works with: PC-BASIC version any

Works with: QBasic

10 I = 1024
20 WHILE I > 0
30   PRINT I
40   i = INT(i/2)
50 WEND
60 END

IS-BASIC

100 LET I=1024
110 DO WHILE I>0
120   PRINT I
130   LET I=IP(I/2)
140 LOOP

IWBASIC

DEF X:INT

X=1024

OPENCONSOLE

WHILE X>0

    PRINT X
    X=X/2 

ENDWHILE
'Output starts with 1024 and ends with 1.

'Putting the following in the loop will produce output starting with 512 and ending with 0:
'X=X/2
'PRINT X

'When compiled as a console only program, a press any key to continue message is automatic.
'I presume code is added by the compiler.
CLOSECONSOLE

'Since this is, in fact, an IWBASIC console program, which compiles and runs.
END

Note: Spacing is not an issue. I just find the code to be more readable with spaces.

Liberty BASIC

All integers are changed to floats if an operation creates a non-integer result. Without using int() the program keeps going until erroring because accuracy was lost.

Works with: Just BASIC

i = 1024
while i > 0
   print i
   i = int( i / 2)
wend
end

Microsoft Small Basic

i = 1024
While i > 0
  TextWindow.WriteLine(i)
  i = Math.Floor(i / 2)
EndWhile

Minimal BASIC

Minimal BASIC have no while construct. Equivalent using conditional jump:

Works with: Commodore BASIC

Works with: Nascom ROM BASIC version 4.7

10 REM Loops/While
20 LET I = 1024
40 IF I <= 0 THEN 80
50 PRINT I
60 LET I = INT(I/2)
70 GOTO 40
80 END

MSX Basic

There is no WHILE construct in MSX Basic. A GOTO construct is used instead.

10 I% = 1024
20 IF I% = 0 THEN END
30 PRINT I%
40 I% = I%/2 : rem INT(I/2)
50 GOTO 20

Solutions GW-BASIC and Minimal BASIC work without changes.

Nascom BASIC

See Minimal BASIC.

If the repeated sequence is short then the loop can be written in one line.

Works with: Nascom ROM BASIC version 4.7

10 REM Loops/While
20 LET I=1024
30 IF I>0 THEN PRINT I:I=INT(I/2):GOTO 30
40 END

NS-HUBASIC

10 I=1024
20 IF I=0 THEN END
30 PRINT I
40 I=I/2
50 GOTO 20

PureBasic

If OpenConsole()  
  
  x.i = 1024
  While x > 0
    PrintN(Str(x))
    x / 2
  Wend

  Print(#CRLF$ + #CRLF$ + "Press ENTER to exit")
  Input()
  CloseConsole()
EndIf

QB64

Dim n As Integer
n = 1024
While n > 0
    Print n
    n = n \ 2
Wend

QBasic

Works with: QBasic version 1.1

Both loops are equivalent

i = 1024
WHILE i > 0
   PRINT i
   i = INT(i / 2)
WEND
PRINT

i = 1024
DO WHILE i > 0
   PRINT i
   i = i \ 2
LOOP
END

QuickBASIC

Works with: QuickBasic version 4.5

i = 1024
WHILE i > 0
  PRINT i
  i = i \ 2
WEND

Run BASIC

i = 1024
while i > 0
   print i
   i = int(i / 2)
wend
end

Sinclair ZX81 BASIC

The distinctive thing about a while loop is that the conditional test happens before the loop body, not after—so that the code in the loop may be executed zero times.

Since we have no integer type, we floor the result of the division each time.

10 LET I=1024
20 IF I=0 THEN GOTO 60
30 PRINT I
40 LET I=INT (I/2)
50 GOTO 20

SmallBASIC

i = 1024
WHILE i > 0
   PRINT i
   i = i \ 2  ' Using \ for integer division instead of /
WEND

TI-83 BASIC

1024→I
While I>0
Disp I
I/2→I
End

TI-89 BASIC

Local i
1024 → i
While i > 0
  Disp i
  intDiv(i, 2) → i
EndWhile

Tiny BASIC

Tiny BASIC have no while construct. Equivalent using conditional jump:

Works with: TinyBasic

10 REM Loops/While
20 LET I = 1024
30 IF I <= 0 THEN GOTO 70
40 PRINT I
50 LET I = I / 2
60 GOTO 30
70 END

Output:

True BASIC

LET i = 1024

DO WHILE i > 0
   PRINT i
   LET i = INT(i / 2)
LOOP
END

VBA

Public Sub LoopsWhile()
    Dim value As Integer
    value = 1024
    Do While value > 0
        Debug.Print value
        value = value / 2
    Loop
End Sub

Visual Basic .NET

Dim x = 1024
Do
    Console.WriteLine(x)
    x = x \ 2
Loop While x > 0

Wee Basic

let number=1024
while number>0.5
print 1 number
let number=number/2
wend
end

XBasic

Works with: Windows XBasic

i% = 1024
DO WHILE i% > 0
  PRINT i%
  i% = i% / 2
LOOP

Yabasic

i = 1024 
while i > 0
  Print i
  i = int(i / 2)
wend
end

bc

i = 1024
while (i > 0) {
    i
    i /= 2
}

Befunge

84*:*>       :v
     ^/2,*25.:_@

blz

num = 1024
while num > 1 # blz will automatically cast num to a fraction when dividing 1/2, so this is necessary to stop an infinite loop
    print(num)
    num = num / 2
end

BQN

BQNcrate's while idiom is the closest equivalent of a while loop in the language.

_while_ ← {𝔽⍟𝔾∘𝔽_𝕣_𝔾∘𝔽⍟𝔾𝕩}

(⌊∘÷⟜2 •Show) _while_ (>⟜0) 1024

Bracmat

1024:?n & whl'(!n:>0 & out$!n & div$(!n.2):?n)

Brat

Converts to integers so output is a little bit shorter and neater.

i = 1024
while { i > 0 } {
    p i
    i = (i / 2).to_i
}

C

int i = 1024;
while(i > 0) {
  printf("%d\n", i);
  i /= 2;
}

In for loop fashion:

int i;
for(i = 1024;i > 0; i/=2){
   printf("%d\n", i);
}

C#

int i = 1024;
while(i > 0){
   System.Console.WriteLine(i);
   i /= 2;
}

C++

int i = 1024;
while(i > 0){
  std::cout << i << std::endl;
  i /= 2;
}

Alternatively, it can be done with for:

for(int i = 1024; i > 0; i /= 2)
  std::cout << i << std::endl;

Instead of i /= 2 one can also use the bit shift operator i >>= 1 on integer variables.

