Command-line arguments: Difference between revisions

no edit summary
(Shorter notes, remove uneeded comments.)
No edit summary
 
(374 intermediate revisions by more than 100 users not shown)
Line 1:
{{task}}{{selection|Short Circuit|Console Program Basics}} [[Category:Basic language learning]][[Category:Programming environment operations]][[Category:Initialization]]Retrieve the list of command-line arguments given to the program. For programs that only print the arguments when run directly, see [[Scripted main]].
{{task}}
 
See also [[Program name]].
Retrieve the list of command-line arguments given to the program.
 
For parsing command line arguments intelligently, see [[Parsing command-line arguments]].
 
Example command line:
Line 7 ⟶ 9:
myprogram -c "alpha beta" -h "gamma"
 
==[[Ada]]{{header|11l}}==
<code>:argv</code> is a list containing all command line arguments, including the program name.
[[Category:Ada]]
<syntaxhighlight lang="11l">:start:
Command line arguments are available through the pre-defined package Ada.Command_Line.
print(‘Program name: ’:argv[0])
print("Arguments:\n":argv[1..].join("\n"))</syntaxhighlight>
 
=={{header|8080 Assembly}}==
with Ada.Command_line; use Ada.Command_Line;
with Ada.Text_IO; use Ada.Text_IO
procedure Print_Commands is
begin
-- The number of command line arguments is retrieved from the function Argument_Count
-- The actual arguments are retrieved from the function Argument
-- The program name is retrieved from the function Command_Name
Put(Command_Name & " ");
for Arg in 1..Argument_Count loop
Put(Argument(Arg) & " ");
end loop;
New_Line;
end Print_Commands;
 
This program runs under CP/M. CP/M includes some very rudimentary argument parsing: it assumes that the
==[[C]]==
first two space-separated arguments are filenames. Apart from that you can get the raw command line, except
[[Category: C]]
that all lowercase letters are made uppercase.
 
If you need any further parsing, the program needs to do that by itself, which led to many people not bothering.
Command line arguments are passed to main. Since the program name is also passed as "argument", the provided count is actually one more than the number of program arguments. Traditionally the argument count is named argc and the array of argument strings is called argv, but that's not mandatory; any (non-reserved) name will work just as well. It is, however, a good idea to stick to the conventional names.
(The CP/M assembler in particular is infamous for abusing the file extension to take arguments.) The following program shows
everything CP/M gives it.
 
<syntaxhighlight lang="8080asm">putch: equ 2 ; CP/M syscall to print character
Be careful on systems that use Unicode or other multibyte character sets. You may need to use a type of _wchar* and multi-byte-character-set-aware versions of printf.
puts: equ 9 ; CP/M syscall to print $-terminated string
arglen: equ 80h ; Length of argument
argmt: equ 81h ; Argument string
fcb1: equ 5Ch ; FCBs
fcb2: equ 6Ch
org 100h
;;; Print all argument(s) as given
lxi d,cmdln ; Print 'Command line: '
mvi c,puts
call 5
lda arglen ; Retrieve the length of the argument
lxi h,argmt ; Pointer to argument string
call plstr
;;; CP/M also assumes that the first two words on the command
;;; line are filenames, and prepares two FCBs with the filenames
;;; in them. If there are no filenames, they will be blank.
lxi d,file1 ; Print the first one
mvi c,puts
call 5
mvi a,11 ; Filenames are 8+3 characters long and padded with
lxi h,fcb1+1 ; spaces
call plstr
lxi d,file2 ; Print the second one
mvi c,puts
call 5
mvi a,11
lxi h,fcb2+1
; ... fall through - on small systems saving bytes is a virtue
;;; This subroutine prints a length-A string in HL.
plstr: ana a ; If A=0, print nothing.
rz
push psw ; Save A and HL registers on the stack
push h ; (CP/M syscalls clobber all registers)
mov e,m ; Print character under HL
mvi c,putch
call 5
pop h ; Restore A and HL registers
pop psw
inx h ; Increment string pointer
dcr a ; Decrement character counter
jnz plstr ; Print next character if not zero
ret
cmdln: db 'Command line: $'
file1: db 13,10,'File 1: $'
file2: db 13,10,'File 2: $'</syntaxhighlight>
 
{{out}}
#include <stdio.h>
 
<pre>A>args
int main(int argc, char* argv[])
A>args
{
Command line:
int i;
File 1:
printf("This program is named %s.\n", argv[0]);
File 2: for (i = 1; i < argc; ++i)
A>args -c "alpha beta" -h "gamma"
printf("the argument #%d is %s\n", i, argv[i]);
Command line: -C "ALPHA BETA" -H "GAMMA"
File 1: -C
File 2: "ALPHA
A>args foobar.baz barbaz.qux
Command line: FOOBAR.BAZ BARBAZ.QUX
File 1: FOOBAR BAZ
File 2: BARBAZ QUX
</pre>
 
=={{header|8086 Assembly}}==
The method for doing this depends on the memory model of your program. For a COM file, everything is contained in one segment, including the command line arguments. The program starts at offset 100h and the command line arguments start at 81h. It's as simple as reading from that memory address.
 
For an EXE file, both the <code>DS</code> and <code>ES</code> registers are set to the program segment prefix as soon as the program begins. You'll need to load from offset 81h to FFh to get the command line arguments.
=={{header|AArch64 Assembly}}==
{{works with|as|Raspberry Pi 3B version Buster 64 bits <br> or android 64 bits with application Termux }}
<syntaxhighlight lang AArch64 Assembly>
/* ARM assembly AARCH64 Raspberry PI 3B */
/* program commandLine64.s */
/************************************/
/* Constantes */
/************************************/
/* for this file see task include a file in language AArch64 assembly*/
.include "../includeConstantesARM64.inc"
 
/************************************/
/* Initialized data */
/************************************/
.data
szCarriageReturn: .asciz "\n"
/************************************/
/* UnInitialized data */
/************************************/
.bss
.align 4
/************************************/
/* code section */
/************************************/
.text
.global main
main: // entry of program
mov fp,sp // fp <- start address
ldr x4,[fp] // number of Command line arguments
add x5,fp,#8 // first parameter address
mov x2,#0 // init loop counter
1:
ldr x0,[x5,x2,lsl #3] // string address parameter
bl affichageMess // display string
ldr x0,qAdrszCarriageReturn
bl affichageMess // display carriage return
add x2,x2,#1 // increment counter
cmp x2,x4 // number parameters ?
blt 1b // loop
 
100: // standard end of the program
mov x0, #0 // return code
mov x8,EXIT
svc 0 // perform the system call
 
qAdrszCarriageReturn: .quad szCarriageReturn
 
/***************************************************/
/* ROUTINES INCLUDE */
/***************************************************/
/* for this file see task include a file in language AArch64 assembly*/
.include "../includeARM64.inc"
 
</syntaxhighlight>
{{Out}}
<pre>
~/.../rosetta/asm4 $ commandLine64 toto tutu
commandLine64
toto
tutu
</pre>
 
=={{header|Ada}}==
In Ada95 and later versions, command line arguments are available through the predefined package Ada.Command_Line. In Ada83, this would be implementation dependent.
 
<syntaxhighlight lang="ada">with Ada.Command_line; use Ada.Command_Line;
with Ada.Text_IO; use Ada.Text_IO;
 
procedure Print_Commands is
begin
-- The number of command line arguments is retrieved from the function Argument_Count
-- The actual arguments are retrieved from the function Argument
-- The program name is retrieved from the function Command_Name
Put(Command_Name & " ");
for Arg in 1..Argument_Count loop
Put(Argument(Arg) & " ");
end loop;
New_Line;
end Print_Commands;</syntaxhighlight>
 
=== Alternative version using Matreshka ===
 
Uses [http://forge.ada-ru.org/matreshka Matreshka]
 
<syntaxhighlight lang="ada">with Ada.Wide_Wide_Text_IO;
 
with League.Application;
with League.Strings;
 
procedure Main is
begin
for J in 1 .. League.Application.Arguments.Length loop
Ada.Wide_Wide_Text_IO.Put_Line
(League.Application.Arguments.Element (J).To_Wide_Wide_String);
end loop;
end Main;</syntaxhighlight>
 
=={{header|Aikido}}==
The arguments are passed to the program as a vector of strings called <em>args</em>
<syntaxhighlight lang="aikido">
 
foreach arg in args {
println ("arg: " + arg)
}
 
</syntaxhighlight>
 
=={{header|Aime}}==
<syntaxhighlight lang="aime">integer i;
 
i = 0;
while (i < argc()) {
o_text(argv(i));
o_byte('\n');
i += 1;
}</syntaxhighlight>
 
=={{header|ALGOL 68}}==
{{works with|ALGOL 68G|Any - tested with release mk15-0.8b.fc9.i386 - ''argc'' and ''argv'' are not part of the standard's prelude}}
<syntaxhighlight lang="algol68">main:(
FOR i TO argc DO
printf(($"the argument #"g(-0)" is "gl$, i, argv(i)))
OD
)</syntaxhighlight>
Linux command:
/usr/bin/a68g Command-line_arguments.a68 - 1 2 3 ...
Output:
<pre>
the argument #1 is /usr/bin/a68g
the argument #2 is ./Command-line_arguments.a68
the argument #3 is -
the argument #4 is 1
the argument #5 is 2
the argument #6 is 3
the argument #7 is ...
</pre>
=={{header|Amazing Hopper}}==
<p>The main function "main(argv,argc)" is a macro-defined in HOPPER.H: get the arguments, and put them into array ARGV; ARGC have total arguments.</p>
<p>Macro MAIN(ARGV, ARGC):</p>
<syntaxhighlight lang="amazing hopper">
#defn main(_V_,_N_) #RAND, main:, V#RNDV=1,_V_={#VOID}, \
_N_=0,totalarg,mov(_N_), \
LOOPGETARG_#RNDV:, {[ V#RNDV ]},push(_V_),++V#RNDV,\
{_N_,V#RNDV},jle(LOOPGETARG_#RNDV),clear(V#RNDV)
</syntaxhighlight>
VERSION 1:
<syntaxhighlight lang="amazing hopper">
#include <hopper.h>
 
main(argv, argc)
{"This program named: "},[1]get(argv),println
for(i=2, {i}lethan(argc),++i)
{"Argument #",i," = "}[i]get(argv),println
next
exit(0)
</syntaxhighlight>
VERSION 2:
<syntaxhighlight lang="amazing hopper">
#include <hopper.h>
 
main:
total arg, argc=0,mov(argc)
{"This program named: ",[&1]},println
i=2
__CNT_ARGS__:
{"Argumento #",i," = ",[&i]}println
++i,{argc,i}jle(__CNT_ARGS__)
exit(0)
</syntaxhighlight>
VERSION 3:
<syntaxhighlight lang="amazing hopper">
#include <hopper.h>
 
main(argv, argc)
i=2
#hl{
print( "This program named: ",argv[1],"\n")
while( i <= argc )
print("Argument #",i," = ",argv[i],"\n")
i += 1
wend
}
exit(0)
</syntaxhighlight>
VERSION 4:
<syntaxhighlight lang="amazing hopper">
#include <natural.h>
#include <hopper.h>
 
main:
get total arguments, and remember as 'total arguments'.
remember ("This program named: "),
now remember ( argument '1' ), and print with a new line.
secondly, declare 'i', take '2', and store in 'i'
do while ( variable 'i' is less or equal than 'total arguments', \
consider this ( {"Argument #",i," = "} );\
remember ( argument 'i' ); put a new line and print it; finally increment 'i' ).
exit(0)
</syntaxhighlight>
ETCETERA...
{{out}}
<pre>
xu@MrDalien:~/Proyectos/xuesp/HOPPER$ hopper src/args1.com 1 "Rosseta code" 100
This program named: src/args1.com
Argumento #2 = 1
Argumento #3 = Rosseta code
Argumento #4 = 100
</pre>
 
=={{header|AppleScript}}==
 
<syntaxhighlight lang="applescript">
#!/usr/bin/env osascript
-- Print first argument
on run argv
return (item 1 of argv)
end run
</syntaxhighlight>
 
=={{header|ARM Assembly}}==
{{works with|as|Raspberry Pi}}
<syntaxhighlight lang="arm assembly">
/* ARM assembly Raspberry PI */
/* program commandLine.s */
 
/* Constantes */
.equ STDOUT, 1 @ Linux output console
.equ EXIT, 1 @ Linux syscall
.equ WRITE, 4 @ Linux syscall
/* Initialized data */
.data
szCarriageReturn: .asciz "\n"
 
/* UnInitialized data */
.bss
.align 4
 
/* code section */
.text
.global main
main: @ entry of program
push {fp,lr} @ saves registers
add fp,sp,#8 @ fp <- start address
ldr r4,[fp] @ number of Command line arguments
add r5,fp,#4 @ first parameter address
mov r2,#0 @ init loop counter
loop:
ldr r0,[r5,r2,lsl #2] @ string address parameter
bl affichageMess @ display string
ldr r0,iAdrszCarriageReturn
bl affichageMess @ display carriage return
add r2,#1 @ increment counter
cmp r2,r4 @ number parameters ?
blt loop @ loop
 
