Parse command-line arguments: Difference between revisions
Content deleted Content added
m →{{header|Wren}}: Changed to Wren S/H |
|||
(8 intermediate revisions by 7 users not shown) | |||
Line 8:
Many languages provide a library (getopt or GetOpt) to parse the raw command line options in an intelligent way.
<br><br>
=={{header|Ada}}==
<syntaxhighlight lang="ada">
-- Show command-line arguments
-- J. Carter 2023 Apr
-- The task is called "Parse command-line arguments", but as parsing requires attaching meaning to arguments, and the task
-- specification does not do so, showing them is all we can reasonably do
with Ada.Command_Line;
with Ada.Text_IO;
procedure Show_Args is
-- Empty
begin -- Show_Args
All_Args : for Arg in 1 .. Ada.Command_Line.Argument_Count loop
Ada.Text_IO.Put_Line (Item => Arg'Image & ": " & Ada.Command_Line.Argument (Arg) );
end loop All_Args;
end Show_Args;
</syntaxhighlight>
{{out}}
<pre>
$ ./show_args nc -v -n -z -w 1 192.168.1.2 1-1000
1: nc
2: -v
3: -n
4: -z
5: -w
6: 1
7: 192.168.1.2
8: 1-1000
</pre>
=={{header|Arturo}}==
<syntaxhighlight lang="arturo">loop.with:'i arg 'a ->
print ["argument" (to :string i)++":" a]
loop args [k,v]->
if k <> "values" ->
print ["found option:" k "with value:" v "(of type:" (to :string type v)++")"]</syntaxhighlight>
'''Sample input:'''
<pre>arturo parse\ command-line\ arguments.art one two --file:four --verbose --loop:3 three</pre>
'''Sample output:'''
<pre>argument 0: one
argument 1: two
argument 2: --file:four
argument 3: --verbose
argument 4: --loop:3
argument 5: three
found option: file with value: four (of type: string)
found option: verbose with value: true (of type: logical)
found option: loop with value: 3 (of type: integer)</pre>
=={{header|AutoHotkey}}==
For AutoHotkey v1.1+
<
if 0 > 0
{
Line 41 ⟶ 98:
}
MsgBox % "Parsed Arguments :`n" msg</
'''Output (MsgBox):'''
<pre>Parsed Arguments :
Line 52 ⟶ 109:
=={{header|AWK}}==
{{works with|gawk}}
<
# -E instead of -f so program arguments don't conflict with Gawk arguments
@include "getopt.awk"
Line 70 ⟶ 127:
if(tval) print "-t = " tval
if(uval) print "-u = " uval
}</
=={{header|Bracmat}}==
Line 80 ⟶ 137:
=={{header|C}}==
The man page for getopt (man 3 getopt) provides better option handling with examples. But if you just want to parse one argument... (adapted from simple database task):
<
int main(int argc, char **argv){
int i;
Line 109 ⟶ 166:
}
return 0;
}</
=={{header|Clojure}}==
Line 116 ⟶ 173:
=={{header|D}}==
The [http://dlang.org/phobos/std_getopt.html getopt module] in D's standard library is inspired by Perl's Getopt::Long module. The syntax of Phobos getopt infers the expected parameter types from the static types of the passed-in pointers.
<
void main(string[] args) {
Line 134 ⟶ 191:
writeln("verbose: ", verbose);
writeln("color: ", color);
}</
{{out|Usage example}}
<pre>C:\getopt_test --verbose --length 12
Line 143 ⟶ 200:
=={{header|Delphi}}==
{{libheader| System.SysUtils}}
<syntaxhighlight lang="delphi">
program Parse_command_line_argument;
Line 179 ⟶ 236:
Readln;
end.</
{{out}}
Parse_command_line_argument.exe
Line 198 ⟶ 255:
Elixir provides an option parser in a library module called <tt>OptionParser</tt>.
<
IO.puts 'Arguments:'
IO.inspect OptionParser.parse(System.argv())</
<
Arguments:
{[a: true, b: true, c: "yes", no_flag: true, verbose: true],
["apple", "banana"], [{"-V", nil}, {"-a", "1"}, {"-b", "t"}]}</
=={{header|FreeBASIC}}==
<
' Program (commandline.exe) invoked like this:
Line 225 ⟶ 282:
Print
Print "Press any key to quit"
Sleep</
{{out}}
Line 244 ⟶ 301:
=={{header|Go}}==
Most simply, implementing the suggested example from the talk page:
<
import (
Line 259 ⟶ 316:
fmt.Println("s:", *s)
fmt.Println("n:", *n)
}</
Example runs:
<pre>
Line 281 ⟶ 338:
The Icon Programming Library provides a procedure for processing command line options. See the library reference for detailed documentation. The code below is an example.