Indeed, in C++,

for(init; cond; update){
  statement;
}

is equivalent to

{
  init;
  while(cond){
    statement;
    update;
  }
}

Caché ObjectScript

WHILELOOP
    set x = 1024
    while (x > 0) {
        write x,!
        set x = (x \ 2)    ; using non-integer division will never get to 0
    }
	
    quit

Output:

SAMPLES>DO ^WHILELOOP
1024
512
256
128
64
32
16
8
4
2
1

Chapel

var val = 1024;
while val > 0 {
        writeln(val);
        val /= 2;
}

ChucK

1024 => int value;

while(value > 0)
{
    <<<value>>>;
    value / 2 => value;
}

Clojure

(def i (ref 1024))

(while (> @i 0)
  (println @i)
  (dosync (ref-set i (quot @i 2))))

2 ways without mutability:

(loop [i 1024]
  (when (pos? i)
    (println i)
    (recur (quot i 2))))


(doseq [i (take-while pos? (iterate #(quot % 2) 1024))]
  (println i))

CLU

start_up = proc ()
    po: stream := stream$primary_output()
    n: int := 1024
    while n>0 do
        stream$putl(po, int$unparse(n))
        n := n/2
    end
end start_up

COBOL

COBOL does not have a while loop construct, but it is does have a PERFORM UNTIL structure, which means that the normal condition used in a while loop must be negated.

       IDENTIFICATION DIVISION.
       PROGRAM-ID. Loop-While.

       DATA DIVISION.
       WORKING-STORAGE SECTION.
       01  I PIC 9999 VALUE 1024.

       PROCEDURE DIVISION.
           PERFORM UNTIL NOT 0 < I
               DISPLAY I
               DIVIDE 2 INTO I
           END-PERFORM

           GOBACK
           .

ColdFusion

Remove the leading space from the line break tag.

With tags:

<cfset i = 1024 /><cfloop condition="i GT 0">  #i#< br />
  <cfset i /= 2 />
</cfloop>

With script:

<cfscript>  i = 1024;
  while( i > 0 )
  {
    writeOutput( i + "< br/ >" );
  }
</cfscript>

Common Lisp

(let ((i 1024))
  (loop while (plusp i) do
        (print i)
        (setf i (floor i 2))))

(loop with i = 1024
      while (plusp i) do
      (print i)
      (setf i (floor i 2)))

(defparameter *i* 1024)
(loop while (plusp *i*) do
      (print *i*)
      (setf *i* (floor *i* 2)))

Cowgol

include "cowgol.coh";

var n: uint16 := 1024;
while n > 0 loop
    print_i16(n);
    print_nl();
    n := n/2;
end loop;

Crack

i = 1024;
while( i > 0 ) {
  cout ` $i\n`;
  i = i/2;
}

Crystal

i = 1024
while i > 0
    puts i
    i //= 2
end

until condition is the negated version, equivalent to while !(condition).

i = 1024
until i <= 0
   puts i
   i //= 2
end

D

import std.stdio;

void main() {
    int i = 1024;

    while (i > 0) {
        writeln(i);
        i >>= 1;
    }
}

Output:

Dao

i = 1024;
while( i > 0 ) i = i / 2;

Dart

void main() {
  var val = 1024;
  while (val > 0) {
    print(val);
    val >>= 2;
  }
}

Instead of the bitshift operator i >>= 2, you can also use i /= 2 on double variables.

void main() {
  num val = 1024;
  while (val > 0) {
    print(val);
    val /= 2;
  }
}

Dc

[ q ] sQ [ d 0!<Q p 2 / lW x ] sW 1024 lW x

DCL

DCL is quite primitive in terms of "control statements", no WHILE, REPEAT, UNLESS or FOR, so must make do with IF/THEN/ELSE and GOTO statements.

$ i = 1024
$Loop:
$ IF ( i .LE. 0 ) THEN GOTO LoopEnd
$ WRITE sys$output F$FAO( "  i = !4UL", i )  ! formatted ASCII output, fixed-width field
$ ! Output alternatives:
$ !   WRITE sys$output F$STRING( i )         ! explicit integer-to-string conversion
$ !   WRITE sys$output i                     ! implicit conversion to string/output
$ i = i / 2
$ GOTO Loop
$LoopEnd:

Delphi

var
  i : Integer;
begin
  i := 1024;

  while i > 0 do
  begin
    Writeln(i);
    i := i div 2;
  end;
end;

Draco

proc nonrec main() void:
    word i;
    i := 1024;
    while i > 0 do
        writeln(i);
        i := i >> 1
    od
corp

Output:

Dragon

i = 1024
while(i > 0){
   showln i
   i >>= 1 //also acceptable: i /= 2
}

DUP

1024[$][$.10,2/\%]# {Short form}

Explanation:

1024                {push 1024 on stack}
    [ ][         ]# {while[condition>0][do]}
     $              {DUP}
        $.          {DUP, print top of stack to STDOUT}
          10,       {print newline}
             2/\%   {2 DIV/MOD SWAP POP}

Alternative, if the interpreter allows using the shift operator:

1024[$][$.10,1»]#

Output:

DWScript

var i := 1024;

while i > 0 do begin
   PrintLn(i);
   i := i div 2;
end;

Dyalect

Translation of: Swift

var i = 1024
while i > 0 {
  print(i)
  i /= 2
}

E

var i := 1024
while (i > 0) {
    println(i)
    i //= 2
}

EasyLang

i = 1024
while i > 0
  print i
  i = i div 2
.

EchoLisp

(set! n 1024)
(while (> n 0) (write n) (set! n (quotient n 2)))
1024 512 256 128 64 32 16 8 4 2 1

EGL

x int = 1024;
while ( x > 0 )
   SysLib.writeStdout( x );
   x = MathLib.floor( x / 2 );
end

Elena

ELENA 6.x:

public program()
{
    int i := 1024;
    while (i > 0)
    {
        console.writeLine(i);
 
        i /= 2
    }
}

Elixir

defmodule Loops do
  def while(0), do: :ok
  def while(n) do
    IO.puts n
    while( div(n,2) )
  end
end

Loops.while(1024)

Emacs Lisp

(let ((i 1024))
  (while (> i 0)
    (message "%d" i)
    (setq i (/ i 2))))

EMal

int i = 1024
while i > 0
  writeLine(i)
  i /= 2
end

Erlang

-module(while).
-export([loop/0]).

loop() ->
	loop(1024).
 
loop(N) when N div 2 =:= 0 ->
	io:format("~w~n", [N]);
	
loop(N) when N >0 ->
	io:format("~w~n", [N]),
	loop(N div 2).

ERRE

   I%=1024
   WHILE I%>0 DO  ! you can leave out >0
     PRINT(I%)
     I%=I% DIV 2  ! I%=INT(I%/2) for C-64 version
   END WHILE

Euphoria

integer i
i = 1024

while i > 0 do
    printf(1, "%g\n", {i})
    i = floor(i/2) --Euphoria does NOT use integer division.  1/2 = 0.5
end while

Even without the floor() the code will in fact end. But it's FAR beyond 1.