100: @ standard end of the program
mov r0, #0 @ return code
pop {fp,lr} @restaur registers
mov r7, #EXIT @ request to exit program
swi 0 @ perform the system call
 
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 2 registres
bx lr @ return
 
</syntaxhighlight>
 
=={{header|Arturo}}==
<syntaxhighlight lang="rebol">loop arg 'a [
print a
]</syntaxhighlight>
 
=={{header|AutoHotkey}}==
From the AutoHotkey [http://www.autohotkey.com/docs/Scripts.htm documentation]:
"The script sees incoming parameters as the variables %1%, %2%, and so on. In addition, %0% contains the number of parameters passed (0 if none). "
<syntaxhighlight lang="autohotkey">Loop %0% ; number of parameters
params .= %A_Index% . A_Space
If params !=
MsgBox, %0% parameters were passed:`n`n %params%
Else
Run, %A_AhkPath% "%A_ScriptFullPath%" -c "\"alpha beta\"" -h "\"gamma\""</syntaxhighlight>
 
=={{header|AWK}}==
 
<syntaxhighlight lang="awk">#!/usr/bin/awk -f
 
BEGIN {
print "There are " ARGC "command line parameters"
for(l=1; l<ARGC; l++) {
print "Argument " l " is " ARGV[l]
}
}</syntaxhighlight>
 
=={{header|Babel}}==
==[[C plus plus|C++]]==
Invoke Babel in interactive mode with arguments using the -i switch:
[[Category:C plus plus]]
 
<syntaxhighlight lang="babel">babel -i Larry Mo Curly</syntaxhighlight>
 
Print the argv list with newlines:
 
<syntaxhighlight lang="babel">argv prn !</syntaxhighlight>
 
{{out}}
<pre>Larry
Mo
Curly
</pre>
 
Print the argv list with spaces:
 
<syntaxhighlight lang="babel">argv prs !</syntaxhighlight>
 
{{out}}
<pre>Larry Mo Curly</pre>
 
To access an individual argument, use the ith operator to select an element from the argv list; print with newline using say:
 
<syntaxhighlight lang="babel">argv 1 ith say !</syntaxhighlight>
 
{{out}}
<pre>Mo
</pre>
 
=={{header|BASIC}}==
{{works with|QuickBASIC}}
 
For most older BASICs that supply the keyword <code>COMMAND$</code>, all arguments are returned in a single string that must then be parsed inside the program. (Unlike modern BASICs, there is often no easy way to retrieve the program's name.)
 
<syntaxhighlight lang="qbasic">PRINT "args: '"; COMMAND$; "'"</syntaxhighlight>
 
Sample output:
args: 'This is a test.'
 
{{works with|FreeBASIC}}
 
FreeBASIC supplies three ways to retrieve the arguments: <code>COMMAND$</code> (which works identically to QuickBASIC's <code>COMMAND$</code>), <code>COMMAND$()</code> (a string array which works like [[#C|C]]'s <code>argv[]</code>), and <code>__FB_ARGV__</code> (an array of pointers which works even more like C's <code>argv[]</code>) and __FB_ARGC__ (which works like C's <code>argc</code>).
 
<syntaxhighlight lang="freebasic">DIM i AS INTEGER
 
PRINT COMMAND$
 
PRINT "This program is named "; COMMAND$(0)
i = 1
DO WHILE(LEN(COMMAND$(i)))
PRINT "The argument "; i; " is "; COMMAND$(i)
i = i + 1
LOOP
 
FOR i = 0 TO __FB_ARGC__ - 1
PRINT "arg "; i; " = '"; *__FB_ARGV__[i]; "'"
NEXT i</syntaxhighlight>
 
Sample output:
C:\>cla 1 2 3
1 2 3
This program is named cla
The argument 1 is 1
The argument 2 is 2
The argument 3 is 3
arg 0 = 'cla'
arg 1 = '1'
arg 2 = '2'
arg 3 = '3'
 
==={{header|BaCon}}===
<syntaxhighlight lang="freebasic">' Command line arguments including program name
PRINT "Entire command line: ", ARGUMENT$
 
SPLIT ARGUMENT$ BY " " TO cli$ SIZE args
PRINT "Skip program name:";
FOR i = 1 TO args - 1
PRINT " " & cli$[i];
NEXT
PRINT</syntaxhighlight>
 
{{out}}
<pre>prompt$ bacon command-line.bac
Converting 'command-line.bac'... done, 9 lines were processed in 0.002 seconds.
Compiling 'command-line.bac'... cc -c command-line.bac.c
cc -o command-line command-line.bac.o -lbacon -lm
Done, program 'command-line' ready.
 
prompt$ ./command-line -c "alpha beta" -h "gamma"
Entire command line: ./command-line -c "alpha beta" -h gamma
Skip program name: -c "alpha beta" -h gamma</pre>
 
=={{header|Batch File}}==
{{works with|Windows NT|4 or later (includes Windows XP and onward)}}
<syntaxhighlight lang="dos">@echo off
setlocal enabledelayedexpansion
 
set Count=0
:loop
if not "%1"=="" (
set /a count+=1
set parameter[!count!]=%1
shift
goto loop
)
 
for /l %%a in (1,1,%count%) do (
echo !parameter[%%a]!
)</syntaxhighlight>
 
Another way of doing it
 
<syntaxhighlight lang="dos">::args2.cmd
@echo off
setlocal enabledelayedexpansion
set fn=%~f0
set p0=%~0
set p*=%*
set /a c=1
:loop
if @%1==@ goto done
set p%c%=%~1
set /a c=c+1
shift
goto loop
:done
set /a c=c-1
set p#=%c%
echo fn=%fn%
echo p0=%p0%
echo p*=%p*%
echo p#=%p#%
for /l %%i in (1,1,%p#%) do (
echo p%%i=!p%%i!
)</syntaxhighlight>
 
Invocation:
 
<syntaxhighlight lang="dos">>args2 foo "bar baz" quux
fn=d:\bin\args2.cmd
p0=args2
p*=foo "bar baz" quux
p#=3
p1=foo
p2=bar baz
p3=quux
</syntaxhighlight>
 
=={{header|BBC BASIC}}==
{{works with|BBC BASIC for Windows}}
<syntaxhighlight lang="bbcbasic">PRINT @cmd$</syntaxhighlight>
 
=={{header|Blue}}==
 
Linux/x86-64
 
<syntaxhighlight lang="blue">
global _start
 
: syscall ( num:eax -- result:eax | rcx ) syscall ;
 
: exit ( status:edi -- noret ) 60 syscall ;
: bye ( -- noret ) 0 exit ;
: die ( err:eax -- noret ) neg exit ;
 
: unwrap ( result:eax -- value:eax ) dup 0 cmp ' die xl ;
: ordie ( result -- ) unwrap drop ;
 
1 const stdout
 
: write ( buf:esi len:edx fd:edi -- ) 1 syscall ordie ;
: print ( buf len -- ) stdout write ;
 
: newline ( -- ) s" \n" print ;
: println ( buf len -- ) print newline ;
 
: find0 ( start:rsi -- end:rsi ) lodsb 0 cmp latest xne ;
: cstrlen ( str:rdi -- len:rsi ) dup find0 swap sub dec ;
: cstr>str ( cstr:rdx -- str:rsi len:rdx ) dup cstrlen xchg ;
 
: print-arg ( arg -- ) cstr>str println ;
 
: _start ( rsp -- noret ) dup @ swap
: print-args ( argc:rcx argv:rsp -- noret )
8 add @ print-arg latest loop
bye
;
</syntaxhighlight>
 
=={{header|BQN}}==
BQN has a system value for getting pre-parsed command line arguments.
 
<syntaxhighlight lang="text">•Show •args</syntaxhighlight>
 
should show the full list of args.
 
=={{header|Bracmat}}==
When Bracmat is started with one or more arguments, each argument is evaluated as if it were a Bracmat expression <i>unless</i> an argument (for example the first one) consumes the next argument(s) by calling <code>arg$</code>. Each invocation of <code>arg$</code> pops one argument from the remaining list of arguments. Calling <code>arg$</code> when no more arguments are available results in failure. The following program iterates over all arguments following the currently evaluated argument and outputs the argument to standard output.
<syntaxhighlight lang="text">whl'(arg$:?a&out$(str$("next arg=" !a)))</syntaxhighlight>
Now run Bracmat with this program as the first argument in a DOS environment:
<pre>bracmat "whl'(arg$:?a&out$(str$(\"next arg=\" !a)))" "a" /b -c 2+3 'd;' "out$(\"13+7=\" 13+7)"</pre>
Instead of starting in interactive mode, Bracmat interprets the first argument, which consumes all following arguments. This is output to standard output:
<pre>next arg=a
next arg=/b
next arg=-c
next arg=2+3
next arg='d;'
next arg=out$("13+7=" 13+7)</pre>
If given an argument index, <code>arg$&lt;<i>arg index</i>&gt;</code> returns the indexed argument without consuming any argument.
<pre>bracmat "0:?n&whl'(arg$!n:?a&out$str$(arg[ !n \"]=\" !a)&1+!n:?n)" "a" /b -c 2+3 'd;' "out$(\"13+7=\" 13+7)"</pre>
Output:
<pre>arg[0]=bracmat
arg[1]=0:?n&whl'(arg$!n:?a&out$str$(arg[ !n "]=" !a)&1+!n:?n)
arg[2]=a
arg[3]=/b
arg[4]=-c
arg[5]=2+3
arg[6]='d;'
arg[7]=out$("13+7=" 13+7)
13+7= 20</pre>
The last line demonstrates that not only the first argument is evaluated, but also the following arguments.
 
If Bracmat is run without arguments, Bracmat starts in interactive mode. In that situation calling <code>arg$</code> fails. The same is true if Bracmat is compiled as a shared library (DLL or so).
 
=={{header|C}}==
 
Command line arguments are passed to main. Since the program name is also passed as "argument", the provided count is actually one more than the number of program arguments. Traditionally the argument count is named argc and the array of argument strings is called argv, but that's not mandatory; any (non-reserved) name will work just as well. It is, however, a good idea to stick to the conventional names.
 
Be careful on systems that use Unicode or other multibyte character sets. You may need to use a type of _wchar* and multi-byte-character-set-aware versions of printf.
 
<syntaxhighlight lang="c">#include <stdlib.h>
#include <stdio.h>
 
int main(int argc, char* argv[])
{
int i;
(void) printf("This program is named %s.\n", argv[0]);
for (i = 1; i < argc; ++i)
(void) printf("the argument #%d is %s\n", i, argv[i]);
return EXIT_SUCCESS;
}</syntaxhighlight>
 
=={{header|C sharp|C#}}==
There are at least two methods to access the command-line arguments. The first method is to access the string array passed to Main. This method only accesses the arguments and not the path to the executable.
<syntaxhighlight lang="csharp">using System;
 
namespace RosettaCode {
class Program {
static void Main(string[] args) {
for (int i = 0; i < args.Length; i++)
Console.WriteLine(String.Format("Argument {0} is '{1}'", i, args[i]));
}
}
}</syntaxhighlight>
 
The second method is to call the Environment.GetCommandLineArgs function. This method also returns the path to the executable as args[0] followed by the actual command line arguments.
<syntaxhighlight lang="csharp">using System;
 
namespace RosettaCode {
class Program {
static void Main() {
string[] args = Environment.GetCommandLineArgs();
for (int i = 0; i < args.Length; i++)
Console.WriteLine(String.Format("Argument {0} is '{1}'", i, args[i]));
}
}
}</syntaxhighlight>
 
=={{header|C++}}==
Command line arguments are passed the same way as in C.
 
This example uses <code><iostream></code>. Traditional C-style Ii/Oo also works.
 
<syntaxhighlight lang="cpp">#include <iostream>
int main(int argc, char* argv[])
{
std::cout << "This program is named " << argv[0] << "\n";
std::cout << "There are " << argc-1 << " arguments given.\n";
for (int i = 1; i < argc; ++i)
std::cout << "the argument #" << i << " is " << argv[i] << "\n";
}
 
int main(int argc, const char* argv[]) {
==[[Clean]]==
std::cout << "This program is named " << argv[0] << '\n'
[[Category:Clean]]
<< "There are " << argc - 1 << " arguments given.\n";
for (int i = 1; i < argc; ++i)
std::cout << "The argument #" << i << " is " << argv[i] << '\n';
}</syntaxhighlight>
 
=={{header|C3}}==
 
Command line arguments are passed to main and will be converted to UTF-8 strings on all platforms.
 
<syntaxhighlight lang="c3">import std::io;
 
fn void main(String[] args)
{
io::printfn("This program is named %s.", args[0]);
for (int i = 1; i < args.len; i++)
{
io::printfn("the argument #%d is %s\n", i, args[i]);
}
}</syntaxhighlight>
 
 
=={{header|Clean}}==
<tt>getCommandLine</tt> from the module <tt>ArgEnv</tt> returns an array of command-line arguments (the first element is the name of the program).
 
<syntaxhighlight lang="clean">import ArgEnv
 
Start = getCommandLine</syntaxhighlight>
 
=={{header|Clojure}}==
 
The value of ''*command-line-args*'' is a sequence of the supplied command line arguments, or ''nil'' if none were supplied.
 