<
procedure main(ARGLIST)
Line 294 ⟶ 351:
i := opttable(i) # assign an integer
...
end</
{{libheader|Icon Programming Library}}
Line 309 ⟶ 366:
:Test if an argument is present:
::<
::This is true if the argument is present and false, if it is not.
Line 315 ⟶ 372:
:Or, find the name of an optional file:
::<
::This is the name of the first file named after the first -f argument, or empty if there was no such file.
Line 321 ⟶ 378:
Other concepts are also possible...
=={{header|jq}}==
{{works with|jq}}
'''Also works with gojq, the Go implementation of jq'''
The jq and gojq programs parse some command-line options and arguments for their own purposes
but also provide two mechanisms allowing for arbitrarily many command-line arguments to be provided to the program:
* the --args option can be used to provide a sequence of shell strings that are converted to JSON strings;
* the --jsonargs option can similarly be used to specify a sequence of JSON values.
For example, assuming a bash or bash-like shell, the invocation
<pre>
jq -n '$ARGS' --args 1 two '[3, "four"]'
</pre>
results in:
<pre>
{
"positional": [
"1",
"two",
"[3, \"four\"]"
],
"named": {}
}
</pre>
whereas:
<pre>
jq -n '$ARGS' --jsonargs 1 '"two"' '[3, "four"]'
</pre>
results in:
<pre>
{
"positional": [
1,
"two",
[
3,
"four"
]
],
"named": {}
}
</pre>
Notice that in the first case, the token `two` has not been quoted, whereas in the second case, it must be presented as `'"two"'`
if it is to be understood as a JSON string.
=={{header|Julia}}==
{{works with|Julia|0.6}}
Line 326 ⟶ 429:
Example taken from the official documentation of [https://carlobaldassi.github.io/ArgParse.jl/stable/ ArgParse docs].
<
function parse_commandline()
Line 357 ⟶ 460:
end
main()</
=={{header|Kotlin}}==
<
fun main(args: Array<String>) = println(args.asList())</
{{out}}
Line 372 ⟶ 475:
=={{header|Mathematica}}/{{header|Wolfram Language}}==
The command line is parsed and stored into a list of strings to ease manual handling by list processing functions.
<
-> {math, -v, -n, -z, -w, 1, 192.168.1.2, 1-1000}</
=={{header|Nim}}==
<
import parseopt
Line 409 ⟶ 512:
main()
</syntaxhighlight>
Sample command line:
Line 441 ⟶ 544:
Unknown option: x
Got arg 3: "1-1000"
</pre>
=={{header|Nu}}==
Parsing can be done through the <code>main</code> function's signature.
<syntaxhighlight lang="nu">
def main [
input: string # File to operate on
output?: string # File to write to
--verbose (-v): # Be verbose
] {
{
Input: $input
Output: $output
Verbose: $verbose
}
}
</syntaxhighlight>
{{out}}
<pre>
~> nu cli.nu input.txt --verbose
╭─────────┬───────────╮
│ Input │ input.txt │
│ Output │ │
│ Verbose │ true │
╰─────────┴───────────╯
~> nu cli.nu input.txt output.txt
╭─────────┬────────────╮
│ Input │ input.txt │
│ Output │ output.txt │
│ Verbose │ false │
╰─────────┴────────────╯
~> nu cli.nu --invalid
Error: nu::parser::unknown_flag
× The `main` command doesn't have flag `invalid`.
╭─[<commandline>:1:1]
1 │ main --invalid
· ────┬────
· ╰── unknown flag
╰────
help: Available flags: --verbose(-v), --help(-h). Use `--help` for more information.