F#

let rec loop n = if n > 0 then printf "%d " n; loop (n / 2)
loop 1024

Factor

1024 [ dup 0 > ] [ dup . 2 /i ] while drop

FALSE

1024[$0>][$."
"2/]#%

Fantom

class Main
{
  public static Void main ()
  {
    Int i := 1024
    while (i > 0)
    {
      echo (i)
      i /= 2
    }
  }
}

Fennel

(var n 1024)
(while (> i 0)
  (print i)
  (set i (// n 2)))

Fermat

Note that the test is against something between 1/2 and 1, rather than 0. Otherwise Fermat will continue printing numbers of the form 1/2^n indefinitely.

n:=1024;
while n>2/3 do !!n;n:=n/2; od;

Forth

: halving ( n -- )
  begin  dup 0 >
  while  cr dup .  2/
  repeat drop ;
1024 halving

Fortran

Works with: Fortran version 90 and later

INTEGER :: i = 1024
DO WHILE (i > 0)
  WRITE(*,*) i
  i = i / 2
END DO

Works with: Fortran version 77 and later

      PROGRAM LOOPWHILE
        INTEGER I

C       FORTRAN 77 does not have a while loop, so we use GOTO statements
C       with conditions instead. This is one of two easy ways to do it.
        I = 1024
   10   CONTINUE
C       Check condition.
        IF (I .GT. 0) THEN
C         Handle I.
          WRITE (*,*) I
          I = I / 2
C         Jump back to before the IF block.
          GOTO 10
        ENDIF
        STOP
      END

Works with: Fortran version IV and 66 and later

      PROGRAM LOOPWHILE
      INTEGER I
C     FORTRAN 66 does not have IF block.
      I = 1024
 10   CONTINUE
      IF (I .LE. 0) GOTO 20
      WRITE (*,*) I
      I = I / 2
      GOTO 10
 20   CONTINUE
      STOP
      END

Fortress

component loops_while
  export Executable

  var i:ZZ32 = 1024
  run() = while i > 0 do
    println(i)
    i := i DIV 2
  end
end

Output:

Frink

i=1024
while i>0
{
   i = i/1
}

GAP

n := 1024;
while n > 0 do
    Print(n, "\n");
    n := QuoInt(n, 2);
od;

GML

i = 1024
while(i > 0)
    {
    show_message(string(i))
    i /= 2
    }

Go

i := 1024
for i > 0 {
  fmt.Printf("%d\n", i)
  i /= 2
}

Groovy

Solution:

int i = 1024
while (i > 0) {
    println i
    i /= 2
}

Output:

Haskell

import Control.Monad (when)

main = loop 1024      
  where loop n = when (n > 0)
                      (do print n
                          loop (n `div` 2))

You can use whileM_ function from monad-loops package that operates on monads:

import Data.IORef
import Control.Monad.Loops

main :: IO ()
main = do r <- newIORef 1024
          whileM_ (do n <- readIORef r
                     return (n > 0))
                  (do n <- readIORef r
                     print n
                     modifyIORef r (`div` 2))

With MonadComprehensions extension you can write it a little bit more readable:

{-# LANGUAGE MonadComprehensions #-}
import Data.IORef 
import Control.Monad.Loops

main :: IO ()
main = do
   r <- newIORef 1024 
   whileM_ [n > 0 | n <- readIORef r] $ do
        n <- readIORef r  
        print n
        modifyIORef r (`div` 2)

Haxe

Using shift right.

var i = 1024;
    
while (i > 0) {
  Sys.println(i);
  i >>= 1;
}

Using integer division.

var i = 1024;
    
while (i > 0) {
  Sys.println(i);
  i = Std.int(i / 2);
}

Output:

hexiscript

let i 1024
while i > 0
  println i
  let i (i / 2)
endwhile

HolyC

U16 i = 1024;
while (i > 0) {
  Print("%d\n", i);
  i /= 2;
}

Icon and Unicon

procedure main()
   local i
   i := 1024
   while write(0 < (i := i / 2))
end

Inform 7

let N be 1024;
while N > 0:
	say "[N][line break]";
	let N be N / 2;

Insitux

Use / for floating-point division, and use (> i 0) for full printing of the sequence (as the output of 1 here is due to the return of the last while statement).

(var i 1024)

(while (> i 1)
  (print i)
  (var i (// i 2)))

Output:

J

J is array-oriented, so there is very little need for loops. For example, one could satisfy this task this way:

,. <.@-:^:*^:a: 1024

J does support loops for those times they can't be avoided (just like many languages support gotos for those time they can't be avoided).

monad define 1024
  while. 0 < y do.
    smoutput y
    y =. <. -: y 
  end.
  i.0 0
)

Note: this defines an anonymous function (monad define, and the subsequent lines) and passes it the argument 1024, which means it will be executed as soon as the full definition is available.

Java

int i = 1024;
while(i > 0){
   System.out.println(i);
   i >>= 1; //also acceptable: i /= 2;
}

With a for loop:

for(int i = 1024; i > 0;i /= 2 /*or i>>= 1*/){
   System.out.println(i);
}

JavaScript

var n = 1024;
while (n > 0) {
  print(n);
  n /= 2;
}

In a functional idiom of JavaScript, however, we can not use a While statement to achieve this task, as statements return no value, mutate state, and can not be composed within other functional expressions.

Instead, we can define a composable loopWhile() function which has no side effects, and takes 3 arguments:

An initial value
A function which returns some derived value, corresponding to the body of the While loop
A conditional function, corresponding to the While test

function loopWhile(varValue, fnDelta, fnTest) {
  'use strict';
  var d = fnDelta(varValue);

  return fnTest(d) ? [d].concat(
    loopWhile(d, fnDelta, fnTest)
  ) : [];
}

console.log(
  loopWhile(
    1024,
    function (x) {
      return Math.floor(x/2);
    },
    function (x) {
      return x > 0;
    }
  ).join('\n')
);

If we assume integer division here (Math.floor(x/2)) rather than the floating point division (x/2) used in the imperative example, we obtain the output:

Joy

DEFINE putln == put '\n putch.
1024 [] [dup putln 2 /] while.

jq

Using recurse/1

# To avoid printing 0, test if the input is greater than 1
1024 | recurse( if . > 1 then ./2 | floor else empty end)

Using recurse/2 (requires jq >1.4)

1024 | recurse( ./2 | floor; . > 0)

Using a filter

def task: if . > 0 then ., (./2 | floor | task) else empty end;
1024|task

Using while/2

If your jq does not include while/2 as a builtin, here is its definition:

def while(cond; update):
  def _while: if cond then ., (update | _while) else empty end;
  _while;

For example:

1024|while(. > 0; ./2|floor)

Jsish

#!/usr/bin/env jsish
/* Loops/While in Jsish */
var i = 1024;

while (i > 0) { puts(i); i = i / 2 | 0; }

/*
=!EXPECTSTART!=
1024
512
256
128
64
32
16
8
4
2
1
=!EXPECTEND!=
*/

Output:

prompt$ jsish -u loopsWhile.jsi
[PASS] loopsWhile.jsi

Julia

n = 1024

while n > 0
    println(n)
    n >>= 1
end

Output:

K

Implementation of the task using anonymous function is given below

{while[x>0; \echo x; x%:2]} 1024

Kotlin

// version 1.0.6

fun main(args: Array<String>) {
    var value = 1024
    while (value > 0) {
        println(value)
        value /= 2
    }
}

Output:

LabVIEW

Use Round Towards -Inf to prevent the integer becoming a float.
This image is a VI Snippet, an executable image of LabVIEW code. The LabVIEW version is shown on the top-right hand corner. You can download it, then drag-and-drop it onto the LabVIEW block diagram from a file browser, and it will appear as runnable, editable code.