<syntaxhighlight lang="clojure">(dorun (map println *command-line-args*))</syntaxhighlight>
 
=={{header|CLU}}==
 
While it is not part of the language standard as specified in the reference manual,
[[Portable CLU]] includes a library function <code>get_argv</code> which returns
all the arguments in order.
 
Note that unlike C, the program name itself is not included in the list of arguments.
 
<syntaxhighlight lang="clu">% This program needs to be merged with PCLU's "useful.lib",
% where get_argv lives.
%
% pclu -merge $CLUHOME/lib/useful.lib -compile cmdline.clu
 
start_up = proc ()
po: stream := stream$primary_output()
args: sequence[string] := get_argv()
for arg: string in sequence[string]$elements(args) do
stream$putl(po, "arg: " || arg)
end
end start_up</syntaxhighlight>
{{out}}
<pre>$ ./cmdline -c "alpha beta" -h "gamma"
arg: -c
arg: alpha beta
arg: -h
arg: gamma</pre>
 
=={{header|COBOL}}==
The COBOL standard appears to say nothing regarding the retrieval of command-line arguments, although methods of retrieving them are provided by most vendors.
 
{{works with|OpenCOBOL}}
{{works with|Visual COBOL}}
 
Getting the arguments in one go, exactly as they were passed in:
<syntaxhighlight lang="cobol"> IDENTIFICATION DIVISION.
PROGRAM-ID. accept-all-args.
DATA DIVISION.
WORKING-STORAGE SECTION.
01 args PIC X(50).
PROCEDURE DIVISION.
main-line.
ACCEPT args FROM COMMAND-LINE
DISPLAY args
GOBACK
.</syntaxhighlight>
 
Getting the arguments one at a time, with arguments being split by whitespace if not in quotes:
<syntaxhighlight lang="cobol"> IDENTIFICATION DIVISION.
PROGRAM-ID. accept-args-one-at-a-time.
DATA DIVISION.
WORKING-STORAGE SECTION.
01 arg PIC X(50) VALUE SPACES.
PROCEDURE DIVISION.
ACCEPT arg FROM ARGUMENT-VALUE
PERFORM UNTIL arg = SPACES
DISPLAY arg
MOVE SPACES TO arg
ACCEPT arg FROM ARGUMENT-VALUE
END-PERFORM
GOBACK
.</syntaxhighlight>
 
Passing arguments from UNIX/Linux Systems to COBOL.
{{works with|OpenCOBOL}}
{{works with|gnuCOBOL}}
<syntaxhighlight lang="cobol">
*>Created By Zwiegnet 8/19/2004
 
IDENTIFICATION DIVISION.
PROGRAM-ID. arguments.
 
ENVIRONMENT DIVISION.
 
DATA DIVISION.
 
 
WORKING-STORAGE SECTION.
 
01 command1 PIC X(50).
01 command2 PIC X(50).
01 command3 PIC X(50).
 
 
PROCEDURE DIVISION.
PERFORM GET-ARGS.
 
*> Display Usage for Failed Checks
ARGUSAGE.
display "Usage: <command1> <command2> <command3>"
STOP RUN.
 
*> Evaluate Arguments
GET-ARGS.
ACCEPT command1 FROM ARGUMENT-VALUE
IF command1 = SPACE OR LOW-VALUES THEN
PERFORM ARGUSAGE
ELSE
INSPECT command1 REPLACING ALL SPACES BY LOW-VALUES
 
 
ACCEPT command2 from ARGUMENT-VALUE
IF command2 = SPACE OR LOW-VALUES THEN
PERFORM ARGUSAGE
ELSE
INSPECT command2 REPLACING ALL SPACES BY LOW-VALUES
 
 
ACCEPT command3 from ARGUMENT-VALUE
IF command3 = SPACE OR LOW-VALUES THEN
PERFORM ARGUSAGE
ELSE
INSPECT command3 REPLACING ALL SPACES BY LOW-VALUES
 
 
*> Display Final Output
display command1 " " command2 " " command3
 
 
STOP RUN.
 
.</syntaxhighlight>
 
=={{header|CoffeeScript}}==
{{works with|Node.js}}
<syntaxhighlight lang="coffeescript">
console.log arg for arg in process.argv
</syntaxhighlight>
 
=={{header|Common Lisp}}==
 
The Common Lisp standard does not specify anything relating to external invocation of a Common Lisp system. The method for getting command-line arguments varies by implementation.
 
The following function could be used to create a uniform way to access the arguments:
 
<syntaxhighlight lang="lisp">(defun argv ()
(or
#+clisp (ext:argv)
#+sbcl sb-ext:*posix-argv*
#+abcl ext:*command-line-argument-list*
#+clozure (ccl::command-line-arguments)
#+gcl si:*command-args*
#+ecl (loop for i from 0 below (si:argc) collect (si:argv i))
#+cmu extensions:*command-line-strings*
#+allegro (sys:command-line-arguments)
#+lispworks sys:*line-arguments-list*
nil))</syntaxhighlight>
 
=={{header|Cowgol}}==
Cowgol includes a function to retrieve command-line arguments in its standard library.
The manner in which arguments are parsed is, however, dependent on the operating system.
 
<syntaxhighlight lang="cowgol">include "cowgol.coh";
include "argv.coh";
 
ArgvInit();
var i: uint8 := 0;
 
loop
var arg := ArgvNext();
if arg == 0 as [uint8] then break; end if;
i := i + 1;
print_i8(i);
print(": '");
print(arg);
print("'\n");
end loop;</syntaxhighlight>
 
{{out}}
On Linux:
<pre>$ ./args -c "alpha beta" -h "gamma"
1: '-c'
2: 'alpha beta'
3: '-h'
4: 'gamma'</pre>
 
On CP/M:
<pre>A>args -c "alpha beta" -h "gamma"
1: '-C'
2: '"ALPHA'
3: 'BETA"'
4: '-H'
5: '"GAMMA"'</pre>
 
=={{header|D}}==
<syntaxhighlight lang="d">void main(in string[] args) {
import std.stdio;
 
foreach (immutable i, arg; args[1 .. $])
writefln("#%2d : %s", i + 1, arg);
}</syntaxhighlight>
 
=={{header|Dart}}==
<syntaxhighlight lang="csharp">main(List<String> args) {
for(var arg in args)
print(arg);
}</syntaxhighlight>
 
=={{header|DCL}}==
case is not preserved unless the parameter is in quotes
<syntaxhighlight lang="dcl">$ i = 1
$ loop:
$ write sys$output "the value of P''i' is ", p'i
$ i = i + 1
$ if i .le. 8 then $ goto loop</syntaxhighlight>
{{out}}
<pre>$ @command_line_arguments -c "alpha beta" -h "gamma"
the value of P1 is -C
the value of P2 is alpha beta
the value of P3 is -H
the value of P4 is gamma
the value of P5 is
the value of P6 is
the value of P7 is
the value of P8 is</pre>
 
=={{header|Delphi}}==
 
<syntaxhighlight lang="delphi">// The program name and the directory it was called from are in
// param[0] , so given the axample of myprogram -c "alpha beta" -h "gamma"
 
for x := 0 to paramcount do
writeln('param[',x,'] = ',param[x]);
 
// will yield ( assuming windows and the c drive as the only drive) :
 
// param[0] = c:\myprogram
// param[1] = -c
// param[2] = alpha beta
// param[3] = -h
// param[4] = gamma
</syntaxhighlight>
 
=={{header|Déjà Vu}}==
Command line arguments are found in <code>!args</code> and <code>!opts</code>.
 
<syntaxhighlight lang="dejavu">for i range 0 -- len !args:
print\( "Argument #" i " is " )
. get-from !args i
 
if has !opts :c:
!print "Ah, the -c option."
 
if has !opts :four:
!. get-from !opts :four</syntaxhighlight>
{{out}}
<pre>$ vu args-3.deja one two -c three --four=five
Argument #0 is "args-3.deja"
Argument #1 is "one"
Argument #2 is "two"
Argument #3 is "three"
Ah, the -c option.
"five"</pre>
 
The order of command line ''options'' is lost.
 
=={{header|Draco}}==
Draco comes with a library function that will return each command line argument
in turn. It simply splits the command line on whitespace, and does not support
quotes.
 
In the example below, the arguments are additionally all made uppercase.
This is however a limitation of the CP/M operating system, and not of Draco.
 
<syntaxhighlight lang="draco">\util.g
 
proc nonrec main() void:
*char par;
word i;
i := 0;
while par := GetPar(); par ~= nil do
i := i + 1;
writeln(i:3, ": '", par, "'")
od
corp</syntaxhighlight>
 
{{out}}
 
<pre>A>param -c "alpha beta" -h "gamma"
1: '-C'
2: '"ALPHA'
3: 'BETA"'
4: '-H'
5: '"GAMMA"'</pre>
 
=={{header|E}}==
 
<syntaxhighlight lang="e">interp.getArgs()</syntaxhighlight>
 
=={{header|Eiffel}}==
 
This class inherits functionality for dealing with command line arguments from class <code lang="eiffel">ARGUMENTS</code>. It uses the feature <code lang="eiffel">separate_character_option_value</code> to return the values by option name for each of the two arguments.
 
<syntaxhighlight lang="eiffel ">class
APPLICATION
inherit
ARGUMENTS
create
make
feature {NONE} -- Initialization
make
-- Print values for arguments with options 'c' and 'h'.
do
print ("Command line argument value for option 'c' is: ")
print (separate_character_option_value ('c') + "%N")
print ("Command line argument value for option 'h' is: ")
print (separate_character_option_value ('h') + "%N")
io.read_line -- Keep console window open
end
end</syntaxhighlight>
 
Output (for command line arguments: -c "alpha beta" -h "gamma"):
<pre>
Command line argument value for option 'c' is: alpha beta
Command line argument value for option 'h' is: gamma
</pre>
 
=={{header|Elena}}==
ELENA 6.x :
<syntaxhighlight lang="elena">import system'routines;
import extensions;
public program()
Start = getCommandLine
{
program_arguments.forEvery::(int i)
{ console.printLine("Argument ",i," is ",program_arguments[i]) }
}</syntaxhighlight>
{{out}}
<pre>
Argument 0 is myprogram.exe
Argument 1 is -c
Argument 2 is alpha beta
Argument 3 is -h
Argument 4 is gamma
</pre>
 
=={{header|Elixir}}==
==[[E]]==
Elixir provides command line arguments via the <tt>System.argv()</tt> function.
[[Category:E]]
<syntaxhighlight lang="elixir">#!/usr/bin/env elixir
IO.puts 'Arguments:'
Enum.map(System.argv(),&IO.puts(&1))</syntaxhighlight>
Example run:
<syntaxhighlight lang="bash">$ ./show-args.exs a b=2 --3 -4
Arguments:
a
b=2
--3
-4</syntaxhighlight>
 
=={{header|Emacs Lisp}}==
interp.getArgs()
 
<syntaxhighlight lang="lisp">(while argv
(message "Argument: %S" (pop argv)))</syntaxhighlight>
 
Invoke script:
 
emacs --script test.el foo bar baz
 
=={{header|Erlang}}==
When used as a script language the arguments is a list to the main/1 function. When compiled use init:get_arguments/0
<syntaxhighlight lang="erlang">3> init:get_arguments().</syntaxhighlight>
result
<syntaxhighlight lang="erlang">[{root,["/usr/erlang/erl5.5"]},
{progname,["erl"]},
{home,["/home/me"]},
{c,["alpha beta"]},
{h,["gamma"]}]</syntaxhighlight>
 
init:get_argument(name) can be used to fetch value of a particular flag
 
<syntaxhighlight lang="erlang">4> init:get_argument(h).
{ok,[["gamma"]]}
5> init:get_argument(c).
{ok,[["alpha beta"]]}</syntaxhighlight>
 
=={{header|Euphoria}}==
<syntaxhighlight lang="euphoria">constant cmd = command_line()
printf(1,"Interpreter/executable name: %s\n",{cmd[1]})
printf(1,"Program file name: %s\n",{cmd[2]})
if length(cmd)>2 then
puts(1,"Command line arguments:\n")
for i = 3 to length(cmd) do
printf(1,"#%d : %s\n",{i,cmd[i]})
end for
end if</syntaxhighlight>
 
=={{header|F_Sharp|F#}}==
The entry-point function accepts the comment line arguments as an array of strings. The following program will print each argument on a separate line.
<syntaxhighlight lang="fsharp">#light
[<EntryPoint>]
let main args =
Array.iter (fun x -> printfn "%s" x) args
0</syntaxhighlight>
 
=={{header|Factor}}==
USING: io sequences command-line ;
(command-line) [ print ] each
 
=={{header|Fancy}}==
<syntaxhighlight lang="fancy">ARGV each: |a| {
a println # print each given command line argument
}</syntaxhighlight>
 
=={{header|Fantom}}==
 
<syntaxhighlight lang="fantom">
class Main
{
public static Void main (Str[] args)
{
echo ("command-line args are: " + args)
}
}
</syntaxhighlight>
 
==[[{{header|Forth]]}}==
[[Category:Forth]]
Access to command line arguments is not a standard feature of Forth, since it is designed to be used without an operating system. The popular GNU implementation gforth runs from a shell and can access command line arguments similar to C: variable '''argc''' contains the count (including the command itself) and '''arg''' is a function that returns the ''nth'' argument as a string.
 
