Is Perl less compact than Ruby? (Kind of competition)

locked_user mtve has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: Is Perl less compact than Ruby? (Kind of competition) by cLive ;-) (Prior) on Feb 18, 2002 at 21:12 UTC
Well, since it doesn't work under strict, you can safely drop: `$n=0;` [download] that's another 5 chars. Also, put pop@n in a sub at the beginning: `sub p{pop@n}` [download] then replace: `pop@n => p` [download] throughout. another 20 chars saved. `sub w{push@n,$_[0]}sub p{pop@n}map{push@L,@C+0;push@C,split//,lc}<>;wh +ile($n<@C){$_=$C[$n++];if(/n/){$0=0;/-/ or$0=$_+$07while($_=index"ht +aoinse",$C[$n++])<7;w$0}if(/t/){$_=p@n;$_?$n=$L[$_-1]:last if p@n}/a/ +&&push@n,1+grep$n>$_,@L;if(/h/){$_=p@n;$n[$#n-abs];$_>0&&splice@n,@n- +2-$_,1}/o/&&print+chr(p@n);/s/&w-(p@n)+p@n;$0=p@n,$_=p@n,w(int$_/$0), +$_%$0if/e/;w($_=getc)?ord:-1if/i/}` [download] cLive ;-) Update* - d'oh, yep, another 16 chars saved by changing p@n to p (updated above) - that means we can also save another 10 chars by whacking push@n,'whatever' into a sub. What's that now? 379 chars? -- seek(JOB,$$LA,0);	[reply] [d/l] [select]
20 bytes from cLive ;-) by locked_user mtve (Deacon) on Feb 18, 2002 at 21:25 UTC
Excellent! `pop@n => p@n` or simply to '`p`', yes?
err, i miscounted by locked_user mtve (Deacon) on Feb 18, 2002 at 21:37 UTC
and another similar optimization on push, now EQUAL TO RUBY - 357 bytes!!! `sub p{pop@n}sub v{push@n,@_}map{push@L,@C+0;push@C,lc=~/./g}<>;while($ +n<@C){$_= $C[$n++];if(/n/){$0=0;/-/ or$0=$_+$0*7while($_=index"htaoinse",$C[$n++ +])<7;v$0} if(/t/){$_=p;$_?$n=$L[$_-1]:last if p}/a/&&v 1+grep$n>$_,@L;if(/h/){$_ +=p;v$n[ $#n-abs];$_>0&&splice@n,@n-2-$_,1}/o/&&print+chr(p);/s/&&v-(p)+p;$0=p, +$_=p,v int$_/$0,$_%$0if/e/;v(($_=getc)?ord:-1)if/i/}` [download] Thank you!!! What's next?
Re: err, i miscounted by locked_user mtve (Deacon) on Feb 18, 2002 at 21:47 UTC
Re: Re: err, i miscounted by dragonchild (Archbishop) on Feb 18, 2002 at 21:56 UTC
Some notes below your chosen depth have not been shown here
Re: err, i miscounted by Anonymous Monk on Feb 18, 2002 at 22:17 UTC
Re: Is Perl less compact than Ruby? (Kind of competition) by tadman (Prior) on Feb 18, 2002 at 19:27 UTC
Before pulling apart the interpreter, you can at least shave a character off the preamble using the old m// trick: `# Formerly: split//,lc # Revised: lc=~m/./g # Revised revision [mtve]: lc=~/./g` [download] Chalk one up for Perl! Update: mtve points out the obvious. D'oh!	[reply] [d/l]
2 bytes from tadman by locked_user mtve (Deacon) on Feb 18, 2002 at 19:39 UTC
Thanks! Worked even without `m`, is it ok?