Lambdatalk

{def loops_while
 {lambda {:i}
  {if {< :i 1}
   then (end of loop)
   else :i {loops_while {/ :i 2}}}}}
-> loops_while

{loops_while 1024}
-> 1024 512 256 128 64 32 16 8 4 2 1 (end of loop)

Alternative solution:

{def while
 {lambda {:i}
  {if {< :i 1}
   then
   else {br}:i {while {/ :i 2}}}}}

{while 1024} -> 

1024 
512 
256 
128 
64 
32 
16 
8 
4 
2 
1

Lang

$n = 1024
while($n > 0) {
	fn.println($n)
	
	$n //= 2
}

Lang5

Translation of: Factor

: /i  / int ; : 0=  0 == ;
: dip  swap '_ set execute _ ; : dupd  'dup dip ;
: 2dip  swap '_x set swap '_y set execute _y _x ;
: while
    do  dupd 'execute 2dip
        rot 0= if break else dup 2dip then
    loop ;

1024 "dup 0 >" "dup . 2 /i" while

langur

var .i = 1024
while .i > 0 {
    writeln .i
    .i \= 2
}

Lasso

local(i = 1024)
while(#i > 0) => {^
	#i + '\r'
	#i /= 2
^}

LDPL

data:
n is number

procedure:
store 1024 in n
while n is greater than 0 do
    display n lf
    divide n by 2 in n
    floor n
repeat

LIL

set num 1024; while {$num > 0} {print $num; set num [expr $num \ 2]}

Backslash is integer division, otherwise LIL would allow the division to go floating point.

Lingo

n = 1024
repeat while n>0
  put n
  n = n/2 -- integer division implicitely returns floor: 1/2 -> 0
end repeat

Lisaac

+ i : INTEGER;
i := 1024;
{ i > 0 }.while_do {
  i.println;
  
  i := i / 2;
};

LiveCode

put 1024 into n
repeat while n > 0
    put n & cr
    divide n by 2
end repeat

Logo

make "n 1024
while [:n > 0] [print :n  make "n :n / 2]

LOLCODE

LOLCODE's loop semantics require an afterthought if a condition is used, thus the nop in the following example. The more idiomatic approach would have been to GTFO of the loop once n had reached 0.

HAI 1.3

I HAS A n ITZ 1024

IM IN YR loop UPPIN YR nop WILE n
    VISIBLE n
    n R QUOSHUNT OF n AN 2
IM OUTTA YR loop

KTHXBYE

Lua

n = 1024
while n>0 do
  print(n)
  n = math.floor(n/2)
end

M2000 Interpreter

Module Checkit {
      Def long A=1024
      While A>0 {
            Print A
            A/=2
      }
}
Checkit

One line

Module Online { A=1024&: While A>0 {Print A: A/=2}} : OnLine

m4

This ‘loop’ is really a tail recursion, which m4 implementations generally do not optimize.

divert(-1)

define(`loop',
  `ifelse(eval(0 < ($1)),1,`$1`
'loop(eval(($1) / 2))')')

divert`'dnl
loop(1024)dnl

Output:

$ m4 loops_while.m4

Make

NEXT=`expr $* / 2`
MAX=10

all: $(MAX)-n;

0-n:;

%-n: %-echo
       @-make -f while.mk $(NEXT)-n MAX=$(MAX)

%-echo:
       @echo $*

Invoking it

|make -f while.mk MAX=1024

Maple

To avoid generating an infinite sequence (1/2, 1/4, 1/8, 1/16, etc.) of fractions after n takes the value 1, we use integer division (iquo) rather than the solidus operation (/).

> n := 1024: while n > 0 do print(n); n := iquo(n,2) end:
                                  1024
                                  512
                                  256
                                  128
                                   64
                                   32
                                   16
                                   8
                                   4
                                   2
                                   1

Mathematica/Wolfram Language

Mathematica does not support integer-rounding, it would result in getting fractions: 1/2, 1/4 , 1/8 and so on; the loop would take infinite time without using the Floor function:

i = 1024;
While[i > 0,
 Print[i];
 i = Floor[i/2];
]

MATLAB / Octave

In Matlab (like Octave) the math is done floating point, then rounding to integer, so that 1/2 will be always 1 and never 0. A 'floor' is used to round the number.

i = 1024;
while (i > 0)
    disp(i);
    i = floor(i/2);
end

A vectorized version of the code is

  printf('%d\n', 2.^[log2(1024):-1:0]);

Maxima

block([n], n: 1024, while n > 0 do (print(n), n: quotient(n, 2)));

/* using a C-like loop: divide control variable by two instead of incrementing it */
for n: 1024 next quotient(n, 2) while n > 0 do print(n);

MAXScript

a = 1024
while a > 0 do
(
    print a
    a /= 2
)

Metafont

Metafont has no while loop, but it can be "simulated" easily.

a := 1024;
forever: exitif not (a > 0);
  show a;
  a := a div 2;
endfor

min

Works with: min version 0.19.3

1024 :n (n 0 >) (n puts 2 div @n) while

MiniScript

i = 1024
while i > 0
    print i
    i = floor(i/2)
end while

Output:

MIRC Scripting Language

alias while_loop {
  var %n = 10
  while (%n >= 0) {
    echo -a Countdown: %n
    dec %n
  }
}

MIXAL

******************************************
* X = M / N WHILE X > 0
* STORE EACH X IN NUMERIC ARRAY
* PRINT ARRAY
*******************************************
M	EQU	1024		
N	EQU	2
LPR	EQU	18		
BUF0	EQU	100		
MSG	EQU	2000		
LENGTH	EQU	500		
	ORIG	3000
START	IOC	0(LPR) 		
	ENTX	M		
CALC	STX	BUF0,1 		
	DIV	=N=		
	SRAX	5		
	INC1	1		
	JXP	CALC		
	ST1	LENGTH		
PRINT	LDA	BUF0,2		
	CHAR			
	STX	MSG		
	OUT	MSG(LPR)	
	INC2	1		
	CMP2	LENGTH		
	JNE	PRINT
	HLT			
	END	START

МК-61/52

1	0	2	4	П0	ИП0	/-/	x<0	15	ИП0
2	/	П0	БП	05	С/П

Modula-2

MODULE DivBy2;
  IMPORT InOut;

  VAR
    i: INTEGER;
BEGIN
  i := 1024;
  WHILE i > 0 DO
    InOut.WriteInt(i, 4);
    InOut.WriteLn;
    i := i DIV 2
  END
END DivBy2.

Modula-3

The usual module code and imports are omitted.

