18
\$\begingroup\$

Good Evening Ladies and Gentleman.

Tonight - we have a code challenge. A competition that will be defined not in the amount of points (a la Asteroids) but simply whether you can finish it (a la Zelda). And for the sake of comparison, the amount of bytes (uh... but you said...).

Your challenge is to compose a single quine that compiles in both WhiteSpace and any other language of your choice. Both quines must have absolutely identical output (which as a quine is absolutely identical to the source code). Your quine must be more then 3 bytes in length.

Your score is the size of your source code in bytes.

Given that this challenge involves WhiteSpace, please escape your answers so that they are clear - preferably in the C style. Your score is the original source in bytes - not the escaped version.

Good luck.

\$\endgroup\$
  • 1
    \$\begingroup\$ Would an xxd hex dump do? \$\endgroup\$ – Ilmari Karonen Aug 2 '13 at 12:26
  • \$\begingroup\$ Yes. I'll accept it. \$\endgroup\$ – lochok Aug 2 '13 at 12:57
19
\$\begingroup\$

Whitespace and Perl, 992 975 characters

Good Evening Ladies and Gentlemen.

To start with, here is my submission encoded in base64, which I feel is the best way to communicate a large Whitespace program. (You really don't want to use an encoding that leaves any whitespace as-is, so there's no point in selecting something more "readable".)
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Here's an excerpt that highlights all of the visible parts of the source. is used to indicate a tab and to indicate a newline.

say<< x2;say<<0,0  ↲
say<< x2;say<<0,0  ↲
↲
⇥⇥↲
   ⇥⇥  ⇥ [... etcetera ... skipping rest of a really long line ...]↲
   ⇥⇥⇥ ⇥⇥[... etcetera ... shorter but still quite a long line ...]↲
 ↲
  ↲
    ↲
 ⇥  ⇥↲
[... etcetera ... whole lotta whitespace in here ...]
⇥⇥↲
 ↲
↲
↲
⇥↲
0↲

Perl was the natural choice for the second language in this challenge, being one of the best general-purpose languages for writing compact quines. My shortest Perl quine is 19 bytes:

say<< x2
say<< x2
 

– and you can see how it was the seed for the Perl half of the double quine. My best Whitespace quine, in comparison, is 541 bytes long. (Though shorter ones do exist – 445 bytes is the best I've seen.)

From the Perl interpreter's point of view, the first line of the double quine's source file contains the two statements that make up the entire program proper, as the remaining contents are two quoted strings. The first string is the repeated line of Perl, and is delimited by the blank third line. The second string is all whitespace, and runs from the fourth line of the source all the way to the 0 delimiter at the bottom of the file.

When taken as a Whitespace program, the first four lines contain three instructions that are largely useless. (Their effect is to push two zero values onto the stack, and then discard the second one.) They're included just to get safely past the newlines that the Perl program requires – the real program starts after that. Rather than quote the unreadable source any further, here's a paraphrasing of the instructions that make up the Whitespace program, in an assembly-like format:

# Representation of "say<< ;say<<0,0  \n" in base 122 as ASCII chars.
PERLCODE = 44892457841068708924520433691075560592081

# Represention of the whitespace program, following the "push PERLCODE"
# instruction, in base 3 (see comments on wsout).
WSCODE = 9823454421986355730445143846606456399449033186160554878002671428613111806443504867738858766142050504887335990409088441824104338753030405625930185

# Set up the stack and the heap. The first three instructions are not
# particularly useful; they're just there to skip past the newlines in
# the Perl code. (Though the initial zero on the stack does get used
# at the very end.)

        push    0
        push    0
        jneg    wsout

        push    WSCODE
        push    PERLCODE
        dup
        dup
        push    0
        copy    1

# Output the first four lines of the file.

perl:   dup
        mod     122
        putchar
        div     122
        dup
        jnzero  perl
        pop
        jzero   perl
        push    68              # represents "jneg wsout"
        call    wsout

# Output the rest of the file.

        copy    1
        call    pushout
        push    2
        call    wsout
        call    pushout
        call    wsout
        putnum
        push    2
        call    wsout
        exit