Re: Is Perl less compact than Ruby? (Kind of competition) by grinder (Bishop) on Feb 18, 2002 at 22:40 UTC
By folding the suggestions encountered so far, and adding in some of my own, I was able to bring the total down to 342 characters. Big gains can be made by observing that `$foo[-2]` refers to indices at the end of an array. That happens twice in this script. In the part where the code pops to $_ and $0, it's cheaper to just call pop @n directly. This was the case with pop@n, and I guess it will still hold with p. sub w as written contains a bug. Sometimes you push more than one element on. But that's okay, @_ is a win over `$_[0]`. I tried changing `expr if cond` to `cond&&expr` but that didn't do anything. `map{push@L,@C+0;push@C,lc=~/./g}<>;sub p{pop@n}sub w{push@n,@_}while($ +n<@C){$_= $C[$n++];if(/n/){$0=0;/-/ or$0=$_+$07while($_=index"htaoinse",$C[$n++ +])<7;w$0} if(/t/){$_=p;$_?$n=$L[$_-1]:last if p}/a/&&w 1+grep{$n>$_}@L;if(/h/){$ +_=p;w$n[- abs$_];$_>0&&splice@n,-1-$_,1}/o/&&print chr p;/s/&&w-p()+p;/e/&&w int + p()/p,p% p;w($_=getc)?ord:-1if/i/}` [download] Note that changing `/s/&&v-(p)+p` to `/s/&&v+p-p` breaks the interpreter. Or at least, the hello.eta program no longer runs correctly, which was the extent of my testing. The interpreter is quite slow. print@_{sort keys %_},$/if%_=split//,'= & a?b:e\f/h^h!j+n,o@o;r$s-t%t#u'	[reply] [d/l] [select]
negative offsets if good by locked_user mtve (Deacon) on Feb 19, 2002 at 07:42 UTC
but a few bugs exists. `p` has side effect of modifying @n, so for example in /e/ part you must call it only twice not four times. in last expression correct logic is `w+($_=getc)?ord:-1if/i/`
Good and bad news by locked_user mtve (Deacon) on Feb 19, 2002 at 08:04 UTC
Good news is that by common efforts we clear first hurdle, 357 bytes. Bad news is that Ruby is also don't stand still, and as you can see here it's now four lines and 306 bytes long! `n=[];p=$<.readlines;$==f=l=c=0;loop{p[l].gsub(/(a)\|(i)\|(n)\|(o)\|(h)\|(e) +\|(s)\|t/){ f>0?$6?n<<c&&c=f=0:c="htaoins".index($&)+c*7: $1?n<<l+2: $2?n<<STDIN.g +etc: $3?f= 1:(a=n.pop;$4?$><<a.chr: $5?n<<n[-1-a.abs]&&a>0&&n.slice!(-2-a):(b=n.p +op;$6?n<< b/a&&n<<b%a: $7?n<<b-a: b!=0&&(l=a-2;break)))};p[l+=1]&&l>=0\|\|break}` [download] How do you like this intrigue?
last version by locked_user mtve (Deacon) on Feb 19, 2002 at 08:36 UTC
is 342 bytes `map{push@L,@C+0;push@C,lc=~/./g}<>;sub p{pop@n}sub w{push@n,@_}while($ +n<@C){$_= $C[$n++];if(/h/){$_=p;w$n[-1-abs];$_>0&&splice@n,-2-$_,1}if(/n/){w 0;/ +-/\|\|w$_+7 *p while($_=index"htaoinse",$C[$n++])<7}$0=p,$_=p,w int$_/$0,$_%$0if/e +/;if(/t/) {$_=p;$_?$n=$L[$_-1]:last if p}/a/?w 1+grep$n>$_,@L:/o/?print chr p:/s +/?w-(p)+p :/i/?w+($_=getc)?ord:-1:0}` [download] Beg you for another 26 bytes, is it possible?
