Roku Developer Program

[RokuMarkn]

"EnTerr" wrote:

That's like saying "10 LIST" is a legal quine (self-printing program).

Yeah, someone once responded to my challenge to write a Perl Quine by writing a program that opened and read the source file and printed it. Hey, I haven't seen anyone write a Brightscript Quine yet. Any takers?

--Mark

[TheEndless]

"RokuMarkn" wrote:

"EnTerr" wrote:

That's like saying "10 LIST" is a legal quine (self-printing program).

Yeah, someone once responded to my challenge to write a Perl Quine by writing a program that opened and read the source file and printed it. Hey, I haven't seen anyone write a Brightscript Quine yet. Any takers?

--Mark

Quine is new to me, but does this count?

For Each file In ListDir("pkg:/source"):print ReadAsciiFile("pkg:/source/" + file):Next

[EnTerr]

"RokuMarkn" wrote:
Hey, I haven't seen anyone write a Brightscript Quine yet. Any takers?

Done! Can i have a cookie?

[spoiler=83 chars:3udiu46y]No cheating! Giving out the answer spoils all the fun, pain and suffering. Here is an alphabetical anagram of my solution (it runs on fw3 too; PM me for the unscrambled version if not wanting to try the quine challenge on your own):

""(((((((((())))))))))++11333333444455668888:=??SSSSScccccccccchhhhhhhhhhrrrrrrrrrr

Can there be an even shorter one, i wonder?![/spoiler:3udiu46y]

[EnTerr]

"TheEndless" wrote:
Quine is new to me, but does this count?

For Each file In ListDir("pkg:/source"):print ReadAsciiFile("pkg:/source/" + file):Next

No, that's cheating. Such rejection may sound arbitrary ("why can't i read external data?"). I don't know if this will help as justification but since above is a one-line program, i should be able to run it in console and it should print exactly what i just typed, which i should be able to copy&paste in the console again to get the same output (etc etc ad nauseam).

The challenge is to do something that feels impossible when you try at first. Have to write a non-empty program/snippet - using only the means of the language - that prints itself. It should survive compilation (if there ever was one for B/S) where source file disappears.

[RokuMarkn]

My first attempt came out to 84 chars, although based on your anagram I think we're using rather different techniques.

--Mark

[RokuMarkn]

Well, here's mine. Don't look if you're still playing.
[spoiler=84 chars:266a1re8]

q=chr(34):p="q=chr(34):p=:?mid(p,1,12);q;p;q;mid(p,13)":?mid(p,1,12);q;p;q;mid(p,13)

[/spoiler:266a1re8]

[EnTerr]

"RokuMarkn" wrote:
My first attempt came out to 84 chars, although based on your anagram I think we're using rather different techniques.

I had success so far only using the form
S="¿":¿
where ¿ is code that prints S="¿":¿ by using S; it starts with "?" and has no " quotes. The anagram amusingly is very obvious which function i liberally (ab)used... courtesy of the order of the alphabet.

I keep thinking eval() may come useful but nothing so far.

Shouldn't we split this thread from around here viewtopic.php?f=34&t=87203#p493015 into a new "quine" one?

[RokuMarkn]

"EnTerr" wrote:

Shouldn't we split this thread from around here viewtopic.php?f=34&t=87203#p493015 into a new "quine" one?

Yes.

--Mark

[EnTerr]

"RokuMarkn" wrote:
Well, here's mine. Don't look if you're still playing.
[spoiler=84 chars:z505pj9s]...[/spoiler:z505pj9s]

Interestingly, there is some space for "optimization" (shortenization? briefization?) there:

• replacing the first MID with LEFT shortens it from 84 to 82 chars


• dropping `q` and using chr(34) in all 4 places keeps length at 82; does not seem to help but wait, wait for it...


• ... omitting ";" wherever possible* (2x3 places) now brings length down to 76, woo-hoo!

Btw. saying that you used different technique made me thinking yesterday what else can i do... and i remembered a string function documented last year (it's available in fw3 though). With its help my personal record now is down to 62! **
[spoiler=62 chars ramagan:z505pj9s]

""(((()))),,,,0013344::==??SSSSSS^^^bbcceehhiirrssssttttttuuuu

[/spoiler:z505pj9s]

(*) i know, border-line criminal but the B/S parser is the de-facto standard on the language (like perl) and it does not require ";" nor extra whitespace in print
(**) i wonder if someone else (e.g. @belltown) can bring that count even lower