PROCEDURE DivBy2() =
  VAR i: INTEGER := 1024;
  BEGIN
    WHILE i > 0 DO
      IO.PutInt(i);
      IO.Put("\n");
      i := i DIV 2;
    END;
  END DivBy2;

Monte

var i := 1024
while (i > 0):
    traceln(i)
    i //= 2

MOO

i = 1024;
while (i > 0)
  player:tell(i);
  i /= 2;
endwhile

Morfa

import morfa.io.print;

var i = 1024;
while(i > 0)
{
    println(i);
    i /= 2;
}

Nanoquery

$n = 1024
while ($n > 0)
    println $n
    $n = $n/2
end while

Neko

var i = 1024

while(i > 0) {
    $print(i + "\n");
    i = $idiv(i, 2)
}

Nemerle

mutable x = 1024;
while (x > 0)
{
    WriteLine($"$x");
    x /= 2;
}

Or, with immutable types, after Haskell:

// within another function, eg Main()
def loop(n : int) : void
{
    when (n > 0)
    {
        WriteLine($"$n");
        loop(n / 2);
    }
}
       
loop(1024)

NetRexx

/* NetRexx */
options replace format comments java crossref savelog symbols nobinary

  say
  say 'Loops/While'

  x_ = 1024
  loop while x_ > 0
    say x_.right(6)
    x_ = x_ % 2 -- integer division
    end

NewLISP

(let (i 1024)
  (while (> i 0)
    (println i)
    (setq i (/ i 2))))

Nim

var n: int = 1024
while n > 0:
  echo(n)
  n = n div 2

Oberon-2

The usual module code and imports are ommited.

PROCEDURE DivBy2*();
  VAR i: INTEGER;
BEGIN
  i := 1024;
  WHILE i > 0 DO
    Out.Int(i,0);
    Out.Ln;
    i := i DIV 2;
  END;
END DivBy2;

Objeck

i := 1024;
while(i > 0) {
   i->PrintLine();
   i /= 2;
};

ObjectIcon

An important thing here is the test for termination is a succeed-or-fail operation, rather than a boolean expression. No boolean value is written to the output; rather, the whole statement fails when n reaches zero. The expression 0 < n either returns n or fails.

import io

procedure main()
   local n
   n := 1024
   while n := write(0 < n) / 2
end

Output:

$ oit -s loops_while-oi.icn && ./loops_while-oi

The following also works, and shows a way to avoid having to code the method of writing the output into the loop itself.

import io

procedure main()
  local coexpression
  coexpression := create generator(1024)
  while write(@coexpression)
end

procedure generator(n)
  while 0 < n do { suspend n; n /:= 2 }
end

To show what I mean, let’s do some (merely illustrative) loop-unrolling where the output gets written.

import io

procedure main()
  local coexpression
  coexpression := create generator(1024)
  write(@coexpression) &
      write(@coexpression) &
      write(@coexpression) &
      write(@coexpression)&
      while (write(@coexpression) &
             write(@coexpression) &
             write(@coexpression) &
             write(@coexpression))
end

procedure generator(n)
  while 0 < n do { suspend n; n /:= 2 }
end

OCaml

let n = ref 1024;;
while !n > 0 do
  Printf.printf "%d\n" !n;
  n := !n / 2
done;;

But it is more common to write it in a tail-recursive functional style:

let rec loop n =
  if n > 0 then begin
    Printf.printf "%d\n" n;
    loop (n / 2)
  end
in loop 1024

Octave

i = 1024;
while (i > 0)
  disp(i)
  i = floor(i/2);
endwhile

The usage of the type int32 is not convenient, since the math is done floating point, then rounding to integer, so that 1/2 will be always 1 and never 0.

Odin

Odin only has one loop type: for

package main

import "core:fmt"

main :: proc() {
    for i := 1024 ; i > 0 ; i /= 2 {
		fmt.println(i)
    }
}

Oforth

1024 while ( dup ) [ dup println 2 / ]

OOC

main: func {
  value := 1024
  while (value > 0) {
    value toString() println()
    value /= 2
  }
}

Oz

Oz' for-loop can be used in a C-like manner:

for I in 1024; I>0; I div 2 do
   {Show I}
end

Alternatively, we can use the while feature of the for-loop with a mutable variable:

declare
  I = {NewCell 1024}
in
  for while:@I > 0 do
     {Show @I}
     I := @I div 2
  end

Panda

Panda doesn't have explicit loops, instead we solve it by using the transitive closure operator. It applies a function to each successive value, each unique value is outputted. Our function halves, we make sure that the result is greater than 0 and add newline.

fun half(a) type integer->integer a.divide(2)
1024.trans(func:half).gt(0) nl

Panoramic

dim x%:rem an integer

x%=1024

while x%>0

     print x%
     x%=x%/2

end_while

rem output starts with 1024 and ends with 1.

terminate

PARI/GP

n=1024;
while(n,
  print(n);
  n/=2
);

Pascal

program divby2(output);

var
  i: integer;

begin
  i := 1024;
  while i > 0 do
    begin
      writeln(i);
      i := i div 2
    end
end.

PeopleCode

Local string &CRLF;
Local number &LoopNumber;
&LoopNumber = 1024;
&CRLF = Char(10) | Char(13);

While &LoopNumber > 0;
 WinMessage(&LoopNumber | &CRLF);
 &LoopNumber = &LoopNumber / 2;
End-While;

Perl

my $n = 1024;
while($n){
    print "$n\n";
    $n = int $n / 2;
}

or written as a for-loop and using the bit-shift operator

for(my $n = 1024; $n > 0; $n >>= 1){
    print "$n\n";
}

until (condition) is equivalent to while (not condition).

my $n = 1024;
until($n == 0){
    print "$n\n";
    $n = int $n / 2;
}

Phix

integer i = 1024
while i!=0 do
    ?i
    i = floor(i/2)  -- (see note)
end while

note: using i=i/2 would iterate over 1000 times until i is 4.94e-324 before the final division made it 0, if it didn't typecheck when it got set to 0.5

Phixmonti

/# Rosetta Code problem: https://rosettacode.org/wiki/Loops/While
by Galileo, 11/2022 #/

include ..\Utilitys.pmt

1024 dup while dup ? 2 / int dup endwhile

Output:

1024
512
256
128
64
32
16
8
4
2
1

=== Press any key to exit ===

PHL

var i = 1024;
while (i > 0) {
	printf("%i\n", i);
	i = i/2;
}

PHP

$i = 1024;
while ($i > 0) {
   echo "$i\n";
   $i >>= 1;
}

Picat

While loop

go =>
 N = 1024,
 while (N > 0)
   println(N),
   N := N // 2
 end.

Recursion

Translation of: Prolog

go2 =>
  while_loop(1024).

while_loop(0) => true.
while_loop(N) =>
  println(N),
  while_loop(N//2).

PicoLisp

(let N 1024
   (while (gt0 N)
      (println N)
      (setq N (/ N 2)) ) )

Pike

int main(){
   int i = 1024;
   while(i > 0){
      write(i + "\n");
      i = i / 2;
   }
}

PL/0

var i;
begin
  i := 1024;
  while i > 0 do
  begin
    ! i;
    i := i / 2
  end;
end.