Re: last version by grinder (Bishop) on Feb 19, 2002 at 11:44 UTC
Down to 340 bytes: `map{push@L,@C+0;push@C,lc=~/./g}<>;sub p{pop@n}sub w{push@n,@_}while($ +n<@C){$_= $C[$n++];if(/h/){$_=p;w$n[-1-abs];$_>0&&splice@n,-2-$_,1}if(/n/){w 0;/ +-/\|\|w$_+7 p while($_=index"htaoinse",$C[$n++])<7}$0=p,$_=p,w int$_/$0,$_%$0if/e +/;if(/t/) {$_=p;$_?$n=$L[$_-1]:last if p}/a/?w 1+grep$n>$_,@L:/o/?print chr p:/s +/?w-(p)+p :/i/&&w($_=getc)?ord:-1}` [download] We can `s/last if p/p&&last/` and that brings it own to 338* bytes: `map{push@L,@C+0;push@C,lc=~/./g}<>;sub p{pop@n}sub w{push@n,@_}while($ +n<@C){$_= $C[$n++];if(/h/){$_=p;w$n[-1-abs];$_>0&&splice@n,-2-$_,1}if(/n/){w 0;/ +-/\|\|w$_+7 p while($_=index"htaoinse",$C[$n++])<7}$0=p,$_=p,w int$_/$0,$_%$0if/e +/;if(/t/) {$_=p;$_?$n=$L[$_-1]:p&&last}/a/?w 1+grep$n>$_,@L:/o/?print chr p:/s/? +w-(p)+p: /i/&&w($_=getc)?ord:-1}` [download] But wait, there's more! change `$0=p,$_=p,w int$_/$0,$_%$0` to `w int($_=p)/($0=p),$_%$0}` Which gives 337* (I think) bytes: `map{push@L,@C+0;push@C,lc=~/./g}<>;sub p{pop@n}sub w{push@n,@_}while($ +n<@C){$_= $C[$n++];if(/h/){$_=p;w$n[-1-abs];$_>0&&splice@n,-2-$_,1}if(/n/){w 0;/ +-/\|\|w$_+7 p while($_=index"htaoinse",$C[$n++])<7}w int($_=p)/($0=p),$_%$0if/e/; +if(/t/) {$_=p;$_?$n=$L[$_-1]:p&&last}/a/?w 1+grep$n>$_,@L:/o/?print chr p:/s/? +w-(p)+p: /i/&&w($_=getc)?ord:-1}` [download] A ternary operator can be replaced by a `&&` and the plus in `w+($_=getc)?ord:-1` trick is not needed, it works without (or, at least, hello.eta still outputs correctly). I have an idea it must be possible to replace the `w$_+7p while($_=index"htaoinse",$C[$n++])<7` section with a redo and/or a pos, but it is so difficult to understand what the program does I am beginning to wonder whether it wouldn't be easier to look at the instruction set and come up with another angle of attack. For instance, it would be nice to be able to do away with the @L array as well. Replace @C and @L by a hash that is keyed on byte offset and the value is the byte. print@_{sort keys %_},$/if%_=split//,'= & *a?b:e\f/h^h!j+n,o@o;r$s-t%t#u'	[reply] [d/l] [select]
1 byte by locked_user mtve (Deacon) on Feb 19, 2002 at 14:19 UTC
Re: Is Perl less compact than Ruby? (Kind of competition) by dragonchild (Archbishop) on Feb 19, 2002 at 17:14 UTC
339. `map{push@L,@C+0;push@C,lc=~/./g}<>;sub p{$_=pop@n}sub w{push@n,@_}whil +e($n<@C){$_=$C[$n++];if(/h/){p;w$n[-1-abs];$_>0&&splice@n,-2-$_,1}if( +/n/){w 0;$b<0\|\|w$b+7p while($b=index"htaoinse",$C[$n++])<7}$0=p,p,w +int$_/$0,$_%$0if/e/;if(/t/){$a=p;$a?$n=$L[$a-1]:last if p}/a/?w 1+gre +p$n>$_,@L:/o/?print chr p:/s/?w-(p)+p:/i/&&w+($_=getc)?ord:-1}` [download] The big change was `sub p{$_=pop@n}` and the subsequent fall-out. pants\ Update:* To 336... `if(/t/){$a=r;$a?$n=$L[$a-1]:last if r} --- $a=r,r&&($a?$n=$L[$a-1]:last)if/t/;` [download] ------ We are the carpenters and bricklayers of the Information Age. Don't go borrowing trouble. For programmers, this means Worry only about what you need to implement.	[reply] [d/l] [select]
29 characters to beat Ruby ... by dragonchild (Archbishop) on Feb 19, 2002 at 18:39 UTC
Another update, bringing to 331. Though I'm not sure the change is legal, it runs the Tic-Tac-Toe program. `map{push@L,push@C,lc=~/./g}<>;sub p{$_=pop@n}sub w{push@n,@_}while($n< +@C){$_=$C[$n++];if(/h/){p;w$n[-1-abs];$_>0&&splice@n,-2-$_,1}if(/n/){ +w 0;$b<0\|\|w$b+7p while($b=index"htaoinse",$C[$n++])<7}$0=p,p,w int$_ +/$0,$_%$0if/e/;$a=p,p&&($a?$n=$L[$a-2]:last)if/t/;/a/?w 2+grep$n>$_,@ +L:/o/?print chr p:/s/?w-(p)+p:/i/&&w+($_=getc)?ord:-1}` [download] The change was to functionally remove the 0 at the beginning of @L. Adjusting the subscripts used to access @L seems* to result in an acceptable performance. ------ We are the carpenters and bricklayers of the Information Age. Don't go borrowing trouble. For programmers, this means Worry only about what you need to implement.	[reply] [d/l]
Re: 29 characters to beat Ruby ... by Matts (Deacon) on Feb 20, 2002 at 09:55 UTC
One thing people often forget is that push@L,values is a semantically equivalent (modulo being faster) to @L=@L,values - a saving of one character per push, or three characters in the above, bringing you down to 328.	[reply]
Re: Re: 29 characters to beat Ruby ... by dragonchild (Archbishop) on Feb 20, 2002 at 13:19 UTC
oops by locked_user mtve (Deacon) on Feb 20, 2002 at 10:46 UTC
increasing on 2 bytes by locked_user mtve (Deacon) on Feb 20, 2002 at 10:02 UTC
Excellent tweak with push! `map{push@L,push@C,lc=~/./g}0,<>;` should fix this. It's nice question but I believe impossibility to jump on first line is some violation of language. It's my fault of incorrect target setting.