'''Interpreter:'''{{works with|gforth |0.6.2}}
<syntaxhighlight lang="forth">\ args.f: print each command line argument on a separate line
: main
argc @ 0 do i arg type cr loop ;
 
main bye</syntaxhighlight>
\ args.f: print each command line argument on a separate line
: main
argc @ 0 do i arg type cr loop ;
main bye
 
Here is output from a sample run.
<syntaxhighlight lang="forth">$ gforth args.f alpha "beta gamma" delta
gforth
args.f
alpha
beta gamma
delta
$</syntaxhighlight>
$
 
=={{header|Fortran}}==
{{works with|Fortran|2003 and later}}
<syntaxhighlight lang="fortran">program command_line_arguments
 
implicit none
integer, parameter :: len_max = 256
integer :: i , nargs
character (len_max) :: arg
nargs = command_argument_count()
!nargs = iargc()
do i = 0, nargs
call get_command_argument (i, arg)
!call getarg (i, arg)
write (*, '(a)') trim (arg)
end do
 
end program command_line_arguments
</syntaxhighlight>
Note: This sample uses the Fortran 2003 intrinsic routines <code>command_argument_count</code> and <code>get_command_argument</code> instead of the nonstandard extensions <code>iargc</code> and <code>getarg</code>. Most Fortran compilers support both.
 
Sample usage:
<syntaxhighlight lang="text">> ./a.out -c "alpha beta" -h "gamma"
./a.out
-c
alpha beta
-h
gamma</syntaxhighlight>
 
=={{header|FreeBASIC}}==
<syntaxhighlight lang="freebasic">' FB 1.05.0 Win64
 
' Program (myprogram.exe) invoke as follows:
' myprogram -c "alpha beta" -h "gamma"
 
Print "The program was invoked like this => "; Command(0) + " " + Command(-1)
Print
Print "Press any key to quit"
Sleep</syntaxhighlight>
 
{{out}}
<pre>
The program was invoked like this => myprogram -c alpha beta -h gamma
</pre>
 
=={{header|Frink}}==
Arguments to a program are available in the <CODE>ARGS</CODE> array variable.
<syntaxhighlight lang="frink">
println[ARGS]
</syntaxhighlight>
 
=={{header|FunL}}==
<syntaxhighlight lang="funl">println( args )</syntaxhighlight>
 
=={{header|Gambas}}==
'''[https://gambas-playground.proko.eu/?gist=a1374aa441520314ad0c7decb1e91c97 Click this link to run this code]'''
<syntaxhighlight lang="gambas">PUBLIC SUB main()
DIM l AS Integer
DIM numparms AS Integer
DIM parm AS String
numparms = Application.Args.Count
FOR l = 0 TO numparms - 1
parm = Application.Args[l]
PRINT l; " : "; parm
NEXT
END</syntaxhighlight>
 
=={{header|Genie}}==
<syntaxhighlight lang="genie">[indent=4]
/*
Command line arguments, in Genie
 
valac commandLine.gs
./commandLine sample arguments 'four in total here, including args 0'
*/
 
init
 
// Output the number of arguments
print "%d command line argument(s):", args.length
 
// Enumerate all command line arguments
for s in args
print s
 
// to reiterate, args[0] is the command
if args[0] is not null
print "\nWith Genie, args[0] is the command: %s", args[0]</syntaxhighlight>
 
{{out}}
<pre>prompt$ valac commandLine.gs
prompt$ ./commandLine -c "alpha beta" -h "gamma"
5 command line argument(s):
./commandLine
-c
alpha beta
-h
gamma
 
With Genie, args[0] is the command: ./commandLine</pre>
 
=={{header|Global Script}}==
 
Command-line arguments are passed to the main program as a linked list of strings (which are also linked lists).
 
This uses the <code>gsio</code> I/O operations, which are designed to be simple to implement on top of Haskell and simple to use. It also uses impmapM, which is a specific specialization of mapM for the HSGS implementation.
<syntaxhighlight lang="global script">λ 'as. impmapM (λ 'a. print qq{Argument: §(a)\n}) as</syntaxhighlight>
 
=={{header|Go}}==
<syntaxhighlight lang="go">
package main
import (
"fmt"
"os"
)
 
func main() {
for i, x := range os.Args[1:] {
fmt.Printf("the argument #%d is %s\n", i, x)
}
}
</syntaxhighlight>
 
=={{header|Groovy}}==
Command-line arguments are accessible via the '''args''' list variable. The following is saved as the file "Echo.groovy":
<syntaxhighlight lang="groovy">println args</syntaxhighlight>
 
The existence of command-line arguments presupposes the existence of a command line interpreter. The following test runs were entered in a cygwin bash shell in a Microsoft Windows XP system:
<pre>$ groovy Echo this is an argument list
[this, is, an, argument, list]
$ groovy Echo -x alkfrew4oij -cdkjei +22
[-x, alkfrew4oij, -cdkjei, +22]
$</pre>
 
For more sophisticated command-line option and option-argument parsing use the '''CliBuilder''' (command-line interface builder) library, which extends the functionality of the Java-based '''Apache Commons CLI''' library to Groovy.
 
=={{header|Harbour}}==
Uses the Harbour-specific hb_PValue() function
<syntaxhighlight lang="visualfoxpro">PROCEDURE Main()
 
LOCAL i
 
FOR i := 1 TO PCount()
? "argument", hb_ntos( i ), "=", hb_PValue( i )
NEXT
 
RETURN</syntaxhighlight>
 
==[[{{header|Haskell]]}}==
[[Category:Haskell]]
 
Defined by the System module, getArgs :: IO [String] provides the command-line arguments in a list.
 
myprog.hs:
<syntaxhighlight lang="haskell">import System
main = getArgs >>= print</syntaxhighlight>
<pre>
myprog a -h b c
=> ["a","-h","b","c"]
</pre>
 
=={{header|HicEst}}==
myprog a -h b c
<syntaxhighlight lang="hicest">DO i = 2, 100 ! 1 is HicEst.exe
=> ["a","-h","b","c"]
EDIT(Text=$CMD_LINE, SePaRators='-"', ITeM=i, IF ' ', EXit, ENDIF, Parse=cmd, GetPosition=position)
IF(position > 0) WRITE(Messagebox) cmd
ENDDO</syntaxhighlight>
 
=={{header|Icon}} and {{header|Unicon}}==
==[[Java]]==
Command line parameters are passed to Icon/Unicon programs as a list of strings.
[[Category:Java]]
<syntaxhighlight lang="icon">procedure main(arglist)
every write(!arglist)
end</syntaxhighlight>
 
{{libheader|Icon Programming Library}} includes [http://www.cs.arizona.edu/icon/library/procs/options.htm options] that parses the command line as switches and arguments and returns the results in a table.
public class Arguments {
 
public static void main(String[] args) {
=={{header|Io}}==
System.out.println("There are " + args.length + " arguments given.");
<syntaxhighlight lang="io">System args foreach(a, a println)</syntaxhighlight>
for(int i = 0; i < args.length; i++)
 
System.out.println("The argument #" + (i+1) + " is " + args[i] + "and is at index " + i);
=={{header|Ioke}}==
}
<syntaxhighlight lang="ioke">System programArguments each(println)</syntaxhighlight>
}
 
=={{header|J}}==
 
The global <code>ARGV</code> holds the command line arguments. Thus, a program to display them:
 
<syntaxhighlight lang="j"> ARGV</syntaxhighlight>
 
In a non-interactive context, we would instead use <code>echo ARGV</code>.
 
=={{header|Jakt}}==
The main function can recieve the command line arguments as an arbitrarily named array of String. argv[0] will be the name of the program.
 
<syntaxhighlight lang="jakt">
fn main(arguments: [String]) {
println("{}", arguments)
}
</syntaxhighlight>
 
=={{header|Java}}==
The arguments will be passed to <code>main</code> as the only parameter.<br />
The array is non-null.
<syntaxhighlight lang="java">
public static void main(String[] args)
</syntaxhighlight>
Running this command
<syntaxhighlight lang="bash">
myprogram -c "alpha beta" -h "gamma"
</syntaxhighlight>
Will produce the following
<pre>
-c
alpha beta
-h
gamma
</pre>
<br />
And alternate demonstration.
<syntaxhighlight lang="java">public class Arguments {
public static void main(String[] args) {
System.out.println("There are " + args.length + " arguments given.");
for(int i = 0; i < args.length; i++)
System.out.println("The argument #" + (i+1) + " is " + args[i] + " and is at index " + i);
}
}</syntaxhighlight>
 
For more sophisticated command-line option and option-argument parsing use the [http://commons.apache.org/cli '''Apache Commons CLI'''] (command-line interface) library.
 
=={{header|JavaScript}}==
{{works with|Node.js}}
<syntaxhighlight lang="javascript">process.argv.forEach((val, index) => {
console.log(`${index}: ${val}`);
});</syntaxhighlight>
{{works with|JScript}}
<syntaxhighlight lang="javascript">var objArgs = WScript.Arguments;
for (var i = 0; i < objArgs.length; i++)
WScript.Echo(objArgs.Item(i));</syntaxhighlight>
{{works with|JScript.NET (compiled with jsc.exe)}}
<syntaxhighlight lang="javascript">import System;
var argv:String[] = Environment.GetCommandLineArgs();
for (var i in argv)
print(argv[i]);</syntaxhighlight>
{{works with|Rhino}}
{{works with|SpiderMonkey}}
<syntaxhighlight lang="javascript">for (var i = 0; i < arguments.length; i++)
print(arguments[i]);</syntaxhighlight>
 
=={{header|Joy}}==
<syntaxhighlight lang="joy">#!/usr/bin/joy
 
argv rest put.</syntaxhighlight>
{{out}}
<pre>["-c" "alpha beta" "-h" "gamma"]</pre>
 
=={{header|jq}}==
{{works with|jq|after February 17, 2017}}
<br/>
jq distinguishes between command-line arguments and command-line options.
Only the former are available programmatically.
 
Specifically, both named and positional command-line arguments are available in the
global constant '''$ARGS''', a JSON object, as follows:
 
$ARGS.positional contains an array of the positional arguments as JSON strings
 
$ARGS.names is a JSON object of giving the mapping of name to value.
 
For example, the invocation:
 
$ jq -n '$ARGS' --args a b
 
yields:<syntaxhighlight lang="json">{
"positional": [
"a",
"b"
],
"named": {}
}</syntaxhighlight>
 
Arguments specified with ''--args'' are always read as JSON strings; arguments specified with ''--jsonargs''
are interpreted as JSON, as illustrated here:
<pre>
$ jq -n '$ARGS' --argjson x 0 --jsonargs 0 '{"a":1}'
{
"positional": [
0,
{
"a": 1
}
],
"named": {
"x": 0
}
}</pre>
 
=={{header|Jsish}}==
<syntaxhighlight lang="javascript">#!/usr/local/bin/jsish
puts(Info.argv0());
puts(console.args);</syntaxhighlight>
 
{{out}}
<pre>$ jsish command-line-arguments.jsi -c "alpha beta" -h "gamma"
/home/btiffin/lang/jsish/command-line-arguments.jsi
[ "-c", "alpha beta", "-h", "gamma" ]</pre>
 
=={{header|Julia}}==
Works when the Julia program is run as a file argument to julia.exe.
<syntaxhighlight lang="julia">using Printf
 
prog = Base.basename(Base.source_path())
 
println(prog, "'s command-line arguments are:")
for s in ARGS
println(" ", s)
end
</syntaxhighlight>
 
{{out}}
<pre>
$ julia command_line_arguments.jl -c "alpha beta" -h "gamma"
command_line_arguments.jl's command-line arguments are:
-c
alpha beta
-h
gamma
</pre>
 
=={{header|Klong}}==
Command line arguments (but not the program name itself) are bound
to the variable ".a". The following program prints them, one argument
per line:
 
<syntaxhighlight lang="k">
.p'.a
</syntaxhighlight>
 
=={{header|Kotlin}}==
{{trans|Java}}
<syntaxhighlight lang="scala">fun main(args: Array<String>) {
println("There are " + args.size + " arguments given.")
args.forEachIndexed { i, a -> println("The argument #${i+1} is $a and is at index $i") }
}</syntaxhighlight>
{{out}}
See Java output.
 