Output:

PL/I

declare i fixed binary initial (1024);

do while (i>0);
   put skip list (i);
   i = i / 2;
end;

PL/SQL

Works with: Oracle

set serveroutput on
declare
  n number := 1024;
begin
  while n > 0 loop
    dbms_output.put_line(n);
    n := trunc(n / 2);
  end loop;
end;
/

Plain English

To run:
Start up.
Show the halvings of 1024.
Wait for the escape key.
Shut down.

To show the halvings of a number:
If the number is 0, exit.
Convert the number to a string.
Write the string to the console.
Divide the number by 2.
Repeat.

Pop11

lvars i = 1024;
while i > 0 do
    printf(i, '%p\n');
    i div 2 -> i;
endwhile;

PostScript

PostScript has no real while loop, but it can easily be created with an endless loop and a check at the beginning:

1024
{
    dup 0 le     % check whether still greater than 0
    { pop exit } % if not, exit the loop
    if
    dup =        % print the number
    2 idiv       % divide by two
}
loop

PowerShell

[int]$i = 1024
while ($i -gt 0) {
    $i
    $i /= 2
}

Prolog

while(0) :- !.
while(X) :-
    writeln(X),
    X1 is X // 2,
    while(X1).

Start the calculation at a top-level like this:

?- while(1024).

Python

n = 1024
while n > 0:
    print n
    n //= 2

Quackery

1024
[ dup 0 > while 
  dup echo cr 2 /
  again ]
drop

Output:

R

i <- 1024L
while(i > 0) 
{
   print(i)
   i <- i %/% 2
}

Racket

Loop/When

#lang racket
(let loop ([n 1024])
  (when (positive? n)
    (displayln n)
    (loop (quotient n 2))))

Macro

#lang racket
(define-syntax-rule (while condition body ...)
  (let loop ()
    (when condition
      body ...
      (loop))))

(define n 1024)
(while (positive? n)
  (displayln n)
  (set! n (sub1 n)))

Raku

(formerly Perl 6)

Here is a straightforward translation of the task description:

my $n = 1024; while $n > 0 { say $n; $n div= 2 }

The same thing with a C-style loop and a bitwise shift operator:

loop (my $n = 1024; $n > 0; $n +>= 1) { say $n }

And here's how you'd really write it, using a sequence operator that intuits the division for you:

.say for 1024, 512, 256 ... 1

REBOL

REBOL [
	Title: "Loop/While"
	URL: http://rosettacode.org/wiki/Loop/While
]

value: 1024
while [value > 0][
	print value
	value: to-integer value / 2
]

ReScript

let n = ref(1024)
while n.contents > 0 {
  Js.log(n.contents)
  n := n.contents / 2
}

Retro

1024 [ cr &putn sip 2 / dup ] while

REXX

version 1, simple

/*REXX program demonstrates a  DO WHILE  with index reduction construct.*/
j=1024                                 /*define the initial value of  J.*/
        do  while  j>0                 /*test if made at the top of  DO.*/
        say j
        j=j%2                          /*in REXX, % is integer division.*/
        end
                                       /*stick a fork in it, we're done.*/

Output:

version 2, right justified

Note that a faster version could be implemented with

DO WHILE x\==0

but that wouldn't be compliant with the wording of the task.

/*REXX program demonstrates a  DO WHILE  with index reduction construct.*/
x=1024                                 /*define the initial value of  X.*/
        do  while  x>0                 /*test if made at the top of  DO.*/
        say right(x,10)                /*pretty output by aligning right*/
        x=x%2                          /*in REXX, % is integer division.*/
        end
                                       /*stick a fork in it, we're done.*/

Output:

version 3, faster WHILE comparison

/*REXX program demonstrates a  DO WHILE  with index reduction construct.*/
x=1024                                 /*define the initial value of  X.*/
        do  while  x>>0                /*this is an  exact  comparison. */
        say right(x,10)                /*pretty output by aligning right*/
        x=x%2                          /*in REXX, % is integer division.*/
        end
                                       /*stick a fork in it, we're done.*/

output is the same as version 2.

version 4, index reduction

/*REXX program demonstrates a  DO WHILE  with index reduction construct.*/
                                       /* [↓] note:   BY   defaults to 1*/  
        do j=1024  by 0  while  j>>0   /*this is an  exact  comparison. */
        say right(j,10)                /*pretty output by aligning right*/
        j=j%2                          /*in REXX, % is integer division.*/
        end
                                       /*stick a fork in it, we're done.*/

output is the same as version 2.

Ring

i = 1024
while i > 0 
      see i + nl
      i = floor(i / 2)
end

Rockstar

The sky is a television on fire.
The floor is ornamental.
The table is hydrochloric.
While the sky is higher than the floor,
Shout the sky.
Put the sky over the table into the sky.
(Rockstar is known for doing that JavaScript quirk, since all numbers are floats, but due to it having the same limits as JavaScript and the most popular interpreter being JS-based, the loop is finite, yet you don't get the result you would get in something like Python.)

RPL

≪ 1024
   WHILE DUP REPEAT
      DUP 1 DISP
      2 /
   END
   DROP
≫

Ruby

i = 1024
while i > 0 do
   puts i
   i /= 2
end

The above can be written in one statement:

puts i = 1024
puts i /= 2 while i > 0

until condition is equivalent to while not condition.

i = 1024
until i <= 0 do
   puts i
   i /= 2
end

Rust

fn main() {
    let mut n: i32 = 1024;
    while n > 0 {
        println!("{}", n);
        n /= 2;
    }
}

SAS

data _null_;
n=1024;
do while(n>0);
  put n;
  n=int(n/2);
end;
run;

Sather

class MAIN is
  main is
    i ::= 1024;
    loop while!(i > 0);
      #OUT + i + "\n";
      i := i / 2;
    end;
  end;
end;

Scala

Library: Scala

Imperative

var i = 1024
while (i > 0) {
  println(i)
  i /= 2
}

Tail recursive

  @tailrec
  def loop(iter: Int) {
    if (iter > 0) {
      println(iter)
      loop(iter / 2)
    }
  }
  loop(1024)

Iterator

  def loop = new Iterator[Int] {
    var i = 1024
    def hasNext = i > 0
    def next(): Int = { val tmp = i; i = i / 2; tmp }
  }
  loop.foreach(println(_))

Stream

Finite stream (1024..0) filtered by takeWhile (1024..1).

  def loop(i: Int): Stream[Int] = i #:: (if (i > 0) loop(i / 2) else Stream.empty)
  loop(1024).takeWhile(_ > 0).foreach(println(_))

Scheme

(do ((n 1024 (quotient n 2)))
    ((<= n 0))
    (display n)
    (newline))

Scilab

Works with: Scilab version 5.5.1

i=1024
while i>0
    printf("%4d\n",i)
    i=int(i/2)
end

Output:

Seed7

$ include "seed7_05.s7i";

const proc: main is func
  local
    var integer: i is 1024;
  begin
    while i > 0 do
      writeln(i);
      i := i div 2
    end while;
  end func;

Output:

SenseTalk

put 1024 into x
log x
Repeat until x = 1
	divide x by 2
	log x
End repeat

SETL

n := 1024;
while n > 0 loop
    print( n );
    n := n div 2;
end loop;

Sidef

var i = 1024
while (i > 0) {
    say i
    i //= 2
}

Simula

Works with: SIMULA-67

begin
  integer i;
  i:=1024;
  while i>0 do
  begin
     outint(i,5);
     i:=i//2-1
  end
end

Output:

 1024  511  254  126   62   30   14    6    2

Slate

#n := 1024.
[n isPositive] whileTrue: 
  [inform: number printString.
   n := n // 2]

Smalltalk

The Block (aka lambda closure) class provides a number of loop messages; with test at begin, test at end and with exit (break).