~> nu cli.nu --help
Usage:
> cli.nu {flags} <input> (output)
Flags:
-v, --verbose - Be verbose
-h, --help - Display the help message for this command
Parameters:
input <string>: File to operate on
output <string>: File to write to (optional)
Input/output types:
╭───┬───────┬────────╮
│ # │ input │ output │
├───┼───────┼────────┤
│ 0 │ any │ any │
╰───┴───────┴────────╯
</pre>
=={{header|PARI/GP}}==
GP exists in a REPL and so it doesn't make sense to parse command-line arguments. But PARI can parse them just like [[#C|C]]:
<
#include <stdio.h>
Line 452 ⟶ 615:
pari_printf("8 + 1 = %Ps\n", addii(int2u(3), gen_1));
return 0;
}</
=={{header|Perl}}==
Line 458 ⟶ 621:
Use the <tt>Getopt::Long</tt> module:
<
use strict;
Line 477 ⟶ 640:
($verbose ? "Verbosity" : "No verbosity"),
" and a length of $length.\n";
</syntaxhighlight>
The output from it is:
Line 495 ⟶ 658:
{{libheader|Phix/basics}}
<!--<
<span style="color: #008080;">with</span> <span style="color: #008080;">javascript_semantics</span>
<span style="color: #004080;">sequence</span> <span style="color: #000000;">res</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">command_line</span><span style="color: #0000FF;">()</span>
<span style="color: #0000FF;">?</span><span style="color: #000000;">res</span>
<!--</
{{out}}
Line 529 ⟶ 692:
<
Opts = new_map(),
process_args(ARGS,Opts),
Line 562 ⟶ 725:
process_args([Opt|As],Pos,Map) :-
Map.put(Pos,[Opt,'']),
process_args(As,Pos+1,Map).</
Note: The parameters to the program should not be one of Picat's own parameters, since they will be parsed (and consumed) by Picat before starting to run the user's program. Here are Picat's parameters:
Line 588 ⟶ 751:
=={{header|PicoLisp}}==
PicoLisp doesn't have a library to get options. Instead, the command line is parsed at startup and handled in the following way: Each command line argument is executed (interpreted) as a Lisp source file, except that if the first character is a hypen '-', then that arguments is taken as a Lisp function call (without the surrounding parentheses). For example, the command line
<
has the effect that
# The file "abc.l" is executed
Line 602 ⟶ 765:
=={{header|PowerShell}}==
Powershell functions and filters handle options organically, with advanced .NET support to handle complex and advanced options including aliases, ranges, sets, datatypes, and more. [https://msdn.microsoft.com/en-us/powershell/reference/5.0/microsoft.powershell.core/about/about_parsing See] [https://msdn.microsoft.com/powershell/reference/5.1/Microsoft.PowerShell.Core/about/about_Functions more] [https://msdn.microsoft.com/en-us/powershell/reference/5.0/microsoft.powershell.core/about/about_functions_advanced here]. However, to parse options 'classically', you can write a custom parser. A slightly messy version (inspired by ruby optparse) that can only handle switches and relies on RegEx:
<syntaxhighlight lang="powershell">
$options = @{
opt1 = [bool] 0
Line 654 ⟶ 817:
return [array]$argv,$options
}#fn
</syntaxhighlight>
Usage (in some function or script):
<syntaxhighlight lang="powershell">
$argv,$options = parseOptions $args $options
Line 664 ⟶ 827:
$bar = $argv | SomeOtherFilter | Baz
}
</syntaxhighlight>
Usage in shell:
<
$> function -c --wxx arg1 arg2
</syntaxhighlight>
Note that this works in Powershell. All of the arguments after the function name will be passed as strings to the function, which then calls them as an array with the automatic variable, $args. The custom parser/function does the work from there, turning the strings into flags and typed arguments. WARNING: This is reinventing the wheel to an extreme degree.
Line 674 ⟶ 837:
Works in SWI-Prolog.
<
main(Argv) :-
Line 718 ⟶ 881:
longflags([port]),
help('The server port.')]
]).</
{{out}}
<syntaxhighlight lang="powershell">
# no options set (use defaults)
$ swipl .\opts.pl
Line 744 ⟶ 907:
--server -s atom=www.google.com The server address.
--port -p integer=5000 The server port.
</syntaxhighlight>
=={{header|Python}}==
Version 2.3+
<syntaxhighlight lang="python">
from optparse import OptionParser
[...]
Line 763 ⟶ 926:
<yourscript> --file=outfile -q
</syntaxhighlight>
=={{header|Racket}}==
Line 769 ⟶ 932:
Racket has a good command-line parsing library, the following demonstrates some of its features:
<
#!/usr/bin/env racket
#lang racket
Line 799 ⟶ 962:
(printf "Log level: ~s\n" loglevel)
(printf "Operations: ~a\n" (string-join ops ", ")))
</syntaxhighlight>
Sample runs:
Line 834 ⟶ 997:
(formerly Perl 6)
At the end of running any top-level code (which can preprocess the arguments if it likes), Raku automatically examines any remaining arguments and transforms them into a call to a <tt>MAIN</tt> routine, if one is defined. The arguments are parsed based on the signature of the routine, so that options are mapped to named arguments.
<syntaxhighlight lang="raku"
say "Bool: $b";
say "Str: $s";
say "Num: $n";
say "Rest: @rest[]";
}</
{{out}}
<pre>$ ./main -h
Line 876 ⟶ 1,039:
╚═════════════════════════════════════════════════════════════════════════════╝
</pre>
<
parse arg opts /*this preserves the case of options. */
opts=space(opts) /*elide superfluous blanks in options. */
Line 915 ⟶ 1,078:
isInt: return datatype(arg(1), 'W') /*return 1 if argument is an integer*/
isNum: return datatype(arg(1), 'N') /*return 1 if argument is a number.*/
sayer: say; say '***error***' arg(1); exit 13</
<pre>
╔═══════════════════════════════════════════════════════════════════════╗
Line 932 ⟶ 1,095:
=== Ruby with 'getoptlong' ===
<
# == Synopsis
Line 1,013 ⟶ 1,176:
nil
end
end</
<pre>$ ./pargs.rb -h
Line 1,047 ⟶ 1,210:
=== Ruby with 'optparse' ===
<
sflag = false
Line 1,098 ⟶ 1,261:
Fruit: #{fruit}
Arguments: #{ARGV.inspect}
EOF</
<pre>$ ruby takeopts.rb -h
Line 1,130 ⟶ 1,293:
Using the [https://docs.rs/structopt StructOpt]:
<
#[derive(StructOpt)]
Line 1,147 ⟶ 1,310:
println!("s: {}", opt.s);
println!("n: {}", opt.n);
}</
Examples:
Line 1,170 ⟶ 1,333:
=={{header|Scala}}==
{{libheader|Scala}}
<
println(s"Received the following arguments: + ${args.mkString("", ", ", ".")}")
}</
=={{header|Standard ML}}==
Line 1,178 ⟶ 1,341:
{{works with|MLton}}
The following code listing can be compiled with both [[SML/NJ]] and [[MLton]]:
<
exception FatalError of string
Line 1,209 ⟶ 1,372:
(* MLton *)
val _ = Test.main (CommandLine.name(), CommandLine.arguments())</
=== SML/NJ ===
[[SML/NJ]] can compile source code to a "heap file", witch can than be executed by the interpreter with arguments given (see [http://stackoverflow.com/questions/5053149/sml-nj-how-to-compile-standalone-executable this entry on stackowerflow.com] for more information).
The <code>source.cm</code> file should lock like this:
<syntaxhighlight lang="text">Group
is
SOURCE_FILE.sml
$/basis.cm</
To compile the program, use <code>ml-build sources.cm</code>. This should create a "heap file" <code>sources.x86-linux</code>, depending on your architecture.
The heap file can be executed with <code>sml @SMLload=sources.x86-linux ARGUMENTS</code>, or the script [http://www.smlnj.org/doc/heap2exec/index.html <code>heap2exec</code>] can be used to make a single executable.
Line 1,226 ⟶ 1,389:
=={{header|Tcl}}==
The following proc detects and removes argument-less (-b) and one-argument options from the argument vector.
<
upvar 1 $_argv argv $_var var
set pos [lsearch -regexp $argv ^$name]
Line 1,238 ⟶ 1,401:
return 0
}
}</
Usage examples:
getopt argv -sep sep ";" ;# possibly override default with user preference
Line 1,248 ⟶ 1,411:
=={{header|Wren}}==
Any command line arguments passed to a Wren CLI script are collected in the same order into a list of strings. If individual arguments require any further parsing, this can then be done using normal Wren code.
<
var args = Process.arguments
Line 1,257 ⟶ 1,420:
var end = Num.fromString(sp[1])
var r = start..end
System.print("The final argument expressed as a Range object is %(r)")</
{{out}}
<pre>
$ wren_cli
The arguments passed are: [-v, -n, -z, -w, 1, 192.168.1.2, 1-1000]
The final argument expressed as a Range object is 1..1000
</pre>
=={{header|XPL0}}==
<pre>parseargs nc -v -n -z -w 1 192.168.1.2 1-1000</pre>
<syntaxhighlight lang "XPL0">int N, C;
[N:= 0;
loop [C:= ChIn(8);
if C = $0D \CR\ then quit;
N:= N+1;
CrLf(0); IntOut(0, N); Text(0, ": ");
repeat ChOut(0, C);
C:= ChIn(8);
until C = $20 \space\;
];
CrLf(0);
]</syntaxhighlight>
{{out}}
<pre>
1: nc
2: -v
3: -n
4: -z
5: -w
6: 1
7: 192.168.1.2
8: 1-1000
</pre>
Line 1,271 ⟶ 1,461:
File myprogram.zkl:
<
argh := Utils.Argh(
T("v","v","print version",fcn{println("Version stub")}),
Line 1,286 ⟶ 1,476:
case("z") { println("zazzle") }
}
}</
<pre>zkl myprogram nc -v -n -z --ip 192.168.1.2 1-1000</pre>
{{out}}
|