Superb dragonchild! by locked_user mtve (Deacon) on Feb 19, 2002 at 18:37 UTC
confirmed. i can't believe it! 336 bytes. only 16 bytes to make it four-liner.
Re: Is Perl less compact than Ruby? (Kind of competition) by jynx (Priest) on Feb 19, 2002 at 19:03 UTC
As for that last section, In my testing it seems to work with the following change (saving 4 more characters): `/i/&&w+($_=getc)?ord:-1 # goes to: /i/&&w(ord getc\|-1)` [download] jynx update: take out the parens on that up there: `/i/&&w ord getc\|-1` [download] for another 1 char savings...	[reply] [d/l] [select]
Re: Re: Is Perl less compact than Ruby? (Kind of competition) by dragonchild (Archbishop) on Feb 19, 2002 at 19:15 UTC
`/i/&& w ord getc\|\|-1` [download] You forgot a '\|'. That brings us to 306. :-) ------ We are the carpenters and bricklayers of the Information Age. Don't go borrowing trouble. For programmers, this means Worry only about what you need to implement.	[reply] [d/l]
Re: Re: Re: Is Perl less compact than Ruby? (Kind of competition) by jynx (Priest) on Feb 19, 2002 at 19:41 UTC
my testing has been limited, i've only been using hello.eta and hello2.eta, but using `\|` seems to be equivalent to `\|\|` there. i tested both ways and they seem to work okay on the two hello programs. did your tests come out differently? if so, which programs should i be testing with (that is, which are more indicative of a real ETA program?) As for the tic-tac-toe program, no matter which of the recent incarnations of the perl program i try, it seems to hang after i've given it input. does it just take a really long time or is it still running? jynx	[reply] [d/l] [select]
a little explanation by locked_user mtve (Deacon) on Feb 20, 2002 at 10:13 UTC
Re: Re: Re: Re: Is Perl less compact than Ruby? (Kind of competition) by dragonchild (Archbishop) on Feb 19, 2002 at 21:42 UTC
Re: Is Perl less compact than Ruby? (Kind of competition) by blakem (Monsignor) on Feb 18, 2002 at 19:35 UTC
Oops... nothing to see here. ~~While I don't have time to actually play this round of golf, the perl version can definately be shortened. For instance, the `while` clause in the first line jumps out as being longer than necessary:~~ ~~`while($n<@C){$_=$C[$n++]};`~~ ~~[download]~~ Becomes: `$_=$C[$n++]while$n<@C;` [download] Shaving 4 chars off of the total..... -Blake	[reply] [d/l] [select]
sorry, no by locked_user mtve (Deacon) on Feb 18, 2002 at 19:42 UTC
It's longer loop.