=={{header|Lang}}==
In lang command line arguments are stored in &LANG_ARGS.
<syntaxhighlight lang="lang">
$ele
foreach($[ele], &LANG_ARGS) {
fn.println($ele)
}
</syntaxhighlight>
Calling a Lang program with command line arguments depends on the implementation.
The following example shows, how this can be achieved in Standard Lang (-- defines the start of the command line arguments for the Lang program):
<syntaxhighlight lang="shell">
$ lang cmdarg.lang -- 2 abc test text
</syntaxhighlight>
{{out}}
<pre>
2
abc
test
text
</pre>
 
=={{header|Lasso}}==
<syntaxhighlight lang="lasso">#!/usr/bin/lasso9
 
iterate($argv) => {
stdoutnl("Argument " + loop_count + ": " + loop_value)
}</syntaxhighlight>
Output:
<syntaxhighlight lang="shell">$ lasso9 arguments.lasso -c "alpha beta" -h "gamma"
Argument 1: arguments.lasso
Argument 2: -c
Argument 3: alpha beta
Argument 4: -h
Argument 5: gamma</syntaxhighlight>
 
=={{header|LFE}}==
 
To demonstrate this, we can start the LFE REPL up with the parameters for this example:
<syntaxhighlight lang="shell">
$ ./bin/lfe -pa ebin/ -c "alpha beta" -h "gamma"
</syntaxhighlight>
 
Once we're in the shell, we can get all the initializing arguments with this call:
<syntaxhighlight lang="lisp">
> (: init get_arguments)
(#(root ("/opt/erlang/r15b03"))
#(progname ("erl"))
#(home ("/Users/oubiwann"))
#(user ("lfe_boot"))
#(pa ("ebin/"))
#(c ("alpha beta"))
#(h ("gamma")))
</syntaxhighlight>
 
We can also get specific arguments if we know their keys:
<syntaxhighlight lang="lisp">
> (: init get_argument 'c)
#(ok (("alpha beta")))
> (: init get_argument 'h)
#(ok (("gamma")))
</syntaxhighlight>
 
=={{header|Liberty BASIC}}==
<syntaxhighlight lang="lb">print CommandLine$</syntaxhighlight>
 
=={{header|Lingo}}==
<syntaxhighlight lang="lingo">put the commandline
-- "-c alpha beta -h gamma"</syntaxhighlight>
 
In latest versions of Mac OS X, the above approach doesn't work anymore. But there is a free "Xtra" (binary plugin/shared library) called "CommandLine Xtra" that works both in Windows and Mac OS X and returns the command-line parsed into a lingo list (array):
 
{{libheader|CommandLine Xtra}}
<syntaxhighlight lang="lingo">put getCommandLineArgs()
-- ["-c", "alpha beta", "-h", "gamma"]</syntaxhighlight>
 
=={{header|Logo}}==
{{works with|UCB Logo|5.6}}
If the command line to a logo script is written
logo file.logo - arg1 arg2 arg3
Then the arguments after the "-" are found in a list in variable :COMMAND.LINE
<syntaxhighlight lang="logo">show :COMMAND.LINE
[arg1 arg2 arg3]</syntaxhighlight>
Alternatively, make the first line of an executable logo script:
#! /usr/bin/logo -
to be able to invoke the script with arguments.
file.logo arg1 arg2 arg3
 
=={{header|LSE64}}==
<syntaxhighlight lang="lse64">argc , nl # number of arguments (including command itself)
0 # argument
dup arg dup 0 = || ,t 1 + repeat
drop</syntaxhighlight>
drop
 
=={{header|Lua}}==
 
The lua scripting language does not use argc and argv conventions for the command line parameters. Instead, the command line parameters to the main script are provided through the global table arg. The script name is placed into element zero of arg, and the script parameters go into the subsequent elements:
 
<syntaxhighlight lang="lua">print( "Program name:", arg[0] )
 
print "Arguments:"
for i = 1, #arg do
print( i," ", arg[i] )
end</syntaxhighlight>
 
=={{header|M2000 Interpreter}}==
function quote$("a") return "a" a string in ""
 
Arguments in command line maybe two kinds, in first part those three letters identifiers with + or - for switches for interpreter and the last part to executed before loading the actual script.
 
For this example we make a script, save to temporary directory, and call it passing arguments. We can use Win as shell substitute in M2000 environment, or the Use statement. Reading the shell statement Win we can see how the command line composed. We call the m2000.exe in the appdir$ (application directory, is the path to M2000.exe), and pass a string as a file with a path. That path will be the current path for the new start of m2000.exe the host for M2000 Interpreter (an activeX dll).
 
<syntaxhighlight lang="m2000 interpreter">
Module Checkit {
Document a$ = {
Module Global A {
Show
Read a$="nothing", x=0
Print a$, x
A$=Key$
}
A: End
}
Dir temporary$
Save.doc a$, "program.gsb"
\\ open if gsb extension is register to m2000.exe
Win quote$(dir$+"program.gsb")
\\ +txt is a switch for interpreter to use string comparison as text (not binary)
\\ so we can send switches and commands before the program loading
Win appdir$+"m2000.exe", quote$(dir$+"program.gsb +txt : Data {Hello}, 100")
\\ no coma after name (we can use "program.gsb" for names with spaces)
Use program.gsb "From Use statement", 200
\\ delete file
Wait 5000
Dos "del "+quote$(dir$+"program.gsb");
\\ open directory
Rem : Win temporary$
}
Checkit
</syntaxhighlight>
 
=={{header|Mathematica}}/{{header|Wolfram Language}}==
myprogram:
<syntaxhighlight lang="mathematica">#!/usr/local/bin/MathematicaScript -script
$CommandLine</syntaxhighlight>
Output:
<pre>{myprogram,-c,alpha beta,-h,gamma}</pre>
 
=={{header|Mercury}}==
<syntaxhighlight lang="text">
:- module cmd_line_args.
:- interface.
 
:- import_module io.
:- pred main(io::di, io::uo) is det.
 
:- implementation.
:- import_module int, list, string.
 
main(!IO) :-
io.progname("", ProgName, !IO),
io.format("This program is named %s.\n", [s(ProgName)], !IO),
io.command_line_arguments(Args, !IO),
list.foldl2(print_arg, Args, 1, _, !IO).
 
:- pred print_arg(string::in, int::in, int::out, io::di, io::uo) is det.
 
print_arg(Arg, ArgNum, ArgNum + 1, !IO) :-
io.format("the argument #%d is %s\n", [i(ArgNum), s(Arg)], !IO).
</syntaxhighlight>
 
=={{header|min}}==
{{works with|min|0.19.3}}
<syntaxhighlight lang="min">args</syntaxhighlight>
 
=={{header|MMIX}}==
<syntaxhighlight lang="mmix">argv IS $1
argc IS $0
i IS $2
 
LOC #100
Main LOC @
SETL i,1 % i = 1
Loop CMP $3,argc,2 % argc < 2 ?
BN $3,1F % then jump to end
XOR $255,$255,$255 % clear $255
8ADDU $255,i,argv % i*8 + argv
LDOU $255,$255,0 % argv[i]
TRAP 0,Fputs,StdOut % write the argument
GETA $255,NewLine % add a newline
TRAP 0,Fputs,StdOut
INCL i,1 % increment index
SUB argc,argc,1 % argc--
BP argc,Loop % argc > 0? then Loop
1H LOC @
XOR $255,$255,$255 % exit(0)
TRAP 0,Halt,0
 
NewLine BYTE #a,0</syntaxhighlight>
 
=={{header|Modula-2}}==
<syntaxhighlight lang="modula2">MODULE try;
 
FROM Arguments IMPORT GetArgs, ArgTable, GetEnv;
FROM InOut IMPORT WriteCard, WriteLn, WriteString;
 
VAR count, item : SHORTCARD;
storage : ArgTable;
 
BEGIN
GetArgs (count, storage);
WriteString ('Count ='); WriteCard (count, 4); WriteLn;
item := 0;
REPEAT
WriteCard (item, 4);
WriteString (' : ');
WriteString (storage^ [item]^);
WriteLn;
INC (item)
UNTIL item = count
END try.</syntaxhighlight>
Example:
<syntaxhighlight lang="modula-2">
jan@Beryllium:~/modula/test$ try jantje zag eens pruimen hangen
Count = 6
0 : try
1 : jantje
2 : zag
3 : eens
4 : pruimen
5 : hangen
</syntaxhighlight>
 
=={{header|Modula-3}}==
Command line parameters are accessed using the <tt>Params</tt> module.
<syntaxhighlight lang="modula3">MODULE Args EXPORTS Main;
 
IMPORT IO, Params;
 
BEGIN
IO.Put(Params.Get(0) & "\n");
IF Params.Count > 1 THEN
FOR i := 1 TO Params.Count - 1 DO
IO.Put(Params.Get(i) & "\n");
END;
END;
END Args.</syntaxhighlight>
 
Output:
<pre>
martin@thinkpad:~$ ./prog
./prog
martin@thinkpad:~$ ./prog 10
./prog
10
martin@thinkpad:~$ ./prog 10 20
./prog
10
20
</pre>
 
=={{header|Nanoquery}}==
{{trans|Ursa}}
<syntaxhighlight lang="nanoquery">//
// command-line arguments
//
 
// output all arguments
for i in range(0, len(args) - 1)
println args[i]
end</syntaxhighlight>
{{out}}
<pre>$ java -jar ../nanoquery-2.3_1700.jar -b cmdline.nq "alpha beta" -h "gamma"
-b
cmdline.nq
alpha beta
-h
gamma</pre>
 
=={{header|Neko}}==
<syntaxhighlight lang="actionscript">/* command line arguments, neko */
var argc = $asize($loader.args)
 
/* Display count and arguments, indexed from 0, no script name included */
$print("There are ", argc, " arguments\n")
 
var arg = 0
while arg < argc $print($loader.args[arg ++= 1], "\n")</syntaxhighlight>
 
{{out}}
<pre>prompt$ nekoc command-line-arguments.neko
prompt$ neko ./command-line-arguments.n -c "alpha beta" -h "gamma"
There are 4 arguments
-c
alpha beta
-h
gamma</pre>
 
=={{header|Nemerle}}==
<syntaxhighlight lang="nemerle">using System;
using System.Console;
 
module CLArgs
{
Main(args : array[string]) : void
{
foreach (arg in args) Write($"$arg "); // using the array passed to Main(), everything after the program name
Write("\n");
def cl_args = Environment.GetCommandLineArgs(); // also gets program name
foreach (cl_arg in cl_args) Write($"$cl_arg ");
}
}</syntaxhighlight>
 
=={{header|NetRexx}}==
In a stand-alone application NetRexx places the command string passed to it in a variable called <tt>arg</tt>.
<syntaxhighlight lang="netrexx">/* NetRexx */
-- sample arguments: -c "alpha beta" -h "gamma"
say "program arguments:" arg
</syntaxhighlight>
'''Output:'''
<pre>
program arguments: -c "alpha beta" -h "gamma"
</pre>
 
=={{header|Nim}}==
<syntaxhighlight lang="nim">import os
echo "program name: ", getAppFilename()
echo "Arguments:"
for arg in commandLineParams():
echo arg</syntaxhighlight>
 
=={{header|Nu}}==
In Nu, the special <code>main</code> function can be declared, which gets passed the cli arguments.
<syntaxhighlight lang="nu">
def main [...argv] {
$argv | print
}
</syntaxhighlight>
{{out}}
<pre>
~> nu cli.nu A B C "Hello World!"
╭───┬──────────────╮
│ 0 │ A │
│ 1 │ B │
│ 2 │ C │
│ 3 │ Hello World! │
╰───┴──────────────╯
</pre>
 
=={{header|Oberon-2}}==
{{works with|oo2c}}
<syntaxhighlight lang="oberon2">
MODULE CommandLineArguments;
IMPORT
NPCT:Args,
Out := NPCT:Console;
 
BEGIN
Out.String("Args number: ");Out.Int(Args.Number(),0);Out.Ln;
Out.String("0.- : ");Out.String(Args.AsString(0));Out.Ln;
Out.String("1.- : ");Out.String(Args.AsString(1));Out.Ln;
Out.String("2.- : ");Out.String(Args.AsString(2));Out.Ln;
Out.String("3.- : ");Out.String(Args.AsString(3));Out.Ln;
Out.String("4.-: ");Out.String(Args.AsString(4));Out.Ln
END CommandLineArguments.
</syntaxhighlight>
{{out}}
<pre>
Args number: 5
0.- : bin/CommandLineArguments
1.- : -c
2.- : alpha beta
3.- : -h
4.-: gamma
</pre>
 
=={{header|Objeck}}==
<syntaxhighlight lang="objeck">
class Line {
function : Main(args : String[]) ~ Nil {
each(i : args) {
args[i]->PrintLine();
};
}
}
</syntaxhighlight>
 
=={{header|Objective-C}}==
 
In addition to the regular C mechanism of arguments to main(), Objective-C also has another way to get the arguments as string objects inside an array object:
<syntaxhighlight lang="objc">NSArray *args = [[NSProcessInfo processInfo] arguments];
NSLog(@"This program is named %@.", [args objectAtIndex:0]);
NSLog(@"There are %d arguments.", [args count] - 1);
for (i = 1; i < [args count]; ++i){
NSLog(@"the argument #%d is %@", i, [args objectAtIndex:i]);
}</syntaxhighlight>
 
=={{header|OCaml}}==
 
The program name is also passed as "argument", so the array length is actually one more than the number of program arguments.
 
<syntaxhighlight lang="ocaml">let () =
Printf.printf "This program is named %s.\n" Sys.argv.(0);
for i = 1 to Array.length Sys.argv - 1 do
Printf.printf "the argument #%d is %s\n" i Sys.argv.(i)
done</syntaxhighlight>
 
=== Using the [https://ocaml.org/api/Arg.html Arg] module ===
<syntaxhighlight lang="ocaml">(* default values *)
let somebool = ref false
let somestr = ref ""
let someint = ref 0
 
let usage = "usage: " ^ Sys.argv.(0) ^ " [-b] [-s string] [-d int]"
 
let speclist = [
("-b", Arg.Set somebool, ": set somebool to true");
("-s", Arg.Set_string somestr, ": what follows -s sets some string");
("-d", Arg.Set_int someint, ": some int parameter");
]
 
let () =
(* Read the arguments *)
Arg.parse
speclist
(fun x -> raise (Arg.Bad ("Bad argument : " ^ x)))
usage;
Printf.printf " %b %d '%s'\n" !somebool !someint !somestr</syntaxhighlight>
<pre>
$ ocaml arg.ml --help
usage: arg.ml [-b] [-s string] [-d int]
-b : set somebool to true
-s : what follows -s sets some string
-d : some int parameter
-help Display this list of options
--help Display this list of options
 
$ ocaml arg.ml -d 4 -b -s blabla
true 4 'blabla'
 
$ ocaml arg.ml
false 0 ''
</pre>
 
=={{header|Odin}}==
 
<syntaxhighlight lang="odin">package main
 
import "core:os"
import "core:fmt"
 
main :: proc() {
fmt.println(os.args)
}
 
// Run: ./program -c "alpha beta" -h "gamma"
// Output: ["./program", "-c", "alpha beta", "-h", "gamma"]
</syntaxhighlight>
 
=={{header|Oforth}}==
 
System.Args returns command line arguments.
 
All arguments that begin with "--" are not included into this list.
The first argument is the program name, so this list is never empty.
 
<syntaxhighlight lang="oforth">System.Args println</syntaxhighlight>
 
=={{header|Oz}}==
===Raw arguments===
Like in C, but without the program name:
<syntaxhighlight lang="oz">functor
import Application System
define
ArgList = {Application.getArgs plain}
{ForAll ArgList System.showInfo}
{Application.exit 0}
end</syntaxhighlight>
 
===Preprocessed arguments===
<syntaxhighlight lang="oz">functor
import Application System
define
ArgSpec =
record(
c(type:string single %% option "--c" expects a string, may only occur once,
optional:false char:&c) %% is not optional and has a shortcut "-c"
h(type:string single %% option "--h" expects a string, may only occur once,
default:"default h" %% is optional and has a default value if not given
char:&h) %% and has a shortcut "-h"
)
Args = {Application.getArgs ArgSpec}
{System.showInfo Args.c}
{System.showInfo Args.h}
{Application.exit 0}
end</syntaxhighlight>
 
=={{header|Pascal}}==
==={{header|Free Pascal}}===
<syntaxhighlight lang="pascal">
Program listArguments(input, output, stdErr);
 
Var
i: integer;
Begin
writeLn('program was called from: ',paramStr(0));
For i := 1 To paramCount() Do
Begin
writeLn('argument',i:2,' : ', paramStr(i));
End;
End.
</syntaxhighlight>
{{out}}
<pre>
./Commandlinearguments -c "alpha beta" -h "gamma"
program was called from: /home/user/Documents/GitHub/rosettacode/Commandlinearguments
argument 1 : -c
argument 2 : alpha beta
argument 3 : -h
argument 4 : gamma
</pre>
 
==[[Perl]]==
[[Category:Perl]]
'''Interpreter:''' [[Perl]] v5.x
 
=={{header|Perl}}==
{{works with|Perl|5.x}}
@ARGV is the array containing all command line parameters
 
<syntaxhighlight lang="perl">my @params = @ARGV;
my $secondparams_size = $@ARGV[1];
my $fifthsecond = $ARGV[41];
my $fifth = $ARGV[4];</syntaxhighlight>
 
If you don't mind importing a module:
==[[Pop11]]==
[[Category:Pop11]]
 
<syntaxhighlight lang="perl">use Getopt::Long;
variable poparglist contains list of command line arguments (as strings).
GetOptions (
One can use iteration over list to process then (for example print).
'help|h' => \my $help,
'verbose|v' => \my $verbose,
);</syntaxhighlight>
 
=={{header|Phix}}==
lvars arg;
<!--<syntaxhighlight lang="phix">(phixonline)-->
for arg in poparglist do
<span style="color: #008080;">with</span> <span style="color: #008080;">javascript_semantics</span>
printf(arg, '->%s<-\n');
<span style="color: #008080;">constant</span> <span style="color: #000000;">cmd</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">command_line</span><span style="color: #0000FF;">()</span>
endfor;
<span style="color: #0000FF;">?</span><span style="color: #000000;">cmd</span>
<span style="color: #008080;">if</span> <span style="color: #000000;">cmd</span><span style="color: #0000FF;">[</span><span style="color: #000000;">1</span><span style="color: #0000FF;">]=</span><span style="color: #000000;">cmd</span><span style="color: #0000FF;">[</span><span style="color: #000000;">2</span><span style="color: #0000FF;">]</span> <span style="color: #008080;">then</span>
<span style="color: #7060A8;">printf</span><span style="color: #0000FF;">(</span><span style="color: #000000;">1</span><span style="color: #0000FF;">,</span><span style="color: #008000;">"Compiled executable name: %s\n"</span><span style="color: #0000FF;">,{</span><span style="color: #000000;">cmd</span><span style="color: #0000FF;">[</span><span style="color: #000000;">1</span><span style="color: #0000FF;">]})</span>
<span style="color: #008080;">else</span>
<span style="color: #7060A8;">printf</span><span style="color: #0000FF;">(</span><span style="color: #000000;">1</span><span style="color: #0000FF;">,</span><span style="color: #008000;">"Interpreted (using %s) source name: %s\n"</span><span style="color: #0000FF;">,</span><span style="color: #000000;">cmd</span><span style="color: #0000FF;">[</span><span style="color: #000000;">1</span><span style="color: #0000FF;">..</span><span style="color: #000000;">2</span><span style="color: #0000FF;">])</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">if</span>
<span style="color: #008080;">if</span> <span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">cmd</span><span style="color: #0000FF;">)></span><span style="color: #000000;">2</span> <span style="color: #008080;">then</span>
<span style="color: #7060A8;">puts</span><span style="color: #0000FF;">(</span><span style="color: #000000;">1</span><span style="color: #0000FF;">,</span><span style="color: #008000;">"Command line arguments:\n"</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">for</span> <span style="color: #000000;">i</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">3</span> <span style="color: #008080;">to</span> <span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">cmd</span><span style="color: #0000FF;">)</span> <span style="color: #008080;">do</span>
<span style="color: #7060A8;">printf</span><span style="color: #0000FF;">(</span><span style="color: #000000;">1</span><span style="color: #0000FF;">,</span><span style="color: #008000;">"#%d : %s\n"</span><span style="color: #0000FF;">,{</span><span style="color: #000000;">i</span><span style="color: #0000FF;">,</span><span style="color: #000000;">cmd</span><span style="color: #0000FF;">[</span><span style="color: #000000;">i</span><span style="color: #0000FF;">]})</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">for</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">if</span>
<!--</syntaxhighlight>-->
When interpreting, the first two elements returned by command_line() are {interpreter,source}.<br>
When compiled, the first two elements are instead {executable,executable}, so the parameters (if any) are consistently the 3rd element onwards.
{{out}}
desktop/Phix (Windows, Linux would be fairly similar but obviously with different paths):
<pre>
C:\Program Files (x86)\Phix>p testcl -c "alpha beta" -h "gamma"
{"C:\\Program Files (x86)\\Phix\\p.exe","C:\\Program Files (x86)\\Phix\\testcl.exw","-c","alpha beta","-h","gamma"}
Interpreted (using C:\Program Files (x86)\Phix\p.exe) source name: C:\Program Files (x86)\Phix\testcl.exw
Command line arguments:
#3 : -c
#4 : alpha beta
#5 : -h
#6 : gamma
C:\Program Files (x86)\Phix>p -c testcl -c "alpha beta" -h "gamma"
{"C:\\Program Files (x86)\\Phix\\testcl.exe","C:\\Program Files (x86)\\Phix\\testcl.exe","-c","alpha beta","-h","gamma"}
Compiled executable name: C:\Program Files (x86)\Phix\testcl.exe
Command line arguments:
#3 : -c
#4 : alpha beta
#5 : -h
#6 : gamma
C:\Program Files (x86)\Phix>testcl -c "alpha beta" -h "gamma"
{"C:\\Program Files (x86)\\Phix\\testcl.exe","C:\\Program Files (x86)\\Phix\\testcl.exe","-c","alpha beta","-h","gamma"}
Compiled executable name: C:\Program Files (x86)\Phix\testcl.exe
Command line arguments:
#3 : -c
#4 : alpha beta
#5 : -h
#6 : gamma
C:\Program Files (x86)\Phix>
</pre>
pwa/p2js: There isn't really a command line, you always get a result of length 2.
<pre>
{"p2js","file:///C:/Program%20Files%20(x86)/Phix/pwa/test.htm"}
Interpreted (using p2js) source name: file:///C:/Program%20Files%20(x86)/Phix/pwa/test.htm
</pre>
 
==[[Python]]{{header|PHP}}==
When PHP is run from the command line, the special variables <tt>$argv</tt> and <tt>$argc</tt> contain the array of arguments, and the number of arguments, respectively. The program name is passed as the first argument.
[[Category:Python]]
 
<syntaxhighlight lang="php"><?php
$program_name = $argv[0];
$second_arg = $argv[2];
$all_args_without_program_name = array_shift($argv);
</syntaxhighlight>
 
=={{header|Picat}}==
Picat has no built-in option parser, and the user must write a specific parser for each use case. The arguments to a Picat programs are available via <code>main/1</code> as a list of strings.
 
<syntaxhighlight lang="picat">main(ARGS) =>
println(ARGS).
main(_) => true.</syntaxhighlight>
 
{{out}}
<pre>picat command_line_arguments.pi -c "alpha beta" -h "gamma"
[-c,alpha beta,-h,gamma]</pre>
 
See the task [http://rosettacode.org/wiki/Command-line_arguments Command-line_arguments] for a simple option parser.
 
=={{header|PicoLisp}}==
There are three ways to handle command-line arguments in PicoLisp:
 
1. Obtain all arguments as a list of strings via '[http://software-lab.de/doc/refA.html#argv argv]'
 
2. Fetch each argument individually with '[http://software-lab.de/doc/refO.html#opt opt]'
 
3. Use the built-in [http://software-lab.de/doc/ref.html#invoc command-line interpretation], where arguments starting with a hyphen are executed as functions.
 
Here we use the third option, as it is not so obvious, sometimes more flexible,
and in fact the most commonly used one for application development.
 
We define 'c' and 'h' as functions, which retrieve their argument with 'opt',
and then '[http://software-lab.de/doc/refL.html#load load]' all remaining
command line arguments.
<syntaxhighlight lang="picolisp">#!/usr/bin/picolisp /usr/lib/picolisp/lib.l
 
(de c ()
(prinl "Got 'c': " (opt)) )
 
(de h ()
(prinl "Got 'h': " (opt)) )
 
(load T)
(bye)</syntaxhighlight>
Output:
<pre>$ ./myprogram -c "alpha beta" -h "gamma"
Got 'c': alpha beta
Got 'h': gamma</pre>
 
=={{header|PL/I}}==
<syntaxhighlight lang="pli">
/* The entire command line except the command word itself is passed */
/* to the parameter variable in PL/I. */
program: procedure (command_line) options (main);
declare command_line character (100) varying;
 
...
 
end program;
</syntaxhighlight>
 
=={{header|Pop11}}==
 
variable poparglist contains list of command line arguments (as strings). One can use iteration over list to process then (for example print).
 
<syntaxhighlight lang="pop11">lvars arg;
for arg in poparglist do
printf(arg, '->%s<-\n');
endfor;</syntaxhighlight>
 
=={{header|PowerBASIC}}==
For versions of PowerBASIC prior to [[PB/Win]] 9 and [[PB/CC]] 5, the only option available is identical to the one used by [[#BASIC|QuickBASIC]] above:
<syntaxhighlight lang="powerbasic">? "args: '"; COMMAND$; "'"</syntaxhighlight>
 
Current versions of PowerBASIC (with the likely exception of [[PB/DOS]]) include <code>COMMAND$()</code> that works similarly to [[FreeBASIC]]'s <code>COMMAND$()</code>, except that you can't retrieve the application's name:
<syntaxhighlight lang="powerbasic">'these two both return ALL args
? COMMAND$
? COMMAND$(0)
 
DO WHILE(LEN(COMMAND$(i)))
PRINT "The argument "; i; " is "; COMMAND$(i)
i = i + 1
LOOP</syntaxhighlight>
 
=={{header|PowerShell}}==
In PowerShell the arguments to a script can be accessed with the <code>$args</code> array:
<syntaxhighlight lang="powershell">$i = 0
foreach ($s in $args) {
Write-Host Argument (++$i) is $s
}</syntaxhighlight>
 
=={{header|Prolog}}==
The command line arguments supplied to a Prolog interpreter can be accessed by passing <code>os_argv</code> to <code>current_prolog_flag/2</code>.
<syntaxhighlight lang="prolog">:-
current_prolog_flag(os_argv, Args),
write(Args).</syntaxhighlight>
 
Alternatively, <code>argv</code> can be used to access only the arguments *not* consumed by the Prolog interpreter.
<syntaxhighlight lang="prolog">:-
current_prolog_flag(argv, Args),
write(Args).</syntaxhighlight>
 
This omits the interpreter name, the input Prolog filename, and any other arguments directed at the Prolog interpreter.
 