[s atEnd] whileFalse: [s next. ...].
[foo notNil] whileTrue: [s next. ...].
[...] doWhile: [ ... someBooleanExpression ].
[...] doUntil: [ ... someBooleanExpression ].
[:exit | ... cold ifTrue:[exit value]. ...] loopWithExit

Examples:

number := 1024.
[ number > 0 ] whileTrue: 
  [ Transcript print: number; nl.
  number := number // 2 ]

number := 1024.
[ number <= 0 ] whileFalse: 
  [ Transcript print: number; nl.
  number := number // 2 ]

Sparkling

var i = 1024;
while i > 0 {
    print(i);
    i /= 2;
}

Spin

Works with: BST/BSTC

Works with: FastSpin/FlexSpin

Works with: HomeSpun

Works with: OpenSpin

con
  _clkmode = xtal1 + pll16x
  _clkfreq = 80_000_000

obj
  ser : "FullDuplexSerial.spin"

pub main | n
  ser.start(31, 30, 0, 115200)

  n := 1024
  repeat while n > 0
    ser.dec(n)
    ser.tx(32)
    n /= 2

  waitcnt(_clkfreq + cnt)
  ser.stop
  cogstop(0)

Output:

1024 512 256 128 64 32 16 8 4 2 1

SPL

n = 1024
>
  #.output(n)
  n /= 2
< n!<1

Output:

SQL PL

Works with: Db2 LUW

version 9.7 or higher.

With SQL PL:

--#SET TERMINATOR @

SET SERVEROUTPUT ON @

BEGIN
 DECLARE I SMALLINT DEFAULT 1024;

 Loop: WHILE (I > 0) DO
  CALL DBMS_OUTPUT.PUT_LINE(I);
  SET I = I / 2;
 END WHILE Loop;
END @

Output:

db2 -td@
db2 => SET SERVEROUTPUT ON @
DB20000I  The SET SERVEROUTPUT command completed successfully.
db2 => BEGIN
...
db2 (cont.) => END @
DB20000I  The SQL command completed successfully.

1024
512
256
128
64
32
16
8
4
2
1

Standard ML

val n = ref 1024;
while !n > 0 do (
  print (Int.toString (!n) ^ "\n");
  n := !n div 2
)

But it is more common to write it in a tail-recursive functional style:

let
  fun loop n =
    if n > 0 then (
      print (Int.toString n ^ "\n");
      loop (n div 2)
    ) else ()
in
  loop 1024
end

Stata

local n=1024
while `n'>0 {
	display `n'
	local n=floor(`n'/2)
}

Suneido

i = 1024
while (i > 0)
    {
    Print(i)
    i = (i / 2).Floor()
    }

Output:

Swift

var i = 1024
while i > 0 {
  println(i)
  i /= 2
}

Tailspin

In Tailspin you can loop by sending a value back to the matchers (by "-> #"). Depending on how you set that up, you create different loops.

1024 -> \(
  <0~..> '$;$#10;' -> !OUT::write
    $ ~/ 2 -> #
\) -> !VOID

Tcl

set i 1024
while {$i > 0} {
    puts $i
    set i [expr {$i / 2}]
}

Plain $T e X$

\newcount\rosetta
\rosetta=1024
\loop
    \the\rosetta\endgraf
    \divide\rosetta by 2
    \ifnum\rosetta > 0
\repeat
\end

TorqueScript

This has to make use of mFloor because torque has automatic type shuffling, causing an infiniteloop.

%num = 1024;
while(%num > 0)
{
    echo(%num);
    %num = mFloor(%num / 2);
}

Transact-SQL

DECLARE @i INT = 1024;
WHILE @i >0
BEGIN
    PRINT @i;
    SET @i = @i / 2;
END;

Trith

1024 [dup print 2 / floor] [dup 0 >] while drop

1024 [dup print 1 shr] [dup 0 >] while drop

TUSCRIPT

$$ MODE TUSCRIPT
i=1024
LOOP
   PRINT i
   i=i/2
   IF (i==0) EXIT
ENDLOOP

Output:

Unicon

See Icon.

Uniface

variables
	numeric I
endvariables

I = 1024
while (I > 0)
	putmess I
	I = (I/2)[trunc]
endwhile

UNIX Shell

Works with: Bourne Again SHell

x=1024
while [[ $x -gt 0 ]]; do 
  echo $x
  x=$(( $x/2 ))
done

UnixPipes

(echo 1024>p.res;tail -f p.res) | while read a ; do
   test $a -gt 0 && (expr $a / 2  >> p.res ; echo $a) || exit 0
done

Ursa

decl int n
set n 1024

while (> n 0)
    out n endl console
    set n (int (/ n 2))
end while

Ursala

Unbounded iteration is expressed with the -> operator. An expression (p-> f) x, where p is a predicate and f is a function, evaluates to x, f(x), or f(f(x)), etc. as far as necessary to falsify p.

Printing an intermediate result on each iteration is a bigger problem because side effects are awkward. Instead, the function g in this example iteratively constructs a list of results, which is displayed on termination.

The argument to g is the unit list <1024>. The predicate p is ~&h, the function that tests whether the head of a list is non-null (equivalent to non-zero). The iterated function f is that which conses the truncated half of the head of its argument with a copy of the whole argument. The main program takes care of list reversal and formatting.

#import nat

g = ~&h-> ^C/half@h ~&

#show+

main = %nP*=tx g <1024>

Output:

Explicit iteration has its uses but there are always alternatives. The same output is produced by the following main program using bit manipulation.

main = %nP*=tK33 1024

V

1024 [0 >] [
   dup puts
   2 / >int
] while

Vala

int i = 1024;
while (i > 0) {
  stdout.printf("%d\n", i);
  i /= 2;
}

Vedit macro language

#1 = 1024
while (#1 > 0) {
    Num_Type(#1)
    #1 /= 2
}

or with for loop:

for (#1 = 1024; #1 > 0; #1 /= 2) {
    Num_Type(#1)
}

Verbexx

//  Basic @LOOP while: verb

@LOOP init:{@VAR n = 1024} while:(n > 0) next:{n /= 2}
{
     @SAY n; 
};

Verilog

module main;
  integer i;
  
  initial begin
      i = 1024;
 
      while( i > 0) begin
        $display(i);
        i = i / 2;
      end
      $finish ;
    end
endmodule