Hooray! by locked_user mtve (Deacon) on Feb 20, 2002 at 21:18 UTC
I must say it again, WE WIN with 290 bytes: `sub p{$_=pop@n}sub w{push@n,@_}$/=$T;$_=uc<>;s#.#+{qw(E 7$0=p;p;w+int$ +_/$0,$_%$0 H 0p;w$n[-1-abs];$_>0&&splice@n,-2-$_,1 I 4w+($_=getc)?ord:-1 A 2w$L+1 + N 5w$N++ S 6w-(p)+p O 3print+chr(p) T 1$t=p;p&&goto"L$t")}->{$&}=~/./&&"\$N?$&> +6?\$N=0:". "w $&+7*p:do{$'};"#eg;s/^/L${\++$i}:\$L=$i;/mg;eval` [download]
Latest results by locked_user mtve (Deacon) on Feb 20, 2002 at 11:26 UTC
Thanks to tadman, cLive ;-) , blakem, dragonchild, grinder, Matts, jynx and PerlMonks.org, Perl version in now officially 333 bytes (or 327 bytes with some cheating), while Ruby is still 306 bytes. `map{push@L,push@C,lc=~/./g}0,<>;sub p{$_=pop@n}sub w{push@n,@_}while($ +n<@C){$_=$ C[$n++];if(/h/){p;w$n[-1-abs];$_>0&&splice@n,-2-$_,1}if(/n/){w 0;$b<0\| +\|w$b+7*p while($b=index"htaoinse",$C[$n++])<7}$0=p,p,w int$_/$0,$_%$0if/e/;$a=p +,p&&($a?$n =$L[$a-1]:last)if/t/;/a/?w 1+grep$n>$_,@L:/o/?print chr p:/s/?w-(p)+p: +/i/&&w+($_ =getc)?ord:-1}` [download]
Completely different algorithm by locked_user mtve (Deacon) on Feb 20, 2002 at 15:18 UTC
Beta version, 325 bytes, expanded view, subject to optimizations: `sub p{$_=pop@n} sub w{push@n,@_} %D=( H=>'0p;w$n[-1-abs];$_>0&&splice@n,-2-$_,1', T=>'1$t=p;p&&goto"L$t"', A=>'2w$L+1', O=>'3print chr p', I=>'4w+($_=getc)?ord:-1', N=>'5w$N++', S=>'6w-(p)+p', E=>'7$0=p;p;w int$_/$0,$_%$0' ); $/=\-1;$_=uc<>; s#.#$D{$&}=~/(.)(.)$/&&"\$N?$1>6?\$N=0:w $1+7p:do{$2};"#eg; s/^/$i++,"L$i:\$L=$i;"/emg; eval; L0:` [download]
Re: Completely different algorithm by japhy (Canon) on Feb 20, 2002 at 19:51 UTC
Here's some more, then. It also adds the (missing?) newline to the end of each command. `sub p{$_=pop@n} sub w{push@n,@_} $/=$T;$_=uc<>; s#.#+{qw( H 0p;w$n[-1-abs];$_>0&&splice@n,-2-$_,1 T 1$t=p;p&&goto"L$t" A 2w$L+1 O 3print+chr+p I 4w+($_=getc)?ord:-1 N 5w$N++ S 6w-(p)+p E 7$0=p;p;w+int$_/$0,$_%$0 )}->{$&}=~/./&&"\$N?$&>6?\$N=0:w $&+7p:do{$'};\n"#eg; s/^/L${\++$i}:\$L=$i;/mg; eval; L0:` [download] _____________________________________________________ Jeff`[japhy]`Pinyan: Perl, regex, and perl* hacker, who'd like a (from-home) job `s++=END;++y(;-P)}y js++=;shajsj<++y(p-q)}?print:??;`	[reply] [d/l]
* japhy's qw is GODLIKE * by locked_user mtve (Deacon) on Feb 20, 2002 at 21:13 UTC
Re: * japhy's qw is GODLIKE * by japhy (Canon) on Feb 20, 2002 at 21:40 UTC
Some notes below your chosen depth have not been shown here
Re: * japhy's qw is GODLIKE * by japhy (Canon) on Feb 20, 2002 at 21:19 UTC
Some notes below your chosen depth have not been shown here
Re: Completely different algorithm by dragonchild (Archbishop) on Feb 20, 2002 at 17:58 UTC
Nicely done! Shave three with `I=>'4w ord getc\|\|-1',`. ------ We are the carpenters and bricklayers of the Information Age. Don't go borrowing trouble. For programmers, this means Worry only about what you need to implement.	[reply] [d/l]
It's cheating by locked_user mtve (Deacon) on Feb 20, 2002 at 19:14 UTC