=={{header|Pure}}==
Arguments are in global variables, argc and argv.
 
<syntaxhighlight lang="pure">
using system;
 
printf "There are %d command line argumants\n" argc;
puts "They are " $$ map (puts) argv;
</syntaxhighlight>
 
=={{header|PureBasic}}==
 
===Reading all parameters===
You can easily read all parameters by using ProgramParameter() without argument.
<syntaxhighlight lang="purebasic">If OpenConsole()
Define n=CountProgramParameters()
PrintN("Reading all parameters")
While n
PrintN(ProgramParameter())
n-1
Wend
Print(#CRLF$+"Press Enter")
Input()
CloseConsole()
EndIf</syntaxhighlight>
===Reading specific parameters===
You can specify which parameter 'n' to read.
<syntaxhighlight lang="purebasic">If OpenConsole()
Define n
PrintN("Reading specific pameters")
For n=0 To CountProgramParameters()
PrintN(ProgramParameter(n))
Next
Print(#CRLF$+"Press Enter")
Input()
CloseConsole()
EndIf</syntaxhighlight>
 
=={{header|Python}}==
''sys.argv'' is a list containing all command line arguments, including the program name. Typically you slice the list to access the actual command line argument:
 
<syntaxhighlight lang="python">import sys
program_name = sys.argv[0]
arguments = sys.argv[1:]
count = len(arguments)</syntaxhighlight>
 
When running a module by invoking Python, the Python interpreter processes and removes some of the arguments, and the module cannot access them. To process command line arguments, run the module directly. sys.argv is a copy of the command line arguments; modifying sys.argv will not change the arguments seen by other processes, e.g. ps. (In other words sys.argv is an object which contains a copy of the process' command line arguments ... modifying that copy is only visible from within the Python program and not externally).
 
For powerful option parsing capabilities check out the [http://docs.python.org/library/optparse.html optparse] module.
 
=={{header|R}}==
 
Following adapted from [http://quantitative-ecology.blogspot.com/2007/08/including-arguments-in-r-cmd-batch-mode.html this post by Forester]:
 
Suppose you want to call your script <tt>test.r</tt> with the arguments <tt>a=1 b=c(2,5,6)</tt>, where <tt>b</tt> is a numeric vector. Suppose you also want to redirect your output to <tt>test.out</tt> (not that you have a choice--I still don't know how to make R send shell-script output to stdout). You would then run
 
<syntaxhighlight lang="r">R CMD BATCH --vanilla --slave '--args a=1 b=c(2,5,6)' test.r test.out</syntaxhighlight>
 
from the commandline, with the following text in <tt>test.r</tt>:
 
<syntaxhighlight lang="r"># Read the commandline arguments
args <- (commandArgs(TRUE))
 
# args is now a list of character vectors
# First check to see if any arguments were passed,
# then evaluate each argument.
if (length(args)==0) {
print("No arguments supplied.")
# Supply default values
a <- 1
b <- c(1,1,1)
} else {
for (i in 1:length(args)) {
eval(parse(text=args[[i]]))
}
}
print(a*2)
print(b*3)</syntaxhighlight>
 
(possibly preceding code that actually does something :-) Your output <tt>test.out</tt> would then contain (e.g., if you <tt>cat</tt> it)
 
<pre>[1] 2
[1] 6 15 18
> proc.time()
user system elapsed
0.168 0.026 0.178
</pre>
 
If you know how to get the output
 
* sent to stdout (i.e., as is normal with shell scripts)
* done without the profiling
 
please update this example!
 
=={{header|Racket}}==
 
The following is the simplest program that prints the command-line arguments:
 
<syntaxhighlight lang="scheme">#lang racket
(current-command-line-arguments)</syntaxhighlight>
 
You can also explicitly print each argument to standard output:
 
<syntaxhighlight lang="scheme">#lang racket
 
(for ([arg (current-command-line-arguments)]) (displayln arg))</syntaxhighlight>
 
=={{header|Raku}}==
(formerly Perl 6)
Perl 5's <code>@ARGV</code> is available as <code>@*ARGS</code>. Alternatively, if you define a subroutine named <code>MAIN</code>, Perl will automatically process <code>@*ARGS</code> according to Unix conventions and <code>MAIN</code>'s signature (or signatures, if your <code>MAIN</code> is a multi sub) and then call <code>MAIN</code> with appropriate arguments; for more detailed information see: https://docs.raku.org/language/create-cli
 
<syntaxhighlight lang="raku" line># with arguments supplied
$ raku -e 'sub MAIN($x, $y) { say $x + $y }' 3 5
8
 
# missing argument:
$ raku -e 'sub MAIN($x, $y) { say $x + $y }' 3
Usage:
-e '...' x y</syntaxhighlight>
 
If the program is stored in a file, the file name is printed instead of <code>-e '...'</code>.
 
=={{header|RapidQ}}==
<syntaxhighlight lang="rapidq">PRINT "This program is named "; Command$(0)
FOR i=1 TO CommandCount
PRINT "The argument "; i; " is "; Command$(i)
NEXT i</syntaxhighlight>
 
=={{header|Raven}}==
 
<syntaxhighlight lang="raven">ARGS print
 
stack (6 items)
0 => "raven"
1 => "myprogram"
2 => "-c"
3 => "alpha beta"
4 => "-h"
5 => "gamma"</syntaxhighlight>
 
=={{header|REALbasic}}==
<syntaxhighlight lang="vb">Function Run(args() as String) As Integer
For each arg As String In args
Stdout.WriteLine(arg)
Next
End Function</syntaxhighlight>
Output (given arguments: ''--foo !bar "bat bang"''):
appName.exe
--foo
!bar
bat bang
 
=={{header|REXX}}==
The entire command line arguments (as a single string) are passed by REXX to the program.
<syntaxhighlight lang="rexx">say 'command arguments:'
say arg(1)</syntaxhighlight>
Input:
 
myprogram -c "alpha beta" -b "gamma"
 
However, in the above example (as shown), it's suggested that (maybe)
only options that start with a minus (-) are to be examined
and assumed to be options.
<syntaxhighlight lang="rexx">parse arg aaa /*get the arguments. */
/*another version: */
/* aaa=arg(1) */
say 'command arguments:'
say aaa
 
opts='' /*placeholder for options. */
data='' /*placeholder for data. */
 
do j=1 to words(aaa)
x=word(aaa,j)
if left(x,1)=='-' then opts=opts x /*Option? Then add to opts.*/
else data=data x /*Must be data. Add to data.*/
end
 
/*the above process adds a leading blank to OPTS and DATA*/
 
opts=strip(opts,'L') /*strip leading blanks. */
data=strip(data,'l') /*strip leading blanks. */
say
say 'options='opts
say ' data='data</syntaxhighlight>
 
;Note to users of Microsoft Windows and Regina REXX:
Note that some REXX pass the command line as is, but Regina REXX lets
the operating system parse it first (for instance Windows), and
Windows will pass anything &nbsp; (along to the program being invoked) &nbsp; inside double quotes (<big>"</big>) to the program as is. &nbsp; Any other data not in double quotes is passed as is.
 
Output from Regina REXX under Windows with the invocation:
 
myprogram -c "alpha beta" -h "gamma"
 
<pre>
command arguments:
-c alpha beta -h gamma
 
options=-c -h
data=alpha beta gamma
</pre>
Output from others REXXes under Windows with the invocation:
 
myprogram -c "alpha beta" -h "gamma"
 
<pre>
command arguments:
-c "alpha beta" -h "gamma"
 
options=-c -h
data="alpha beta" "gamma"
</pre>
 
 
;Notes to UNIX users:
The rexx programming language does not preserve command line parameters containing spaces. This renders it unsuitable for use for wrapperscript applications, where filenames containing spaces need to be preserved, because there is no way to differentiate between a space within a parameter, and a space used to separate parameters.
 
;Scenario:
If a script is called as follows:
 
ccwrapper "foo bar.c" "bar bar.c"
 
From the shell:
 
argv[0] = ccwrapper
argv[1] = foo bar.c
argv[2] = bar bar.c
 
It is a requirement of a wrapper that !argv[1] and !argv[2] are preserved when passed to the target application (a C compiler, in this example). Current implementations of rexx treat the command line arguments as one long argument:
 
arg() = 1
arg(1) = "foo bar.c bar bar.c"
 
The [parser] would separates the command line arguments by spaces. this results in !argv[1] and !argv[2] becoming split, so the target application would be called with different arguments:
 
argv[1] = foo
argv[2] = bar.c
argv[3] = bar
argv[4] = bar.c
 
This has a different meaning to the compiler, so the arguments forwarded from rexx are rendered useless.
 
;Workaround:
A workaround would be to create a wrapper around the rexx interpreter that encodes the command-line before calling rexx. The rexx application then decodes it. Some rexx interpreters, such as [[regina]] also provide a !-a switch as a workaround.
 
=={{header|Ring}}==
When running a module by invoking Python, Python process and remove some of the arguments, and the module can not access them. To process command line arguments, run the module directly. sys.argv is a copy of the command line arguments; modifying sys.argv will not change the arguments seen by other processes, e.g. ps.
<syntaxhighlight lang="ring">
see copy("=",30) + nl
see "Command Line Parameters" + nl
see "Size : " + len(sysargv) + nl
see sysargv
see copy("=",30) + nl
for x = 1 to len(sysargv)
see x + nl
next
</syntaxhighlight>
 
==[[{{header|Ruby]]}}==
[[Category:Ruby]]
Command line arguments are available in the constant Object::ARGV.
 
myprog:
<syntaxhighlight lang="ruby">#! /usr/bin/env ruby
p ARGV</syntaxhighlight>
 
myprog a -h b c
=> ["a","-h","b","c"]
 
==[[Tcl]]{{header|Rust}}==
<syntaxhighlight lang="rust">use std::env;
[[Category: Tcl]]
 
fn main(){
The pre-defined variable <tt>argc</tt> contains the number of arguments passed to the routine, <tt>argv</tt> contains the arguments as a list. Retrieving the second argument might look something like this:
let args: Vec<_> = env::args().collect();
println!("{:?}", args);
}</syntaxhighlight>
Run:
<syntaxhighlight lang="text">./program -c "alpha beta" -h "gamma"
["./program", "-c", "alpha beta", "-h", "gamma"]</syntaxhighlight>
 
=={{header|S-lang}}==
if { $argc > 1 } { puts [lindex $argv 1] }
The command-line arguments exist in the array __argv:
<syntaxhighlight lang="s-lang">variable a;
foreach a (__argv)
print(a);
</syntaxhighlight>Note 1: This array can be changed by calling
 
__set_argc_argv(new_argv);
 
Note 2: When a script is run as a parameter to slsh, __argv does not
include slsh. __argv[0] is simply the name of the script.
 
The same is true if running a script via the editor jed, as in:
 
jed -script FILE [arg ...]
 
However, when [x/w]jed is entered normally, __argv consists of the
command-line for the editor itself, with __argv[0] == jed or /path/to/jed
or the equivalent.
 
=={{header|Sather}}==
<syntaxhighlight lang="sather">class MAIN is
main(args:ARRAY{STR}) is
loop
#OUT + args.elt! + "\n";
end;
end;
end;</syntaxhighlight>
 
As in C (and others), the first element is the command itself (exactly as it is written in the command line and after shell variable expansion); e.g.
 
<pre>$ /home/home/rosetta/sather/a.out arg1 arg2 arg3</pre>
 
prints
 
<pre>/home/home/rosetta/sather/a.out
arg1
arg2
arg3</pre>
 
=={{header|Scala}}==
{{libheader|Scala}}
Calling Scala from command line means invoking a method called <code>main</code>, defined on an
<code>object</code>, whose type is <code>(Array[String]):Unit</code>, meaning it receives an
array of strings, and returns unit. That array contains the command line arguments.
 
<syntaxhighlight lang="scala">object CommandLineArguments extends App {
println(s"Received the following arguments: + ${args.mkString("", ", ", ".")}")
}</syntaxhighlight>
 
When running a Scala script, where the whole body is executed, the arguments get stored in an array of strings called <code>argv</code>:
 
<syntaxhighlight lang="scala">println(s"Received the following arguments: + ${argv.mkString("", ", ", ".")}")</syntaxhighlight>
 
=={{header|Scheme}}==
 
<syntaxhighlight lang="scheme"> (define (main args)
(for-each (lambda (arg) (display arg) (newline)) args))</syntaxhighlight>
 
=={{header|Seed7}}==
 
<syntaxhighlight lang="seed7">$ include "seed7_05.s7i";
 
const proc: main is func
local
var integer: i is 0;
begin
writeln("This program is named " <& name(PROGRAM) <& ".");
for i range 1 to length(argv(PROGRAM)) do
writeln("The argument #" <& i <& " is " <& argv(PROGRAM)[i]);
end for;
end func;</syntaxhighlight>
 
=={{header|Sidef}}==
Command line arguments are available in the ARGV array.
<syntaxhighlight lang="ruby">say ARGV;</syntaxhighlight>
{{out}}
<pre>% myprog -c "alpha beta" -h "gamma"
['-c', 'alpha beta', '-h', 'gamma']
</pre>
 
=={{header|Slate}}==
<syntaxhighlight lang="slate">StartupArguments do: [| :arg | inform: arg]</syntaxhighlight>
 
=={{header|Slope}}==
Command line arguments are available in the '''sys-args''' list.
<syntaxhighlight lang="slope">(display sys-args)</syntaxhighlight>
 
=={{header|Smalltalk}}==
{{works with|GNU Smalltalk}}
<syntaxhighlight lang="smalltalk">(1 to: Smalltalk getArgc) do: [ :i |
(Smalltalk getArgv: i) displayNl
]</syntaxhighlight>
 
{{works with|Smalltalk/X}}
<syntaxhighlight lang="smalltalk">Smalltalk commandLineArguments printCR.
Smalltalk commandLineArguments do:[:each | each printCR]</syntaxhighlight>
 
=={{header|SparForte}}==
As a structured script.
<syntaxhighlight lang="ada">#!/usr/local/bin/spar
pragma annotate( summary, "printargs" )
@( description, "Retrieve the list of command-line arguments given to the program." )
@( description, "Example command line: " )
@( description, "myprogram -c 'alpha beta' -h 'gamma'" )
@( category, "tutorials" )
@( author, "Ken O. Burtch" )
@( see_also, "http://rosettacode.org/wiki/Command-line_arguments" );
pragma license( unrestricted );
 
pragma software_model( nonstandard );
pragma restriction( no_external_commands );
 
procedure printargs is
begin
put_line( "The command is '" & command_line.command_name & "'" );
for Arg in 1..command_line.argument_count loop
put( "Argument" ) @ (Arg ) @ ( " is '" ) @
( command_line.argument(Arg) ) @ ("'");
new_line;
end loop;
end printargs;</syntaxhighlight>
 
=={{header|Standard ML}}==
 
<syntaxhighlight lang="sml">print ("This program is named " ^ CommandLine.name () ^ ".\n");
val args = CommandLine.arguments ();
Array.appi
(fn (i, x) => print ("the argument #" ^ Int.toString (i+1) ^ " is " ^ x ^ "\n"))
(Array.fromList args);</syntaxhighlight>
 
=={{header|Swift}}==
 
<syntaxhighlight lang="swift">let args = Process.arguments
println("This program is named \(args[0]).")
println("There are \(args.count-1) arguments.")
for i in 1..<args.count {
println("the argument #\(i) is \(args[i])")
}</syntaxhighlight>
 
Alternately:
 
{{works with|Swift|1.2+}}
<syntaxhighlight lang="swift">println("This program is named \(String.fromCString(Process.unsafeArgv[0])!).")
println("There are \(Process.argc-1) arguments.")
for i in 1 ..< Int(Process.argc) {
println("the argument #\(i) is \(String.fromCString(Process.unsafeArgv[i])!)")
}</syntaxhighlight>
{{works with|Swift|1.0-1.1}}
<syntaxhighlight lang="swift">println("This program is named \(String.fromCString(C_ARGV[0])!).")
println("There are \(C_ARGC-1) arguments.")
for i in 1 ..< Int(C_ARGC) {
println("the argument #\(i) is \(String.fromCString(C_ARGV[i])!)")
}</syntaxhighlight>
 
=={{header|Tailspin}}==
<syntaxhighlight lang="tailspin">
$ARGS -> !OUT::write
</syntaxhighlight>
{{out}}
<pre>
[-c, alpha beta, -h, gamma]
</pre>
 
=={{header|Tcl}}==
 
The predefined global variable <tt>argc</tt> contains the number of arguments passed to the program after the script being executed, <tt>argv</tt> contains those arguments as a list. (The name of the script is in the <tt>argv0</tt> global variable, and the name of the executable interpreter itself is returned by the command <code>info nameofexecutable</code>.) Retrieving the second argument might look something like this:
 
<syntaxhighlight lang="tcl">if { $argc > 1 } {
puts [lindex $argv 1]
}</syntaxhighlight>
 
(Tcl counts from zero, thus <tt>[lindex $list 1]</tt> retrieves the second item in the list)
 
==[[{{header|Toka]]}}==
[[Category:Toka]]
 
Arguments are stored into an array. The first element in the array
Line 181 ⟶ 2,665:
number of arguments is provided by #args.
 
<syntaxhighlight lang="toka">[ arglist array.get type cr ] is show-arg
[ dup . char: = emit space ] is #=
1 #args [ i #= show-arg ] countedLoop</syntaxhighlight>
 
 
==[[UNIX Shell]]==
=={{header|Transd}}==
<syntaxhighlight lang="Scheme">#lang transd
 
MainModule: {
_start: (λ
(textout "Number of arguments: " @numArgs
"\nArgument list: " @progArgs)
)
}</syntaxhighlight>
 
Linux command ('tree3' is the name of Transd interpreter):
tree3 progname -c "alpha beta" -h "gamma"
{{out}}
<pre>
Number of arguments: 5
Argument list: ["progname", "-c", "alpha beta", "-h", "gamma"]
</pre>
 
=={{header|TXR}}==
 
Command line arguments in TXR's pattern-based extraction language can be treated as the lines of a text stream, which is arranged using the directive <code>@(next :args)</code>. Thus TXR's text parsing capabilities work over the argument list.
 
This <code>@(next :args)</code> should be written as the first line of the TXR program, because TXR otherwise interprets the first argument as the name of an input file to open.
 
<syntaxhighlight lang="txr">@(next :args)
@(collect)
@arg
@(end)
@(output)
My args are: {@(rep)@arg, @(last)@arg@(end)}
@(end)</syntaxhighlight>
 
<pre>$ ./txr args.txr
My args are: {}
$ ./txr args.txr 1
My args are: {1}
$ ./txr args.txr 1 2 3
My args are: {1, 2, 3}</pre>
 
Arguments are also available via two predefined variables: <code>*full-args*</code> and <code>*args*</code>, which are lists of strings, such that <code>*args*</code> is a suffic of <code>*full-args*</code>. <code>*full-args*</code> includes the arguments that were processed by TXR itself; <code>*args*</code> omits them.
 
Here is an example program which requires exactly three arguments. Note how <code>ldiff</code> is used to compute the arguments that are processed by TXR (the interpreter name, any special arguments and script name), to print an accurate usage message.
 
<syntaxhighlight lang="txrlisp">(tree-case *args*
((a b c) (put-line "got three args, thanks!"))
(else (put-line `usage: @(ldiff *full-args* *args*) <arg1> <arg2> <arg3>`)))</syntaxhighlight>
{{out}}
<pre>$ txr command-line-args.txr 1 2
usage: txr command-line-args.txr <arg1> <arg2> <arg3>
$ txr command-line-args.txr 1 2 3 4
usage: txr command-line-args.txr <arg1> <arg2> <arg3>
$ txr command-line-args.txr 1 2 3
got three args, thanks!</pre>
 
=={{header|UNIX Shell}}==
===[[Bourne Shell]]===
To retrieve the entire list of arguments:
<syntaxhighlight lang="bash">WHOLELIST="$@"</syntaxhighlight>
To retrieve the second and fifth arguments:
<syntaxhighlight lang="bash">SECOND=$2
FIFTH=$5</syntaxhighlight>
 
=={{header|Ursa}}==
In Ursa, all command line arguments (including the program name as invoked) are contained in the string stream args.
<syntaxhighlight lang="ursa">#
# command-line arguments
#
 
# output all arguments
for (decl int i) (< i (size args)) (inc i)
out args<i> endl console
end for</syntaxhighlight>
 
Sample shell session in the Bourne shell:
<pre>$ ursa cmdlineargs.u "alpha beta" -h "gamma"
cmdlineargs.u
alpha beta
-h
gamma
$</pre>
 
=={{header|Ursala}}==
Command line arguments are accessible to an application
through a data structure initialized by the run-time system.
This example application does nothing but display the data
structure on standard output.
<syntaxhighlight lang="ursala">#import std
 
#executable ('parameterized','')
 
clarg = <.file$[contents: --<''>+ _option%LP]>+ ~command.options</syntaxhighlight>
Here is a bash terminal session.
<pre>$ clarg -c alpha,beta -h gamma --foo=bar,baz
<
option[
keyword: 'c',
parameters: <'alpha','beta'>],
option[
position: 1,
keyword: 'h',
parameters: <'gamma'>],
option[
position: 2,
longform: true,
keyword: 'foo',
parameters: <'bar','baz'>]></pre>
 
=={{header|V}}==
The arguments to the program are stored in the stack,
 
args.v
<syntaxhighlight lang="v">$stack puts
 
./args.v a b c
=[args.v a b c]</syntaxhighlight>
 
=={{header|vbScript}}==
 
<syntaxhighlight lang="vbscript">
'Command line arguments can be accessed all together by
 
For Each arg In Wscript.Arguments
Wscript.Echo "arg=", arg
Next
 
'You can access only the named arguments such as /arg:value
 
For Each arg In Wscript.Arguments.Named
Wscript.Echo "name=", arg, "value=", Wscript.Arguments.Named(arg)
Next
 
'Or just the unnamed arguments
 
For Each arg In Wscript.Arguments.Unnamed
Wscript.Echo "arg=", arg
Next
</syntaxhighlight>
 
=={{header|Visual Basic}}==
 
Like [[#BASIC|Qbasic]], Visual Basic returns all of the args in the built-in variable <code>Command$</code>:
<syntaxhighlight lang="vb">Sub Main
MsgBox Command$
End Sub</syntaxhighlight>
 
=={{header|Visual Basic .NET}}==
 
This syntax will tokenize the command line arguments. Tokens are normally delimited by spaces, but spaces can be part of a token if surrounded by quotes.
 
<syntaxhighlight lang="vbnet">Sub Main(ByVal args As String())
For Each token In args
Console.WriteLine(token)
Next
End Sub</syntaxhighlight>
 
=={{header|V (Vlang)}}==
This assumes that the following script, myscript.v, is run as follows:
$ v run myscript.v -c "alpha beta" -h "gamma"
<syntaxhighlight lang="v (vlang)">import os
fn main() {
for i, x in os.args[1..] {
println("the argument #$i is $x")
}
}</syntaxhighlight>
 
{{out}}
<pre>
the argument #0 is -c
the argument #1 is alpha beta
the argument #2 is -h
the argument #3 is gamma
</pre>
 
=={{header|Wren}}==
This assumes that the following script, Command-line_arguments.wren, is run as follows:
$ wren Command-line_arguments.wren -c "alpha beta" -h "gamma"
<syntaxhighlight lang="wren">import "os" for Process
 
System.print(Process.arguments)</syntaxhighlight>
 
{{out}}
<pre>
[-c, alpha beta, -h, gamma]
</pre>
 
=={{header|XPL0}}==
Characters following the program name are copied into a buffer that is accessible as device 8.
This displays the example command line with quote marks stripped off.
<syntaxhighlight lang="xpl0">int C;
[loop [C:= ChIn(8);
if C = \EOF\$1A then quit;
ChOut(0, C);
];
CrLf(0);
]</syntaxhighlight>
 
{{out}}
<pre>
-c alpha beta -h gamma
</pre>
 
=={{header|zkl}}==
File myprogram.zkl:
<syntaxhighlight lang="zkl">System.argv.println();
vm.arglist.println();</syntaxhighlight>
zkl myprogram -c "alpha beta" -h "gamma"
{{out}}
<pre>
L("/home/craigd/Projects/ZKL/Bin/zkl","myprogram","-c","alpha beta","-h","gamma")
L("-c","alpha beta","-h","gamma")
</pre>
 
{{omit from|Axe}}
{{omit from|Brainf***}}
{{omit from|bc}}
{{omit from|Commodore BASIC}}
{{omit from|dc}}
{{omit from|EasyLang|Programs cannot be executed outside the IDE, not even in a command line.}}
{{omit from|GUISS}}
{{omit from|Lotus 123 Macro Scripting}}
{{omit from|M4}}
{{omit from|Maxima}}
{{omit From|Metafont}}
{{omit from|MUMPS|Maybe some version can do this. DSM on VMS and Cache on neither VMS nor Windows can.}}
{{omit from|PARI/GP}}
{{omit from|PostScript}}
{{omit from|Retro}}
{{omit from|SmileBASIC|No command line.}}
{{omit from|TI-89 BASIC|Does not have an external OS/command processor.}}
{{omit from|Vim Script}}
{{omit from|ZX Spectrum Basic|There are no command line parameters.}}
45

edits