# pushout: Output a Whitespace push instruction, using the number on
# the top of the stack as the instruction's argument. (Recursion is
# used to output the bits MSB-first.)

pushout:
        push    0
        dup
        call    wsout
        call    wsout
bits:   dup
        jzero   bitend
        dup
        mod     2
        swap
        div     2
        call    bits
bitend: call    wsout
        ret

# wsout: Output a sequence of whitespace characters as represented by
# the number on the top of the stack. The number is read in base 3,
# LSB-first, with 0 = SPC, 1 = TAB, 2 = NL. Calling wsout with a value
# of zero will output a single space.

wsout:
        dup
        mod     3
        mul     -23             # some ugly math that transforms
        mod     -24             # (0, 1, 2) into (32, 9, 10)
        add     32
        putchar
        div     3
        dup
        jnzero  wsout
        pop
        ret

The giant numbers at the top are what we Whitespace users have to use in lieu of actual strings. Don't bother trying to run this on a Whitespace interpreter that doesn't have proper bignum support.

Finally, here's the program again, but this time with C-style escapes, since it was specifically requested:

say<< x2;say<<0,0  \nsay<< x2;say<<0,0  \n\n\t\t\n   \t\t  \t  \t \t\t  \t  \t\t \t \t \t\t\t\t\t  \t\t \t\t\t \t\t\t\t\t \t \t \t\t \t \t   \t\t\t\t\t   \t\t  \t \t\t \t \t\t \t    \t\t\t\t \t  \t \t\t \t    \t \t\t   \t      \t\t\t \t\t\t\t\t \t\t\t \t\t\t\t     \t\t\t \t      \t\t\t\t\t      \t\t   \t\t    \t\t\t\t\t\t\t\t  \t \t \t   \t\t\t   \t\t\t\t\t  \t\t\t\t \t    \t \t\t\t\t\t \t     \t  \t \t      \t    \t     \t\t   \t\t\t \t \t\t\t\t\t\t \t\t \t \t       \t\t  \t           \t\t   \t \t\t\t \t     \t\t\t\t\t\t \t    \t\t\t\t \t\t\t\t  \t\t\t   \t\t\t \t  \t   \t\t\t\t  \t\t \t\t\t   \t\t\t \t\t\t \t\t\t  \t\t \t  \t \t\t\t \t \t \t\t     \t \t \t  \t  \t \t\t     \t \t    \t\t\t\t  \t \t\t \t \t\t \t  \t \t \t\t \t\t \t\t \t\t\t \t\t\t    \t\t\t  \t  \t\n   \t     \t\t\t\t\t \t\t \t \t \t \t  \t\t\t   \t   \t    \t\t \t         \t   \t\t\t\t \t\t    \t\t \t \t  \t\t\t \t\t\t\t \t\t\t   \t    \t \t \t\t\t  \t\t\t\t \t \t\t \t \t \t\t\t\t \t  \t \t\t \t   \t\n \n  \n    \n \t  \t\n\n  \t\n \n    \t\t\t\t \t \n\t \t\t\t\n     \t\t\t\t \t \n\t \t  \n   \t\t\n\t  \n\n\t\t\t\n \n\n\n\t \t\n   \t   \t  \n\n \t\n \t  \t\n\n \t \n   \t \n\n \t\n\n \t \n\n \t\n\t\n \t   \t \n\n \t\n\n\n\n\n   \n   \n \n \n \t\n\n \t\n\n    \n \n \n\t \n \n    \t \n\t \t\t \n\t   \t \n\t \t \n \t  \n\n  \n \n    \t\t\n\t \t\t  \t\t \t\t\t\n\t  \n  \t\t\t   \n\t \t\t   \t     \n\t   \t\n     \t\t\n\t \t  \n   \t\t\n\t  \n\n\t\t\n \n\n\n\t\n0\n
\$\endgroup\$
  • 2
    \$\begingroup\$ Wow! I thought brainfuck would be the first solution up. \$\endgroup\$ – boothby Aug 3 '13 at 5:27
  • 1
    \$\begingroup\$ I thought about it, but the problem is that Brainfuck quines are even longer than Whitespace quines. Between the two of them the result would have been huge. \$\endgroup\$ – breadbox Aug 3 '13 at 5:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.