Vim Script

let i = 1024
while i > 0
    echo i
    let i = i / 2
endwhile

V (Vlang)

fn main() {
    mut i := 1024
    for i > 0 {
        println(i)
        i /= 2
    }
}

Output:

Wart

i <- 1024
while (i > 0)
  prn i
  i <- (int i/2)

Whitespace

Pseudo-assembly equivalent:

push 1024

0:
    dup onum push 10 ochr
    push 2 div dup
    push 0 swap sub
        jn 0
        pop exit

Wren

var i = 1024
while (i > 0) {
    System.print(i)
    i = (i / 2).floor
}

Output:

X86 Assembly

; NASM 64 bit X86-64 assembly on Linux

global main
extern printf

segment .data

printffmt db `%ld\n`,0

segment .text

main:	                     
    push rbp                    
    mov rbp,rsp 
    
; used rbx and r12 because printf preserves these values
    
    mov rbx,1024                 ; start with 1024
    mov r12,2                    ; load 2 as divisor

.toploop                         ; top of while loop    
    cmp rbx,0                    ; compare to 0
    jle .done                    ; exit 0 or less
    
    lea rdi,[printffmt]          ; print number in rsi
    mov rsi,rbx                  ; mov to rsi as argument
    call printf

; calculate n/2 and save
    xor rdx,rdx                  ; clear rdx for division
    mov rax,rbx                  ; mov number to rax for division
    idiv r12                     ; divide by 2
    mov rbx,rax                  ; save n/2

    jmp .toploop                 ; next loop

.done
    xor rax,rax                  ; return code 0
    leave                        ; fix stack
    ret                          ; return

XLISP

The specification calls for an integer value and for the loop to run WHILE that value is greater than zero. In a dynamically typed language like XLISP, variables cannot be declared as integer or real; but the same result is obtained by looping WHILE the value of the variable i is greater than or equal to one.

(DEFINE I 1024)

(WHILE (>= I 1)
    (PRINT I)
    (DEFINE I (/ I 2)))

XPL0

code CrLf=9, IntOut=11;
int I;
[I:= 1024;
while I>0 do
        [IntOut(0, I);  CrLf(0);
        I:= I>>1;       \(same as I/2 for positive I)
        ];
]

Z80 Assembly

Works with: CP/M 3.1 version YAZE-AG-2.51.2 Z80 emulator

Works with: ZSM4 macro assembler version YAZE-AG-2.51.2 Z80 emulator

Use the /S8 switch on the ZSM4 assembler for 8 significant characters for labels and names

	;
	; while loop, dividing 1024 repeatedly by 2, using Z80 assembly language
	;
	; Runs under CP/M 3.1 on YAZE-AG-2.51.2 Z80 emulator
	; Assembled with zsm4 on same emulator/OS, uses macro capabilities of said assembler
	; Created with vim under Windows
	;
	; Thanks to https://wikiti.brandonw.net for the idea for the conversion routine hl -> decimal ASCII
	;
	;
	; 2023-05-19 Xorph
	;

	;
	; Useful definitions
	;

	bdos	equ 05h		; Call to CP/M BDOS function
	strdel	equ 6eh		; Set string delimiter
	wrtstr	equ 09h		; Write string to console

	nul	equ 00h		; ASCII control characters
	cr	equ 0dh
	lf	equ 0ah

	cnull	equ '0'		; ASCII character constants

	;
	; Macros for BDOS calls
	;

setdel 	macro	char		; Set string delimiter to char
	ld	c,strdel
	ld	e,char
	call	bdos
	endm

print 	macro	msg		; Output string to console
	ld	c,wrtstr
	ld	de,msg
	call	bdos
	endm

newline	macro			; Print newline
	ld	c,wrtstr
	ld	de,crlf
	call	bdos
	endm

	;
	; =====================
	; Start of main program
	; =====================
	;

	cseg

	setdel	nul		; Set string delimiter to 00h
	ld	ix,value	; Register ix points to memory location of counter

while:
	ld	a,(ix)		; Z80 has no 16 bit compare, so we check the value byte by byte for 0
	or	a		; In contrast to other CPUs, loading a register does NOT set the flags
	jr	nz,docalc	; or-ing the accumulator with itself sets the flags and is faster than "cp 0"
	ld	a,(ix+1)
	or	a
	jr	z,endprog	; If both bytes are 0, end program - this jump could be optimized away
				; and replaced with a direct "ret z", but we want to simulate a "real"
				; while loop, so we continue (jump to) after the last loop statement

docalc:
	ld	hl,(value)	; Print the current value, followed by newline
	ld	iy,buffer	; Register iy points to memory location for current value as text for printout
	call	dispHL		; dispHL modifies iy, so it must be reset to the buffer on every iteration

	print	buffer
	newline

	srl	(ix+1)		; Neither has the Z80 a 16 bit shift operation for dividing by 2...
	rr	(ix)		; Shift the MSB of value right and then rotate the LSB with carry to the right

	jr	while		; Next iteration

endprog:
	ret			; Return to CP/M

	;
	; ===================
	; End of main program
	; ===================
	;

	;
	; Helper routines - notice that the Z80 does not have a divide instruction
	; Notice further that CP/M does not have any support for pretty-printing
	; formatted numbers and stuff like that. So we have to do all this by hand...
	;

	;
	; Converts the value (unsigned int) in register hl to its decimal representation
	; Register iy has memory address of target for converted value
	; String is terminated with nul character (\0)
	;

dispHL:
	ld	b,1		; Flag for leading '0'
	irp	x,<-10000,-1000,-100,-10,-1>
	ld	de,x		; Subtract powers of 10 and determine digit
	call	calcdig
	endm

	ld	a,nul		; Terminate result string with nul
	ld	(iy+0),a

	ret			; End of conversion routine

calcdig:
	ld	a,cnull-1	; Determine the digit character
incrdig:
	inc	a		; Start with '0'
	add	hl,de		; As long as subtraction is possible, increment digit character
	jr	c,incrdig

	sbc	hl,de		; If negative, undo last subtraction and continue with remainder
	cp	cnull		; Check for leading '0', these are ignored
	jr	nz,adddig
	bit	0,b		; Use bit instruction for check if flag set, register a contains digit
	ret	nz		; If '0' found and flag set, it is a leading '0' and we return
adddig:
	ld	b,0		; Reset flag for leading '0', we are now outputting digits
	ld	(iy+0),a	; Store character in memory and set iy to next location
	inc	iy

	ret			; End of conversion helper routine

	;
	; ================
	; Data definitions
	; ================
	;

	dseg

value:	defw	1024d		; Starting value for loop, 16 bit little endian
crlf:	defb	cr,lf,nul	; Generic newline
buffer:	defs	10		; Buffer for conversion of number to text

Output:

E>whilelp
1024
512
256
128
64
32
16
8
4
2
1

Zig

const std = @import("std");

pub fn main() void {
    var i: u11 = 1024;
    while (i > 0) : (i /= 2)
        std.debug.print("{}\n", .{i});
}

zkl

n:=1024; while(n>0){println(n); n/=2;}

Output: