Here's a tough one:

Make a program that print's it own source code.
 

DOH edit expired...

Rules: The program must have at least 1 real(not whitespace) line of source code

meisawesome:

Here's a tough one:

Make a program that print's it own source code.
 

In Brainfuck:

->+>+++>>+>++>+>+++>>+>++>>>+>+>+>++>+>>>>+++>+>>++>+>+++>>++>

++>>+>>+>++>++>+>>>>+++>+>>>>++>++>>>>+>>++>+>+++>>>++>>+++++

+>>+>>++>+>>>>+++>>+++++>>+>+++>>>++>>++>>+>>++>+>+++>>>++>>+

++++++++++++>>+>>++>+>+++>+>+++>>>++>>++++>>+>>++>+>>>>+++>>+

++++>>>>++>>>>+>+>++>>+++>+>>>>+++>+>>>>+++>+>>>>+++>>++>++>+

>+++>+>++>++>>>>>>++>+>+++>>>>>+++>>>++>+>+++>+>+>++>>>>>>++>

>>+>>>++>+>>>>+++>+>>>+>>++>+>++++++++++++++++++>>>>+>+>>>+>>

++>+>+++>>>++>>++++++++>>+>>++>+>>>>+++>>++++++>>>+>++>>+++>+

>+>++>+>+++>>>>>+++>>>+>+>>++>+>+++>>>++>>++++++++>>+>>++>+>>

>>+++>>++++>>+>+++>>>>>>++>+>+++>>+>++>>>>+>+>++>+>>>>+++>>++

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

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

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

Note that whitespace is not preserved, but is semantically meaningless.  With apologies to Brian Raiter.

In Perl:

#! /usr/bin/perl

use strict;
use warnings;

use LWP::UserAgent;
use HTML::PullParser;

my $ua = new LWP::UserAgent;

my $req = new HTTP::Request(GET => "http://forums.worsethanfailure.com/forums/permalink/114219/114219/ShowThread.aspx");
my $resp = $ua->request($req);

my $p = new HTML::PullParser(doc => $resp->decoded_content,
                             report_tags => [qw/a pre br/],
                             start => '"S", tagname, attr',
                             end => '"E", tagname',
                             text => '"T", dtext',
                             unbroken_text => 1,
                             );

while (my $token = $p->get_token()) {
  last if $token->[0] eq "S" && $token->[1] eq "a" && ($token->[2]->{name} || "") eq "114219";
}

while (my $token = $p->get_token()) {
  last if $token->[0] eq "S" && $token->[1] eq "pre";
}

my $code = "";
while (my $token = $p->get_token()) {
  last if $token->[0] eq "E" && $token->[1] eq "pre";
  $code .= "\n" if $token->[0] eq "S" && $token->[1] eq "br";
  $code .= $token->[1] if $token->[0] eq "T";
}

print($code);

;-)

iwpg:

In Perl:

#! /usr/bin/perl

use strict;
use warnings;

use LWP::UserAgent;
use HTML::PullParser;

my $ua = new LWP::UserAgent;

my $req = new HTTP::Request(GET => "http://forums.worsethanfailure.com/forums/permalink/114219/114219/ShowThread.aspx");
my $resp = $ua->request($req);

my $p = new HTML::PullParser(doc => $resp->decoded_content,
                             report_tags => [qw/a pre br/],
                             start => '"S", tagname, attr',
                             end => '"E", tagname',
                             text => '"T", dtext',
                             unbroken_text => 1,
                             );

while (my $token = $p->get_token()) {
  last if $token->[0] eq "S" && $token->[1] eq "a" && ($token->[2]->{name} || "") eq "114219";
}

while (my $token = $p->get_token()) {
  last if $token->[0] eq "S" && $token->[1] eq "pre";
}

my $code = "";
while (my $token = $p->get_token()) {
  last if $token->[0] eq "E" && $token->[1] eq "pre";
  $code .= "\n" if $token->[0] eq "S" && $token->[1] eq "br";
  $code .= $token->[1] if $token->[0] eq "T";
}

print($code);

;-)

HAHAHA that's fucked up.

you're gunna get a FILE-NOT-FOUND error soon though, ain't ya?

This stopped being difficult when languages gained self-referential capabilities
In Perl:

seek DATA,0,0;print while <DATA>
__END__

meisawesome:

Here's a tough one:

Make a program that print's it own source code. 

int main(int argc, char *argv[]) { char *s="int main(int argc, char *argv[]) { char *s=%c%s%c; printf(s,34,s,34,10); return 0; }%c"; printf(s,34,s,34,10); return 0; }

 

 

Disclaimer: I've written this little program all by myself, without peeking into existing solutions (but knowing the basic concept). Anyway, due to the nature of this challenge, it's likely that a very similar or even identical solution already exists somewhere on the web.
 

 

 

Now to something less boring than C: the self-replicating SQL*PLUS script

set feedback off
set head off
set linesize 32767
create or replace function selfrep return varchar2 is
s varchar2(2000) := 'set feedback offAset head offAset linesize 32767Acreate or replace function selfrep return varchar2 isAs varchar2(2000) := BCB;AbeginAreturn replace(replace(replace(s,chr(65),chr(10)),chr(66),chr(39)),chr(67),s);Aend;A/Aselect selfrep from dual;A';
begin
return replace(replace(replace(s,chr(65),chr(10)),chr(66),chr(39)),chr(67),s);
end;
/
select selfrep from dual;

I have heard, that level 2 version of this problem is to write source code for program A, which outputs source code for program B, which in turn outputs source code for program A.

I am too stupid to solve these by myself, but it would be interesting to see if this could be solved by using CodeDom in .Net.
 

sirhegel:

I have heard, that level 2 version of this problem is to write source code for program A, which outputs source code for program B, which in turn outputs source code for program A.

I am too stupid to solve these by myself, but it would be interesting to see if this could be solved by using CodeDom in .Net.

Here we go, Level 2... the only difference between program A and program B is one character in a comment.. better than nothing, I guess...It's a modified version of the SQL*PLUS script.

 

 

set feedback off
set head off
set linesize 32767
create or replace function selfrep return varchar2 is
s varchar2(2000) := 'set feedback offAset head offAset linesize 32767Acreate or replace function selfrep return varchar2 isAs varchar2(2000) := BCB;AbeginA-- DAreturn replace(replace(replace(replace(replace(replace(replace(s,chr(65),chr(10)),chr(66),chr(39)),chr(67),s),chr(68),chr(70)),chr(69),chr(71)),chr(70),chr(69)),chr(71),chr(68));Aend;A/Aselect selfrep from dual;A';
begin
-- D
return replace(replace(replace(replace(replace(replace(replace(s,chr(65),chr(10)),chr(66),chr(39)),chr(67),s),chr(68),chr(70)),chr(69),chr(71)),chr(70),chr(69)),chr(71),chr(68));
end;
/
select selfrep from dual;

in php:

<?php 

$self = substr($_SERVER['SCRIPT_NAME'], 1);
highlight_string(file_get_contents($self));

?>

Who can beat 21 characters?

<?readfile(__FILE__);

Pap:

Who can beat 21 characters?

How about: 

Rules: The program must have at least 1 real(not whitespace) line of source code

I'm more interested in self-replicating WTFs... you know, the ones where the comments are more retarded than the original story.  Anyone have any favorites of these they would care to link to?

Pap:

Rules: The program must have at least 1 real(not whitespace) line of source code

Bah, how boring. 

Anyone remember the Wang 2200?  The shortest self-reproducing program for it was:

 ^Error 42

 

If you typed that in, the response would be:

^Error 42

 

iwpg:

Pap:

Who can beat 21 characters?

How about: 

Seriously though, Unix shell:

cat $0

Python:

import sys
print file(sys.argv[0]).read()

This is a no-brainer in JavaScript:

  function program() {
    alert(program);
    return "Brillant";
  }


How about a program that opens its source in word, prints the screenshot that lands on a wooden table, triggers a shot by a digital camera and then exports the picture to a website ?

In Powershell:

function TypeMe() { Get-Content "Function:\TypeMe" }

Found this one on the internets:

http://p-nand-q.com/python/scripts/quine6.txt

 

windows batch file...

rem

 

sirhegel:

I have heard, that level 2 version of this problem is to write source code for program A, which outputs source code for program B, which in turn outputs source code for program A.

#include <stdio.h>

int main()
{
    const int id = 0;
    char *prog = "#include <stdio.h>%c%cint main()%c{%c    const int id = %d;%c    char *prog = %c%s%c;%c    printf(prog, 10, 10, 10, 10, id^1, 10, 34, prog, 34, 10, 10, 10);%c    return 0;%c}";
    printf(prog, 10, 10, 10, 10, id^1, 10, 34, prog, 34, 10, 10, 10);
    return 0;
}

Python (without using read...):

s="s=%c%s%c%cprint(s%c(34,s,34,10,37))"
print(s%(34,s,34,10,37))

 

Hehe. One year a zero-length 'quine' one the IOCCC prize for "Worst abuse of the rules".

In English

"Repeat exactly what I just said"

 

I'd really like to see a quine written *in* Whitespace.

<%@ Page Language="C#" %>
<script runat="server">
protected void Page_Init(object sender, EventArgs e){
    Response.Write(System.IO.File.ReadAllText(Request.PhysicalPath));
}
</script>

Atari 8K BASIC:

10 LIST

Heh.

 

Bash :)

echo $BASH_COMMAND

CREATE PROC SP_ME
AS
SELECT ROUTINE_DEFINITION
    FROM INFORMATION_SCHEMA.ROUTINES
    WHERE ROUTINE_NAME = 'SP_ME'
    AND ROUTINE_TYPE='PROCEDURE'

Rexx

/* */ say sourceline(1)

 

#!/bin/bash
x=20; d='bupr
rabq
"}1:d:q{$" a- bupr
bq ;))--d;0=>d;2-k=d((ebs
"740\" r- bupr
31gbe | "q$" a- bupr
"740\" ra- bupr
rabq
"}1:d:q{$" a- bupr
bq ;))--d;k=>d;1-}q#{$=d((ebs
]k-y[$=k ;}q#{$=y
`31gbe | "q$" bupr`=q
=q ;02=k
ufno/avo/!#'
d=`echo "$d" | rot13`
l=${#d}; x=$[l-x]
for((q=${#d}-1;q>=x;q--)); do
  echo -n "${d:q:1}"
done
echo -ne "\047"
echo -n "$d" | rot13
echo -e "\047"
for((q=x-2;q>=0;q--)); do
  echo -n "${d:q:1}"
done
echo

#!/bin/cat

<?php echo implode("\n", file(__FILE__)); ?>

Because someone had to do it

(My apologies for the length)

<?xml version="1.0" encoding="UTF-8" ?>
<?xml-stylesheet type="text/xsl" href="http://forums.worsethanfailure.com/forums/AddPost.aspx?PostID=119432#" ?>
<s:stylesheet version="1.0" xmlns:s="http://www.w3.org/1999/XSL/Transform">
    <s:output method="html" />
    <s:strip-space elements="*" />
    <s:preserve-space elements="s:text" />
    <s:template name="line">
        <br />
        <s:for-each select="ancestor::*">
            <s:text><![CDATA[    ]]></s:text>
        </s:for-each>
    </s:template>
    <s:template name="tagHead">
        <s:text><![CDATA[<]]></s:text>
        <span class="tagName">
            <s:value-of select="name()" />
        </span>
        <s:for-each select="./@*">
            <s:text><![CDATA[ ]]></s:text>
            <span class="attribName">
                <s:value-of select="name()" />
            </span>
            <s:text><![CDATA[="]]></s:text>
            <span class="attribValue">
                <s:value-of select="." />
            </span>
            <s:text><![CDATA["]]></s:text>
        </s:for-each>
    </s:template>
    <s:template name="tagTail">
        <s:text><![CDATA[</]]></s:text>
            <span class="tagName">
                <s:value-of select="name()" />
            </span>
        <s:text><![CDATA[>]]></s:text>
    </s:template>
    <s:template match="//*">
        <s:call-template name="line" />
        <s:call-template name="tagHead" />
        <s:text><![CDATA[ />]]></s:text>
    </s:template>
    <s:template match="//*[child::*]">
        <s:call-template name="line" />
        <s:call-template name="tagHead" />
        <s:text><![CDATA[>]]></s:text>
        <s:apply-templates />
        <s:call-template name="line" />
        <s:call-template name="tagTail" />
    </s:template>
    <s:template match="//s:text|//style|//title|//span[@class='system']">
        <s:call-template name="line" />
        <s:call-template name="tagHead" />
        <s:text><![CDATA[>]]></s:text>
        <span class="cdataMarkup">
            <s:text><![CDATA[<![CDATA[]]></s:text>
        </span>
        <span class="textValue">
            <s:value-of select="text()" />
        </span>
        <span class="cdataMarkup">
            <s:text><![CDATA[]]]]></s:text>
            <s:text><![CDATA[>]]></s:text>
        </span>
        <s:call-template name="tagTail" />
    </s:template>
    <s:template match="/s:stylesheet">
        <html>
            <head>
                <title><![CDATA[Quine!!!]]></title>
                <style><![CDATA[
                    pre {
                        background-color:rgb(255,255,201);
                        border-color:rgb(255,255,0);
                        border-width:3px;
                        border-style:solid;
                        padding:3px;
                    }
                    .tagName        {color:DarkBlue;font-weight:bold;}
                    .attribName     {color:Blue;}
                    .attribValue    {color:DarkGoldenRod;font-style:italic;}
                    .cdataMarkup    {color:DarkMagenta;}
                    .textValue      {color:DarkOliveGreen;font-style:italic;}
                    .system         {color:BlueViolet;}
                ]]></style>
            </head>
            <body>
                <H1>
                    <s:value-of select="//title" />
                </H1>
                <code>
                    <pre>
                        <span class="system"><![CDATA[<?xml version="1.0" encoding="UTF-8" ?>]]></span>
                        <br />
                        <span class="system"><![CDATA[<?xml-stylesheet type="text/xsl" href="http://forums.worsethanfailure.com/forums/AddPost.aspx?PostID=119432#" ?>]]></span>
                        <br />
                        <s:for-each select="/s:stylesheet">
                            <s:call-template name="tagHead" />
                        </s:for-each>
                        <span class="system"><![CDATA[ xmlns:s]]></span>    
                        <s:text><![CDATA[="]]></s:text>
                        <span class="attribValue">
                            <s:text><![CDATA[http://www.w3.org/1999/XSL/Transform]]></s:text>
                        </span>
                        <s:text><![CDATA[">]]></s:text>
                        <s:apply-templates />
                        <br />
                        <s:for-each select="/s:stylesheet">
                            <s:call-template name="tagTail" />
                        </s:for-each>
                    </pre>
                </code>
            </body>
        </html>
    </s:template>
</s:stylesheet>

(Apparently the WTF forum tries to be "smart" and messes up my stylesheet links... those href="http://forums.worsethanfailure.com/..."s are supposed to be just "#".)

An interesting concept for all languages, here in PHP:

 

 <?php
<b>Parse error</b>:  syntax error, unexpected '<' in <b>/opt/lampp/htdocs/selfrep.php</b> on line <b>2</b>
?>

 

 
Of course, this is if your php files are in /opt/lampp/htdocs/, and the file is called "selfrep.php", but it works :P

 

For bash, ksh and probably others: z='echo "$x$z$y\"$x\" ; y=\"$y\" ; $z"' ; x="z='" ; y="' ; x=" ; echo "$x$z$y\"$x\" ; y=\"$y\" ; $z"

sirhegel:

I have heard, that level 2 version of this problem is to write source code for program A, which outputs source code for program B, which in turn outputs source code for program A.

I am too stupid to solve these by myself, but it would be interesting to see if this could be solved by using CodeDom in .Net.
 

 

A recent (a couple years ago or so) back, the IOCCC had such an entry. It started out formatted as a Japanese character. Compile, run, outputs source code in the form of another Japanese character. Compile/run, you get a third, and so on for at least 6 times, at which point, you ended up with the beginning character.
 

[i did the #!/bin/cat, then noticed someone else had after the delete time limit ran out]

 

dillybar1:

windows batch file...

rem

 

 No, that doesn't print out its source code.
 

gremlin:

 No, that doesn't print out its source code.
 

Sure it does. The only time it doesn't is if ECHO OFF was used before running it. Sample output of "TESTBAT.BAT" that only contains the statement "REM":


C:\>testbat

C:\>REM

C:\>echo off
testbat
echo on

C:\>

This is called a quine and it has been done a lot already. I have made one quine program in MS-DOS: @echo off %1 %2 call %0 goto e %% call %0 goto e %%3 echo.%%4 echo :f goto f :e echo.%4@echo off echo.%4%31 %32 echo.%4call %30 goto e %3%3 echo.%4call %30 goto e %3%33 echo.%3%34 echo.%4echo :f echo.%4goto f echo.%4:e :f

O! It's a kimian quine!

 http://sigfpe.blogspot.com/2008/02/third-order-quine-in-three-languages.html not mine but damn, how do you come up with that :|

 

AbbydonKrafts:

gremlin:

 No, that doesn't print out its source code.
 

Sure it does. The only time it doesn't is if ECHO OFF was used before running it. Sample output of "TESTBAT.BAT" that only contains the statement "REM":


C:\>testbat

C:\>REM

C:\>echo off
testbat
echo on

C:\>

@echo off

type printme.bat

 

OK, since the thread is resurrected I decided I'll contribute as well:

puts [format [set s {puts [format [set s {%s}] $s]}] $s]

Guessing the language is left as an exercise to the reader 9=]. (Note, to make it perfect, the file must be terminated by a newline.)

Spectre:

puts [format [set s {puts [format [set s {%s}] $s]}] $s]

Guessing the language is left as an exercise to the reader

That's expect, you pervert. Why don't you write something in COBOL while you're at it?

asuffield:

Spectre:

puts [format [set s {puts [format [set s {%s}] $s]}] $s]

Guessing the language is left as an exercise to the reader

That's expect, you pervert. Why don't you write something in COBOL while you're at it?

Tcl actually, but I guess Expect would run it too.