Accepted

Score	Golfer	Submit Time	Code
64.58	Ton Hospel	Sat Jul 6 22:05:39 2002	-ln0 / \Q$&/||print$1while s/( |(?= ?\G))\Q$&/ /g,/(\S+) /g/$#+
64.58	Ton Hospel	Sun Jul 7 11:46:18 2002	-ln0 /(?<= )\Q$& /||print$1while s/( |\G)\Q$&/ /g,/(\S+) /g/$#+
			the dual solution
65.50	Ton Hospel	Wed Jul 3 16:46:46 2002	-ln0 / \Q$&\E$/m||print$&while s/(^|\G)\Q$&\E\s/ /mg,/(\S+)/g/$#+
			Do everything with one statement A pity I needed to add that () But maybe distinguishing $1 and $& will prove useful
65.50	Ton Hospel	Wed Jul 3 12:05:33 2002	-ln0 / \Q$&\E$/m||print$&while s/(^|\G)\Q$&\E\s/ /mg,/\S+/g;1/!//
68.52	Mtv Europe	Sun Jul 7 15:37:14 2002	-ln0 / \Q$& /||print$&while s/^\Q$&\E\s|^(.+) \1$/$1/mg,/\S+/g;a%!??
			another two bytes, improved tie and a little obfu at the end :) perhaps my last, thanks to referees and to all.
70.48	Ton Hospel	Tue Jul 2 14:29:42 2002	-ln0 s/\G \Q$& / /+/ \Q$& /||print$&while s/^\Q$&\E\s//mg,/\S+/g;1/!//
70.63	Mtv Europe	Sat Jul 6 18:56:10 2002	-p0 $\.="$& "x!/ \Q$& /,s/^\Q$&\E[ ]|^(.+) \1$/$1/mgwhile/\S+/g;a/!//
			strangely it runs so long so i began to wonder will it be accepted?
71.49	Ton Hospel	Tue Jul 2 13:12:08 2002	-ln0 s/\G \Q$& / /-/ \Q$& /||print($&)+s/^\Q$&\E\s//mgwhile/\S+/g;1/!//
71.51	Ton Hospel	Tue Jul 2 13:26:12 2002	-p0 s/\G \Q$& / /+/ \Q$& / or$\.="$& ",s/^\Q$&\E\s//mgwhile/\S+/g;1/!//
72.49	Ton Hospel	Tue Jul 2 12:49:11 2002	-ln0 s/\G \Q$& / /+/ \Q$& /||print($&)+s/^\Q$&\E\s/ /mgwhile/\S+/g;1/!//
73.46	Albert Dvornik	Fri Jul 5 22:27:31 2002	-ln0 s/\G \Q$& / /./ \Q$& /||s/^(\Q$&\E)\s//gm-print$1while/\S+/g;lc&&die
			Got rid of $q; two characters are better than nothing...
73.49	Ton Hospel	Tue Jul 2 12:40:32 2002	-ln0 s/\G \Q$& / /+/ \Q$& /||print($&)+s/^\Q$&\E\s/\n/mgwhile/\S+/g;1/!//
74.47	Wladimir Palant	Sun Jul 7 16:51:26 2002	-ln0 $a="\Q$&",s/\G $a$//m|/ $a$/m||(print$&)&s/^$a\s//gmwhile/\S+/g;a%!$_
			Improving the result by 0.01, have no better idea
74.48	Wladimir Palant	Fri Jul 5 22:18:43 2002	-ln0 $a="\Q$&",s/\G $a$//m|/ $a$/m||print($&)&s/^$a\s//gmwhile/\S+/g;1%!$_
74.50	pom	Sat Jul 6 22:00:14 2002	-ln0 s/^(.+) \1$/$1/m|/ \Q$& /||print($&)^s/^\Q$&\E\s//gmwhile/\S+/g;1/!$_
			Finally, I can get rid of the "die" at the end. I had another way with "exec$_" which I found quite devious, but 1/!$_ is one stroke shorter so there it is.
74.54	Ton Hospel	Tue Jul 2 00:41:12 2002	-ln0 /^(.+) (?!\1 )\Q$& /m||print($&)+s/^(\Q$&\E\s)+/ /mgwhile/\S+/g;1/!//
75.46	Albert Dvornik	Fri Jul 5 01:19:06 2002	-ln0 $q="\Q$&",s/\G $q$//m,/ $q /||s/^($q)\s//mg+print$1while/\S+/g;lc&&die
			Handling only one "node node" at a time is enough.
75.50	Ala Qumsieh	Mon Jul 8 04:50:52 2002	-ln0 s/^(.*) \1$/$1/mg;/ \Q$& /||print($&),s/^\Q$&\E\s//mgwhile/\S+/g;$_&&&
			Those brackets arount the print are killing me!
76.47	Wladimir Palant	Fri Jul 5 14:27:39 2002	-ln0 $a="\Q$&",s/\G $a$//m|/ $a$/m||print($&)&s/^$a\s//gmwhile/\S+/g;$_&&die
76.49	pom	Thu Jul 4 15:34:33 2002	-ln0 s/^(.+) \1$/$1/m|/ \Q$& /||print($&)^s/^\Q$&\E\s//gmwhile/\S+/g;??&&die
			Never in my wildest dreams would I have thought that you could write something like s///gmwhile with no whitespace.
76.50	Ala Qumsieh	Mon Jul 8 03:51:25 2002	-ln0 s/^(.+) \1$/$1/gm;/ \Q$& /||print($&),s/^\Q$&\E\s//mgwhile/\S+/g;$_&&&;
76.52	Qingning Huo	Thu Jul 4 14:52:05 2002	-ln0 $*=qr/\Q$&/,s/^($*) $* //,/ $* /||s/^($*)\s//g&print$1while/\S+/g;a/!$_
76.54	Ton Hospel	Mon Jul 1 23:20:05 2002	-ln0 /^(?!\Q$&\E ).* \Q$& /m||print($&)+s/^(\Q$&\E\s)+/ /mgwhile/\S+/g;1/!//
76.58	Mtv Europe	Sat Jul 6 13:28:37 2002	-p0 s/^(.+) \1$/$1/mg;$\.="$& "x!/ \Q$& /,s/^\Q$&\E[ ]//mgwhile/\S+/g;a/!//
			hmm, still works, strange.
77.46	Wladimir Palant	Fri Jul 5 14:13:25 2002	-ln0 $a="\Q$&",s/\G $a$//m|/ $a$/m||print$&x1,!s/^$a\s//gmwhile/\S+/g;$_&&die
77.49	Ala Qumsieh	Sun Jul 7 04:48:23 2002	-ln0 s/^(.+) \1$/$1/mg;/ \Q$& /||print($&),s/^\Q$&\E\s//mgwhile/\S+/g;$_&&die
77.49	pom	Wed Jul 3 16:27:21 2002	-ln0 s/^(.+) \1$/$1/m|/ \Q$& /||print($&)^s/^\Q$&\E\s//gm while/\S+/g;$_&&die
			AT LAST I can get rid of one of those damn \Q...\E characters!!!
77.52	Qingning Huo	Wed Jul 3 17:50:52 2002	-ln0 $*=qr/\Q$&/,s/^($*) $* //,/ $*$/||s/^($*)\s//g&print$1while/^\S+/g;a/!$_
77.52	Qingning Huo	Wed Jul 3 21:23:16 2002	-ln0 $*=qr/\Q$&/,s/^($*) $* //,/ $* /||s/^($*)\s//g&print$1while/^\S+/g;a/!$_
			improved tie-breaker
77.54	Mtv Europe	Sat Jul 6 13:10:15 2002	-ln0 s/^(.+) \1$/$1/mg;/ \Q$& /||print($&)+s/^\Q$&\E[ ]//mgwhile/\S+/g;a/!//
			just found new a/// operator
78.47	Wladimir Palant	Fri Jul 5 13:25:16 2002	-0n $a="\Q$&",s/\G $a$//m|/ $a$/m||print"$&\n",!s/^$a\s//gmwhile/\S+/g;$_&&die
78.53	Ton Hospel	Mon Jul 1 18:21:23 2002	-ln0 /^(?!\Q$&\E ).* \Q$& /m||print($&)+s/^(\Q$&\E\s)+/ /mgwhile/^\S+/gm;1/!//
78.53	Juuso Salonen	Sun Jul 7 20:43:35 2002	-ln0 S//\S+/g,/(?<!(^|\A)\Q$&\E) \Q$& /m||print($&)+s/^(\Q$&\E\s)+//mgwhile/./
79.46	Albert Dvornik	Thu Jul 4 19:52:49 2002	-ln0 $q="\Q$&",s/^$q $q$/$q/mg,/ $q /||s/^($q)\s//mg+print$1while/\S+/g;lc&&die
79.49	Ala Qumsieh	Thu Jul 4 15:06:43 2002	-ln0 s/^(.+) \1$/$1/mg;/ \Q$& /||print($&),s/^\Q$&\E\s//gmwhile/^\S+/mg;$_&&die
79.49	pom	Wed Jul 3 14:21:22 2002	-ln0 s/^(.+) \1$/$1/m|/ \Q$&\E /||print($&)^s/^\Q$&\E\s//gm while/\S+/g;$_&&die
79.51	Qingning Huo	Wed Jul 3 13:48:24 2002	-ln0 $*=qr/\Q$&/,s/^($*) $* //,/ $*$/||s/^($*)\s//g&print$1while/^\S+/g;$_&&die
			$* is fun.
79.53	Mtv Europe	Sat Jul 6 00:36:38 2002	-ln0 s/^(.+) \1$/$1/mg;/ \Q$& /||print($&)+s/^\Q$&\E[ ]//mgwhile/\S+/g;//&&die
			sleep
80.49	pom	Wed Jul 3 08:26:53 2002	-ln0 s/^(.+) \1$/$1/m|/ \Q$&\E$/m||print($&)^s/^\Q$&\E\s//gm while/\S+/g;$_&&die
			OK this one should work. There is one test case missing from the test program, something like: a aa c aa b aa
80.51	Ton Hospel	Mon Jul 1 18:01:02 2002	-Xln0 /^(?!\Q$&\E ).* \Q$& /m||print($&)+s/^(\Q$&\E\s)+/ /mgwhile/^\S+/gm;exit//
80.52	Juuso Salonen	Thu Jul 4 18:42:28 2002	-ln0 s// /;S//(\S+)/g,/(?<! \Q$1\E) \Q$1 /||(print$1)&s/ (\Q$1\E\s)+/ /gwhile/./
81.49	Ala Qumsieh	Thu Jul 4 04:56:19 2002	-ln0 s/^(.*) \1$/$1/mg; / \Q$& /||print($&),s/^\Q$&\E\s//gmwhile/^\S+/mg; $_&&die
			Sorry for keeping you busy :)
81.51	Ton Hospel	Mon Jul 1 16:39:51 2002	-Xln0 /^(?!\Q$&\E ).* \Q$& /m||print"$&",!s/^(\Q$&\E\s)+/ /mgwhile/^\S+/gm;exit//
82.49	pom	Tue Jul 2 16:56:47 2002	-ln0 s/^(.+) \1$/$1/gm;/ \Q$&\E$/m||print($&)^s/^\Q$&\E\s//gm while/\S+/gm;$_&&die
			It's always when you post a solution that you find how to improve it
82.53	Mtv Europe	Fri Jul 5 22:09:14 2002	-lan0 s/^(\S+) \1$/$1/mg;/ \Q$& /||print($&)+s/^\Q$&\E[ ]//mgwhile/\S+/g;/./&&die
			just a surplus space
83.37	Keith Calvert Ivey	Sun Jul 7 23:06:00 2002	-ln0 s/^(.+) \1$/$1/gm;print(($s)=grep$s!~/ \Q$_ /,split)/s/^\Q$s\E\s//gmwhile$s=$_
			Finally, a breakthrough -- though I fear I'm still stuck in a local minimum
83.47	Qingning Huo	Wed Jul 3 04:43:50 2002	-ln0 $r=qr/\Q$&/,s/^($r) $r //m,/ $r$/m||s/^($r)\s//gm&print$1while/^\S+/gm;$_&&die
83.49	pom	Tue Jul 2 16:42:25 2002	-ln0 s/^(.+) \1$/$1/gm;/ \Q$&\E$/m||print($&)^s/^\Q$&\E\s//gm while/^\S+/gm;$_&&die
			I used one of Ton's trick to gain a stroke!
83.49	Wladimir Palant	Fri Jul 5 10:19:53 2002	-0 $_=<>;$a="\Q$&",s/\G $a$//m|/ $a$/m||print"$&\n",!s/^$a\s//gmwhile/\S+/g;$_&&die
83.49	Albert Dvornik	Thu Jul 4 18:30:13 2002	-ln0 s/^(.*) \1$/$1/mg;$q="\Q$&",/ $q$/m||s/^($q)\s//mg&&print$1while/\S+/g;$_&&die
			Using //g is more compact than split.
83.50	Stephen Turner	Sun Jul 7 17:00:03 2002	-lan0 for$a((@F)x@F){/^(?!\Q$a\E ).* \Q$a /m||s/(?<!\S)\Q$a\E\s//g&&print$a}$_&&die
			Whoops, nearly missed a simple four-stroke improvement.
83.53	Mtv Europe	Fri Jul 5 22:01:12 2002	-lan0 s/^(\S+) \1$/$1/mg;/ \Q$& /||print($&)+s/^\Q$&\E[ ]//mg while/\S+/g;/./&&die
			using ton-'s favourite technique - throwing away unnecessary chars
84.46	Michael Thelen	Thu Jul 4 05:17:15 2002	-ln0 s/^(.*) \1$/$1/mg;s!.!/ \Q$& /||(print$&)&s/^\Q$&\E\s//mg&redo while/\S+/g;&a!e
			My program fits on one line! Well, besides the #!perl line, anyway.
84.47	pom	Tue Jul 2 16:33:41 2002	-ln0 s/^(.+) \1$/$1/gm;/ \Q$&\E$/m||print$&and s/^\Q$&\E\s//gm while/^\S+/gm;$_&&die
			Whew, I'm having more and more trouble understanding what my solution does...
84.50	Marko Nippula	Fri Jul 5 11:54:37 2002	-lna0 for$a((@F)x@F){$z="\Q$a",/^(?!$z ).* $z /m||s/(\G|\s)$z\s/ /g&&print$a}&#if/./
85.48	Qingning Huo	Tue Jul 2 13:02:14 2002	-ln0 $r=qr/\Q$&/,s/^($r) \1$/$1/m,/ $r$/m||s/^($r)\s//gm&print$1while/^\S+/gm;$_&&die
85.49	Wladimir Palant	Fri Jul 5 09:55:42 2002	-0 $_=<>;s/\G \Q$&\E$//m|/ \Q$&\E$/m||print"$&\n",!s/^\Q$&\E\s//gmwhile/\S+/g;$_&&die
85.51	Michael Thelen	Thu Jul 4 04:51:52 2002	-ln0 s/^(.*) \1$/$1/mg;/./&&${/ \Q$& /||(print$&)&s/^\Q$&\E\s//mg&redo while/\S+/g}/0
			Finally got rid of that dang $t, and now the program almost fits on one line. I can feel more strokes here to be shaved.
86.47	Ala Qumsieh	Thu Jul 4 04:52:37 2002	-ln0 s/^(.*) \1$/$1/mg; /^.* \Q$& /ms||print($&),s/^\Q$&\E\s//gmwhile/^\S+/mg; $_&&die
86.50	Michael Thelen	Wed Jul 3 23:57:19 2002	-ln0 s/^(.*) \1$/$1/mg;/ \Q$& /||($t=print$&)&s/^\Q$&\E\s//mgwhile/\S+/g;///$t--&&redo
			Continuing the variation on a theme. I have a feeling this may be my last improvement for a while...
86.54	Juho Snellman	Sun Jul 7 23:32:23 2002	-ln0 $_ x=3;s/(^| )(\Q$1\E\s)+/$1&&&1/megwhile/ (\S+)\s(?!\C* (?!\1 ).+ \1 )/&&print$1
			Replaced x=2 with x=3, which allowed changing a ^ from the regexp to \n, which allowed removing a /m-modifier from the regexp.
87.51	Stephen Turner	Sun Jul 7 15:18:24 2002	-lan0 for$a((@F)x y/ //){/^(?!\Q$a\E ).* \Q$a /m||s/(?<!\S)\Q$a\E\s//g&&print$a}$_&&die
			Uglier, but better tie-breaker.
87.53	Stephen Turner	Sun Jul 7 15:18:04 2002	-lan0 for$a((@F)x y/ //){$b="\Q$a";/^(?!$b ).* $b /m||s/(?<!\S)$b\s//g&&print$a}$_&&die
			Into the 80's. Now if I can only lose another 23 strokes in the next 13 hours, I could still win this!
87.54	Juho Snellman	Thu Jul 4 22:20:04 2002	-ln0 $_ x=2;s/(^| )(\Q$1\E\s)+/$1&&&1/megwhile/^(\S+)\s(?!\C*^(?!\1 ).+ \1 )/m&&print$1
			Finally had time to construct a test-case for the suspicious bit of code in my solution. Any sub-84 solution from me is going to fail on "a b b c a a". Time for v1.9? :-)
87.57	Eugene van der Pijll	Mon Jul 8 02:12:30 2002	-lan0 sub f{$p{$_}||=1/!$g{$_}+++map(&f,$x=~/^(.*) (?!\1 )\Q$_ /mg)+print}$x=$_;f for@F
88.47	Ala Qumsieh	Thu Jul 4 04:45:36 2002	-ln0 s/^(.*) \1$/$1/mg; /^.* \Q$&\E /ms||print($&),s/^\Q$&\E\s//gmwhile/^\S+/mg; $_&&die
			inching closer .. and closer ..
88.50	Michael Thelen	Wed Jul 3 23:53:39 2002	-ln0 s/^(.*) \1$/$1/mg;/ \Q$& /||($t=print$&)&s/^\Q$&\E\s//mgwhile/\S+/g;1/$t--,//&&redo
			Okay, now we'll do illegal division by zero. How many ways can we get perl to die? ;-)
88.51	Wladimir Palant	Fri Jul 5 01:36:36 2002	-0 $_=<>;s/^(\S+) \1$/$1/gm;/ \Q$&\E$/m||print"$&\n",!s/^\Q$&\E\s//gmwhile/\S+/g;$_&&die
88.52	Mtv Europe	Fri Jul 5 20:55:40 2002	-lan0 s/^(\S+) \1$/$1/mg;/ \Q$& /||print($&)+s/^\Q$&\E[ ]//mg+redo while/\S+/g;/./&&die
			small step
89.50	Qingning Huo	Tue Jul 2 11:43:23 2002	-ln0 $0=$&,s/^(\Q$0\E) \1$/$1/m,/ \Q$0\E$/m||s/^\Q$0\E\s//gm&print$0while/^\S+/gm;$_&&die
89.50	Michael Thelen	Wed Jul 3 23:45:47 2002	-ln0 s/^(.*) \1$/$1/mg;/ \Q$& /||($t=print$&)&s/^\Q$&\E\s//mgwhile/\S+/g;$t--?//&&redo:&a
			Changed 'die' to &a (undefined sub). I've tried three completely different approaches and this one's still worked the best, but I'm not so confident in it anymore.
89.51	Wladimir Palant	Thu Jul 4 22:25:40 2002	-0 $_=<>;s/^(\S+) \1$/$1/mg;/ \Q$&\E$/m||print("$&\n"),s/^\Q$&\E\s//gmwhile/\S+/g;//&&die
89.52	Juuso Salonen	Thu Jul 4 16:13:15 2002	-ln0 s// /;0//(\S+)/g,/(?<! \Q$1\E) \Q$1 /?3:do{$a=$1;s/ (\Q$1\E\s)+/ /g;print$a}while/./
			never gonna catch me, Wladimir! :)
90.46	Amir Karger	Sun Jul 7 04:36:58 2002	-ln0 A1:for$n(split){$c="\Q$n\E\\s";/^(?!$c).+ $c/m||s/^($c)?$c//gm&goto A.print$n}$_&&die
			The .02 lead seemed a bit too slim, so...
90.49	Michael Thelen	Wed Jul 3 03:47:24 2002	-ln0 s/^(.*) \1$/$1/mg;/ \Q$& /||($t=print$&)&s/^\Q$&\E\s//mgwhile/\S+/g;$t--?//&&redo:die
			Ditched the outer while loop for a 3 stroke gain. Yay!
90.52	Mtv Europe	Fri Jul 5 20:28:52 2002	-lan0 s/^(\S+) \1$/$1/mg;/ \Q$& /||print($&)&&s/^\Q$&\E[ ]//mg&&redo while/\S+/g;/./&&die
			hmm, i was surprised that it works since sometimes $& $+ $- and so on are dying so unexpectedly.
90.57	Stephen Turner	Sun Jul 7 11:47:40 2002	-lan0 for$a((@F)x y/ //){$b="\Q$a";/^(?!$b ).* $b /m||s/^$b( $b | | )//mg&&print$a}$_&&die
91.46	Amir Karger	Sun Jul 7 04:34:32 2002	-ln0 A1:for$n(split){$c="\Q$n\E\\s";/^(?!$c)\S+ $c/m||s/^($c)?$c//gm&goto A.print$n}$_&&die
			It's getting uglier and uglier, but I'm going to slip back into tenth!
91.48	Keith Calvert Ivey	Sat Jul 6 02:42:02 2002	-ln0 s/^(.+) \1$/$1/gm;s/^\Q$"\E\s//gm,print$"while$"=$_,($")=grep$"!~/ \Q$_ /,split;S/!/./
			Saved a space by changing $s to $"
91.57	Stephen Turner	Sun Jul 7 11:17:01 2002	-lan0 for$a((@F)x y/ //){$b="\Q$a";/^(?!$b ).* $b /m||s/^$b( $b | | )//mg&&print$a}/./&&die
92.37	Keith Calvert Ivey	Wed Jul 3 23:34:29 2002	-ln0 s/^(.+) \1$/$1/gm;s/^\Q$s\E\s//gm,print$s while$s=$_,($s)=grep$s!~/ \Q$_ /,split;S/!/./
			It's now possible to change the for(;;) to a while modifier. Doesn't save any strokes, but lowers the tiebreaker.
92.38	Keith Calvert Ivey	Wed Jul 3 13:16:35 2002	-ln0 for(s;^(.+) \1$;$1;gm;$s=$_,($s)=grep$s!~/ \Q$_ /,split;print$s){s}^\Q$s\E\s}}gm}S/!/./
			Moved print to save a stroke, then added gratuitous obfuscation
92.43	Amir Karger	Fri Jul 5 15:53:03 2002	-ln0 A1:for$n(split){$c="\Q$n\E";s/^$c $c$/$n/gm;/ $c\n/||s/^$c\s//gm&goto A.print$n}$_&&die
			Squeeeeeeezing...
92.47	Ala Qumsieh	Thu Jul 4 04:35:05 2002	-ln0 s/^(.*) \1$/$1/mg; while(/^\S+/mg){/^.* \Q$&\E /ms||print$&and s/^\Q$&\E\s//gm} $_&&die
			Don't ask me how I did .. it just happened!
92.50	Juho Snellman	Wed Jul 3 19:50:05 2002	-ln0 $_ x=2;{s%^(\S+)\s(?!\C*^(?!\1 ).+ \1 )%print$1;s-(^| )(\Q$1\E\s)+-$1&&die-meg;redo%em}
			This is just my 82 stroke solution with extra checks. I haven't constructed a test which the 82 doesn't pass, but I'm pretty sure there is one.
92.55	Wladimir Palant	Thu Jul 4 22:23:19 2002	-0 $_=<>;s/^(\S+) \1$/$1/mg;/ \Q$&\E$/m||print("$&\n"),s/^\Q$&\E( |$)//gmwhile/\S+/g;//&&die
92.57	Stephen Turner	Sat Jul 6 09:01:41 2002	-lan0 for$a((@F)x y/ //){$b="\Q$a";/^(?!$b ).* $b /m||s/^$b( ($b )?| )//mg&&print$a}/./&&die
93.46	Keith Calvert Ivey	Wed Jul 3 12:51:14 2002	-ln0 for(s/^(.+) \1$/$1/gm;$s=$_,($a)=grep$s!~/ \Q$_ /,/\S+/g;print$a){s/^\Q$a\E\s//gm}t/!/./
93.49	Michael Thelen	Wed Jul 3 03:39:03 2002	-ln0 s/^(.*) \1$/$1/mg;${/ \Q$& /||($t=print$&)&s/^\Q$&\E\s//mgwhile/\S+/g;$t--||die}while/./
			None of my other algorithms have panned out, so I'll just keep pushing parentheses around...
94.45	Amir Karger	Fri Jul 5 04:10:31 2002	-ln0 s/^(.+) \1$/$1/gm;A:for$n(split){$c="\Q$n\E\\s";/ $c/||s/^$c//gm&print($n)&goto A}$_&&die
			We've broken the 100 barrier! And I'm in the top 10 for what might be the first time ever.
94.46	Juho Snellman	Tue Jul 2 18:18:12 2002	-ln0 $_ x=2;{s%^(\S+)\s(?!.*((?!\1)\S+) \1 )%sqrt$;{$1}--;print$1;s|^(\Q$1\E\s)*||mg;redo%sem}
			Sandtrap-insurance.
94.48	Qingning Huo	Tue Jul 2 10:19:57 2002	-ln0 ($y=$&)=~s/\W/\\$&/g,s/^($y) $y$/$1/m,/ $y$/m||s/^($y)\s//gm&print$1while/^\S+/gm;$_&&die
			I should have found this earlier.
95.48	Qingning Huo	Tue Jul 2 05:55:26 2002	-ln0 ($y=$&)=~s/\W/\\$&/g,s/^($y) $y$/$1/m,!/ $y$/m&&s/^($y)\s//gm&print$1while/^\S+/gm;$_&&die
95.50	Michael Thelen	Wed Jul 3 02:04:33 2002	-ln0 s/^(.*) \1$/$1/mg;${/ \Q$& /||s/^(\Q$&\E)\s//mg&($t=print$1)while/\S+/g;$t--||die}while/./
			Except for the obvious improvement of removing unnecessary /m modifiers, I've hit a wall.
96.48	Qingning Huo	Tue Jul 2 05:59:18 2002	-ln0 ($y=$&)=~s/\W/\\$&/g,s/^($y) $y$/$1/gm,!/ $y$/m&&s/^($y)\s//gm&print$1while/^\S+/gm;$_&&die
			do we need /g in the second RE?
96.52	Qingning Huo	Mon Jul 1 22:44:12 2002	-ln0 ($y=$&)=~s/\W/\\$&/g,!/^(.+) (?!\1$)$y$/m&&s/^($y)\s($y )?//gm&print$1while/^\S+/gm;$_&&die
			and again.
96.59	Eugene van der Pijll	Mon Jul 8 01:55:23 2002	-lan0 sub f{1/!$g{$_}++;$p{$_}||=map(&f,$x=~/^(.*) (?!\1 )\Q$_ /mg)+print;$g{$_}=0}$x=$_;f for@F
97.47	Jukka Suomela	Fri Jul 5 21:55:27 2002	-alp ($_,my$y)=map{bless$x{$_}||=[$_]}@F;$_-$y&&push@$_,$y}DESTROY{$d?dump:print$_[0][0]}{$d=%x=q
97.49	Michael Thelen	Tue Jul 2 21:32:12 2002	-ln0 s/^(.*) \1$/$1/mg;${/ \Q$& /m||s/^(\Q$&\E)\s//mg&($t=print$1)while/\S+/mg;$t--||die}while/./
			Inching closer...
97.53	Qingning Huo	Mon Jul 1 22:41:33 2002	-ln0 ($y=$&)=~s/\W/\\$&/g,!/^(.+) (?!\1$)$y$/m&&s/^($y)\s($y )?//gm&&print$1while/^\S+/gm;$_&&die
			finally below 100.
97.55	pom	Tue Jul 2 15:55:30 2002	-0777n s/^(.+) \1$/$1/gm;while(/^\S+/gm){if(!/ \Q$&\E$/m){print"$& ";s/^\Q$&\E[ ]//gm}}/\S/&&die
			Oops, this one should work better... Still under 100!
98.48	Chris Dolan	Mon Jul 8 04:23:01 2002	-nal push@{$a{pop@F}},@F}map&a,keys%a;sub a{$z eq$_||0/$.--*&a*$.++for@{$a{$z=$_}};$p{$_}++||print
			This is just a minor improvement on my previous submission to remove the "die". It's a bit slow: takes about a minute on my 500MHz box.
98.57	Wladimir Palant	Thu Jul 4 16:39:07 2002	-0 $_=<>;s/^(\S+) \1$/$1 /mg;/ \Q$1\E$/m||print("$1\n"),s/^\Q$&\E(.*)/$1 /gmwhile/(\S+) /g;//&&die
98.58	Eugene van der Pijll	Mon Jul 8 01:46:45 2002	-lan0 sub f{1/!$g{$_}++;$p{$_}++or map(&f,$x=~/^(.*) (?!\1 )\Q$_ /mg),print;$g{$_}=0}$x=$_;f for@F
99.44	Albert Dvornik	Wed Jul 3 23:10:42 2002	-ln01 s/^(.*) \1$/$1/mg;A:while($_){for$k(split){/ \Q$k\E$/m||s/^\Q$k\E\s//mg+print($k)+next A}die}
			Less than 100!
99.53	Michael Thelen	Tue Jul 2 05:14:26 2002	-ln0 ${/^(?!\Q$&\E ).* \Q$& /m||s/^(\Q$&\E)( \1)?\s//mg&($t=print$1)while/\S+/mg;$t--||die}while/./
			Yay! Broke 100!
99.53	Juuso Salonen	Thu Jul 4 15:52:00 2002	-ln0 s// /;0//(\S+)/g,/(?<! \Q$1\E) \Q$1 /?3:do{$a=$1;s/ (\Q$1\E\s)+/ /g;print$a;/./?3:exit}while 3
100.50	Jukka Suomela	Tue Jul 2 19:20:48 2002	-ap my($x,$y)=map{$x{$_}||=bless[$_]}@F;$x-$y&&push@$x,$y}DESTROY{$0?$\.=$_[0][0].$/:dump}{%x=1;$0=0
100.57	Wladimir Palant	Thu Jul 4 15:54:37 2002	-0 $_=<>;s/^(\S+) \1$/$1 /mg;/ \Q$1\E$/m||print("$1\n"),s/^\Q$&\E(.*)/$1 /gmwhile/(\S+) /g;/\S/&&die
101.53	Qingning Huo	Mon Jul 1 18:20:44 2002	-ln0 ($y=$&)=~s/\W/\\$&/g,/$y/&!/^(.+) (?!\1$)$y$/m&&print($&)+s/^$y\s($y )?//gmwhile/^\S+/gm;$_&&die
			It's closer...
102.46	Jukka Suomela	Tue Jul 2 18:35:55 2002	-ap my($x,$y)=map{$s{$_}||=bless[$_]}@F;$x==$y||push@$x,$y}DESTROY{$\.=$_[0][0].$/;$d&&dump}{%s=0;$d=1
102.50	Keith Calvert Ivey	Wed Jul 3 12:37:21 2002	-ln0 for(s/^(.+) \1$/$1 /gm;$s=$_,($a)=grep$s!~/ \Q$_ /,/\S+/g;){s/^\Q$a\E (.*)/$1 /gm;print$a}t/!/\S/
102.53	Qingning Huo	Mon Jul 1 18:03:47 2002	-ln0 ($y=$&)=~s/\W/\\$&/g,/$y/&!/^(\S+) (?!\1$)$y$/m&&print($&)&s/^$y\s($y )?//gmwhile/^\S+/gm;$_&&die
			can I pass 100?
102.53	Michael Thelen	Tue Jul 2 05:01:49 2002	-ln0 ${/^(?!\Q$&\E ).* \Q$&\E$/m||s/^(\Q$&\E)( \1)?\s//mg&&($t=print$1)while/\S+/mg;$t--||die}while/./
			More small improvements, hoping for the big one...
103.48	Chris Dolan	Sun Jul 7 21:57:42 2002	-nal push@{$a{pop@F}},@F}map&a,keys%a;sub a{$z eq$_||($.--||die,&a,$.++)for@{$a{$z=$_}};$o{$_}++||print
			Gained my stroke back
103.50	Chris Dolan	Sun Jul 7 21:42:56 2002	-nal push@{$a{pop@F}},@F}&a for%a;sub a{$z eq$_||($.--||die,&a,$.++)for@{$a{$z=$_}};$o{$_}++.ref||print
104.48	Chris Dolan	Sun Jul 7 21:55:35 2002	-nal push@{$a{pop@F}},@F}&a for keys%a;sub a{$z eq$_||($.--||die,&a,$.++)for@{$a{$z=$_}};$o{$_}++||print
			This is one stroke longer than my best, but it runs 30% faster, so maybe the refs will like it better...
105.48	Chris Dolan	Sun Jul 7 21:09:37 2002	-nal push@{$a{pop@F}},@F}sub a{$.--<0&&die;$z eq$_||&a for@{$a{$z=$_}};$.++;$o{$_}++||print}for(keys%a){a
105.53	Mtv Europe	Fri Jul 5 19:05:40 2002	-ln0 s/^(\S+) \1$/$1/mg;$a=$_;A:die if map$a!~/ \Q$_\E /&&print&&$a=~s/^\Q$_\E[ ]//mg&&goto A,$a=~/\S+/g
			my first attempt, not sure you will accept it since i use windows platform :)
106.47	Lars Mathiesen	Tue Jul 2 14:57:24 2002	$_=join'',<>;s/^(.*) \1$/$1/mg;s/^\Q$y\E\s//mg,print"$y "while$x=$_,($y)=grep$x!~/ \Q$_\E /,split;/ /&&die
			Grotty \Q\E, grotty fix for reflexives. Room for improvement...
106.50	Qingning Huo	Mon Jul 1 17:36:37 2002	-ln0 while(/^\S+/gm){($y=$&)=~s/\W/\\$&/g;s/^$y\s($y\s)?//gm if/$y/&!/^(\S+) (?!\1$)$y$/m&&print$&}$_&&die
			Hope this works.
106.54	Michael Thelen	Tue Jul 2 04:51:07 2002	-ln0 ${$t=0;/^(?!\Q$&\E ).* \Q$&\E$/m||s/^(\Q$&\E)( \1)?\s//mg&&++$t&&print$1while/\S+/mg;$t||die}while/./
			Got rid of a few parens and braces... now to get rid of $t.
107.47	Juho Snellman	Mon Jul 1 18:27:53 2002	-ln0 s/^(.+) \1$/$1/gm;$_.="\t $_";s/^\Q$1\E\s//mgwhile s/^(\S+)\s(?!.* \1 )(.*\t)/print$1;$2/mse;/\S/&&die
			Ah, I should have remembered that regexes always win... :-)
108.47	Amir Karger	Fri Jul 5 02:29:40 2002	-n0 s/^(\S+) \1$/$1/gm; A:for$n (split){ if(!/ \Q$n\E\n/){ s/^\Q$n\E\s//gm; print"$n\n"; goto A } } $_&&die
			Woo-hoo! Still got more to do...
108.49	David Lowe	Wed Jul 3 23:34:47 2002	-lna @w{@F}=@F;push@g,$_}{&%if map{print,delete$w{$w},redo if$w=$_,!grep/ \Q$w\E$/&&$`ne$w&&$w{$`}ne'',@g}%w
108.53	Juuso Salonen	Thu Jul 4 15:09:57 2002	-ln0 s// /;/(\S+)/g,($a=$1)=~//?/(?<! \Q$a\E) \Q$a /?3:do{s/ (\Q$a\E\s)+/ /g;print$a;/./?3:exit}:3/$/while+3
108.55	Sec	Wed Jul 3 09:45:31 2002	-lap @t{@q{@F}=@F}.=" $F[0] "x!/^\Q$F[1] /}for(keys%q){1/(($_)=grep!$t{$_},%q);$t{$w=$_}+=map{s/ \Q$w //g}%t
			Who had thought that even "die" could be made shorter ;-)
108.56	Eugene van der Pijll	Mon Jul 8 01:39:16 2002	-lan0 sub f{$g{$_}++&&&g;$p{$_}++or map(&f,$x=~/(.*) \Q$_ /g),print;$g{$_}=0}s/^(.*) \1$/$1/mg;$x=$_;f for@F
109.59	Wladimir Palant	Thu Jul 4 15:51:03 2002	-0 $_=<>;s/^(\S+) \1$/$1 /mg;/ \Q$1\E$/m||$a{$1}++||print("$1\n"),s/^\Q$&\E(.*)/$1 /mwhile/(\S+) /g;/\S/&&die
110.49	Keith Calvert Ivey	Wed Jul 3 12:33:21 2002	-p0 for(s/^(.+) \1$/$1 /gm;$s=$_,($a)=grep$s!~/ \Q$_ /,/(.+) /g;){s/^\Q$a\E (.*)/$1 /gm;s/\z/$a /}s/\s+///!/ /
110.49	David Lowe	Wed Jul 3 23:25:43 2002	-lna @w{@F}=@F;push@g,"@F"}{&%if map{print,delete$w{$w},redo if$w=$_,!grep/ \Q$w\E$/&&$`ne$w&&$w{$`}ne'',@g}%w
			pass the test, dammit
110.53	Michael Thelen	Tue Jul 2 04:35:57 2002	-ln0 ${$t=0;for$m(/^\S+/mg){/^(?!\Q$m\E ).* \Q$m\E$/m||s/^(\Q$m\E)( \1)?\s//mg&&++$t&&print$m}$t||die}while/./
111.49	Chris Dolan	Thu Jul 4 05:08:55 2002	-nal push@{$a{pop@F}},@F}sub a{$c{$_}++&&die;$z eq$_||&a for@{$a{$z=$_}};$c{$_}--;$o{$_}++||print}for(keys%a){a
			Make heavy use of default var. Yahoo!
111.51	Juuso Salonen	Wed Jul 3 16:59:43 2002	-ln0 s// /;/(\S+)/g,($a=$1)=~//?/(?<! \Q$a\E) \Q$a /?0:do{while(s/^\Q$a\E\s//m){}print$a;//?3:exit}:3/$/while+3
111.53	Michael Thelen	Tue Jul 2 03:50:57 2002	-ln0 while(/./){$t=0;for$m(/^\S+/mg){/^(?!\Q$m\E ).* \Q$m\E$/m||s/^(\Q$m\E)( \1)?\s//mg&&++$t&&print$m}$t||die}
			I think a regex is the way to go for this one. But it's dang hard to construct! (This isn't it:-)
111.54	Honza Pazdziora	Thu Jul 4 09:01:07 2002	-n0 s/\S+$/$& $& $&/gm;X:for$x(/\S+/g){/^(?!\Q$x\E ).+ \Q$x /m or s/^\Q$x\E .*$//mg,(print"$x "),goto X}/./&&&g
112.56	Terje K	Sun Jul 7 04:23:55 2002	-nl $h{_}{$'}=/ /;$h{$'}{$`}=1if$`ne$'}{sub z{$u{$_}++&&die,$y{$_}||=z($_)+print,$u{$_}--for keys%{$h{+pop}}}z _
113.54	Eugene van der Pijll	Mon Jul 8 00:12:57 2002	-lap0 $_ x=@F;s#(.*) (.*)#$g{$2}>($g{$1}x=1)||log@F-($g{$2}=$g{$1}+($1ne$2))#ge}for(sort{$g{$a}<=>$g{$b}}keys%g){
113.56	Eugene van der Pijll	Mon Jul 8 01:25:42 2002	-lan0 sub f{$g{$_}++&&&g;if(!$p{$_}++){&f for $x=~/(.*) \Q$_\E\n/g;print}$g{$_}=0}s/^(.*) \1$/$1/mg;$x=$_;f for@F
			Eureka!
113.56	Terje K	Sat Jul 6 18:42:38 2002	-nl $h{$;}{$'}=/ /;$h{$'}{$`}=1if$`ne$'}{sub z{$u{$_}++&&die,$y{$_}++||z($_)&print,$u{$_}--for keys%{$h{+pop}}}z$
114.51	Keith Calvert Ivey	Wed Jul 3 12:04:07 2002	-p0 for(s/ (.+)/$& $1 /g,s/^(.+) \1 //gm;$s=$_,($a)=grep$s!~/ \Q$_ /,/(.+) /g;){s/^\Q$a\E .* //gm;s/\z/$a /}t/!/ /
			Saved a stroke with for(;;) in place of while().
114.53	Eugene van der Pijll	Mon Jul 8 00:04:59 2002	-lap0 $_ x=@F;s#(.*) (.*)#$g{$2}>($g{$1}x=1)or@F>($g{$2}=$g{$1}+($1ne$2))or&f#ge}for(sort{$g{$a}<=>$g{$b}}keys%g){
114.53	Jasper McCrea	Fri Jul 5 14:54:21 2002	-lp0 @e{@a=split}=1;$o=join$\,keys%e;1/!eval'grep/ /&$o=~s#^(\Q$\'\E)$((.| )* )(\Q$`\E)$#$4$2$1#m,/.*/g;'x@a;$_=$o
			mmm.. bacon..
114.54	Eugene van der Pijll	Sun Jul 7 23:22:12 2002	-lap0 $_ x=@F;s#(.*) (.*)#$g{$2}>($g{$1}x=1)or($g{$2}=$g{$1}+($1ne$2))<@F||&f#ge}for(sort{$g{$a}<=>$g{$b}}keys%g){
114.55	Stephen Turner	Fri Jul 5 22:01:07 2002	-ln0 map{for$a(/\S+/gm){$b="\Q$a";/^(?!$b ).* $b$/m||s/^$b(?!\S) ?//mg&&print($a)&s/^$b$//mg}}($_)x y/ //;/./&&die
			Down to only 1.75 * Ton's score.
114.56	Terje K	Sat Jul 6 15:05:41 2002	-nl $h{$"}{$'}=/ /;$h{$'}{$`}=1if$`ne$'}{sub z{$u{$_}++&&die,$y{$_}++||z($_)&print,$u{$_}--for keys%{$h{+pop}}}z$"
115.51	Keith Calvert Ivey	Wed Jul 3 03:46:12 2002	-p0 s/ (.+)/$& $1 /g;s/^(.+) \1 //gm;while($s=$_,($a)=grep$s!~/ \Q$_ /,/(.+) /g){s/^\Q$a\E .* //gm;s/\z/$a /}t/!/ /
			Found a better way of dying.
115.54	Sec	Tue Jul 2 20:26:34 2002	-lap @t{@F}.=" $F[0] "if$F[0]ne$F[1];@q{@F}=@F}for(keys%q){(($_)=grep!$t{$_},%q)||die;$t{$,=$_}+=map{s/ \Q$, //g}%t
			I tried using grep a few times already. But suddenly it gave a gain %)
115.57	Terje K	Sat Jul 6 14:53:38 2002	-nl $h{$"}{$'}=/ /;$h{$'}{$`}=1if$`ne$'}{sub z{map{$u{$_}++&&die,$y{$_}++||z($_)&print,$u{$_}--}keys%{$h{+pop}}}z$"
115.65	Terje K	Sat Jul 6 13:02:56 2002	-nl $h{$'}{$"}=/ /;$h{$`}{$'}=$`ne$'}{sub z{map$h{$_}{$_[0]}&&($u{$_}++&&die,$y{$_}++||z($_)&print,$u{$_}--),%h}z$"
116.55	Honza Pazdziora	Wed Jul 3 14:53:11 2002	-n0 s/\S+$/$& $& $&/gm;X:for$x(/(.+) /mg){/^(?!\Q$x\E ).+ \Q$x\E$/m or s/^\Q$x\E .*$//mg,print("$x "),goto X}/./&&&g
117.49	Amir Karger	Wed Jul 3 20:49:54 2002	-nal $p{$l}{$r}=($l,$r)=@F;$p{$r}{$r}=0}{sub z{keys%p}for(z){for$%(z){delete$p{$%},print$%if!grep$p{$_}{$%},z}}z&&die
			Not good for my tiebreak score, but...
117.50	Keith Calvert Ivey	Wed Jul 3 03:42:37 2002	-p0 s/ (.+)/$& $1 /g;s/^(.+) \1 //gm;while($s=$_,($a)=grep$s!~/ \Q$_ /,/(.+) /g){s/^\Q$a\E .* //gm;s/\z/$a /}/ /&&die
117.57	Andrew Savige	Mon Jul 8 04:17:01 2002	-aln / /,@n{@F}+=$`ne$'&!$p{$_}++}{0-$n{$x=$_}or$z+=print,map{/ /,$n{$'}-=$`eq$x}%p,"$x $x"for(@m=keys%n)x@m;$z-@m&&&
			I am sooooo close to a break-through new algorithm. What do you mean, there's only half an hour left? I wonder if `/ has a stroke or two in reserve.
117.65	Terje K	Sat Jul 6 13:00:12 2002	-nl / /;$h{$'}{$"}=1;$h{$`}{$'}=$`ne$'}{sub z{map$h{$_}{$_[0]}&&($u{$_}++&&die,$y{$_}++||z($_)&print,$u{$_}--),%h}z$"
118.47	Juho Snellman	Mon Jul 1 15:53:49 2002	@a=<>;{map{s/^(.+) \1$/$1/;if((($a)=split)&&!grep/ \Q$a\E$/,@a){print"$a ";s/^\Q$a\E\s//for@a;redo}}@a}map/\S/&&die,@a
			I'm afraid that this isn't how Ton did his 88, but at least it takes me to unshared second place... Well, back to the drawing board.
118.48	Chris Dolan	Thu Jul 4 04:45:39 2002	-nal push@{$a{$F[1]}},@F}sub a{$c{my$z=pop}++&&die;$z eq$_||a($_)for@{$a{$z}};$c{$z}--;$o{$z}++||print}for(keys%a){a$_
			Lucky bonus: last item in for is self, so $_ is already set (saves me two strokes)
118.55	Yanick Champoux	Sun Jul 7 16:45:28 2002	-lpa push@;,[@F[$F[0]eq$_..1]]for@F}while(/.+/,map{@$_=grep$&ne$_,@$_}@;){($_)=map{(($x)=@$_)x!grep$$_[1]eq$x,@;}@;or&
			Would that dirty trick be my final swing at this hole?
118.57	Andrew Savige	Sun Jul 7 23:15:52 2002	-aln / /,@n{@F}+=$`ne$'&!$p{$_}++}{ 0-$n{$x=$_}or$z+=print,map{/ /,$n{$'}-=$`eq$x}%p,"$x $x"for(@m=keys%n)x@m;$z-@m&&&
			Caught `/ at last! Will he respond though?
118.59	Wladimir Palant	Thu Jul 4 15:21:33 2002	-0 $_=<>;s/^(\S+) \1$/$1 /mg;while(/^(\S+) /mg){$a{$1}++||print("$1\n"),s/^\Q$1\E (.*)/$1 /m if!/ \Q$1\E$/m};/\S/&&die
119.54	Yanick Champoux	Sat Jul 6 03:13:33 2002	-lpa push@a,[@F[$F[0]eq$_..1]]for@F}while(/.+/,map{@$_=grep$&ne$_,@$_}@a){($_)=map{(($x)=@$_)x!grep$$_[1]eq$x,@a}@a or&
			Woohoo! Sub-120!
119.55	Honza Pazdziora	Wed Jul 3 14:00:51 2002	-n0 s/\S+$/$& $& $&/gm;X:{for$x(/^(.+) /mg){/^(?!\Q$x\E ).+ \Q$x\E$/m or s/^\Q$x\E .*$//mg,print("$x "),redo X}/./&&&g}
120.48	Chris Dolan	Wed Jul 3 21:51:06 2002	-nal sub a{$c{my$z=pop}++&&die;$z eq$_||a($_)for@{$a{$z}};$c{$z}--;$o{$z}++||print$z}push@{$a{$F[1]}},@F}for(keys%a){a$_
			Oops, extra copy-n-paste space
120.48	Amir Karger	Wed Jul 3 19:50:48 2002	-nal $p{$l}{$r}=($l,$r)=@F;$p{$r}{$r}=0}{sub z{keys%p}for(z){for$n(z){grep{$p{$_}{$n}}z or delete$p{$n},print$n}}%p&&die
			Probably my limit for a while. I removed \n's for a "true" best score.
121.48	Chris Dolan	Wed Jul 3 21:50:43 2002	-nal sub a{$c{my$z=pop}++&&die;$z eq$_||a($_)for@{$a{$z}};$c{$z}--;$o{$z}++||print$z} push@{$a{$F[1]}},@F}for(keys%a){a$_
			Shaving...
121.55	Keith Calvert Ivey	Tue Jul 2 13:09:05 2002	-p0 s/ (.+)/$& $1 /g;s/^(.+) \1 //gm;while(%f=/ (.+)( )/g,($a)=grep!$f{$_},/(.+) /g){s/^\Q$a\E .* //gm;s/\z/$a /}/ /&&die
121.55	Eugene van der Pijll	Sun Jul 7 23:12:26 2002	-lap0 @g{@F}=0;$_ x=@o=keys%g;s#(.*) (.*)#$g{$2}>$g{$1}or($g{$2}=$g{$1}+($1ne$2))<@o||&f#ge}for(sort{$g{$a}<=>$g{$b}}@o){
121.56	Andrew Savige	Sun Jul 7 22:54:33 2002	-aln @n{@F}+=!s/^(.+) \1$//&!$p{$_}++}{0-$n{$x=$_}or$z+=print,map{/ /,$n{$'}-=$`eq$x}%p,"$x $x"for(@m=keys%n)x@m;$z-@m&&&
			Still fighting! Puff! Puff! I've got to get into shape for these vigorous sprints to the finish line. Hope `/ is worried.
121.66	Terje K	Sat Jul 6 12:54:01 2002	-nl /(.+) /;$h{$'}{$"}=1;$h{$1}{$'}=$1ne$'}{sub z{map$h{$_}{$_[0]}&&($u{$_}++&&die,$y{$_}++||z($_)&print,$u{$_}--),%h}z$"
122.51	Sec	Tue Jul 2 19:41:02 2002	-lan @t{@F}.=" $F[0] "if$F[0]ne$F[1];@q{@F}=@F}for(keys%q){map$t{$_}||do{$t{$,=$_}+=print;map{s/ \Q$, //g}%t;next},%q;die;
			My last solution is actually invalid according to the rules (see seperate mail for a test case). Fix it.
122.52	Scott Hall	Sun Jul 7 17:55:23 2002	-nl / /;$f{$_}++||push@{$a{$'}},$`eq$'?():$`}sub v{$b{$_}++?die:&v,--$b{$_}for@{$a{$_}};$p{$_}++||print}for(keys%a){my%b;v
122.55	Keith Calvert Ivey	Tue Jul 2 12:06:40 2002	-p0 s/ (.+) /$&$1 /g;s/^(.+) \1 //gm;while(%f=/ (.+)( )/g,($a)=grep!$f{$_},/(.+) /g){s/^\Q$a\E .* //gm;s/\z/$a /}/ /&&die
			Had to change $ to \z to handle the new case with just "a a".
122.58	Jasper McCrea	Thu Jul 4 15:05:32 2002	-lp0 @e{/\S+/g}=1;$o=join" ",keys%e;1/!eval'grep/(\S+) (\S+)/&$o=~s/^(\Q$2\E)$((.| )*)^(\Q$1\E)$/$4$2$1/m,/.*/g;'x99;$_=$o
			passes on 1.8 for me, anyway. Of course, this does not assure that the solution is valid :)
122.60	Wladimir Palant	Thu Jul 4 15:16:31 2002	-0 $_=<>;s/^(\S+) \1$/$1 /mg;while(/^(\S+) (.*)/mg){$a{$1}++||print("$1\n"),s/^\Q$1\E (.*)/$1 /m if!/ \Q$1\E$/m};/\S/&&die
122.66	Terje K	Fri Jul 5 08:11:15 2002	-nl /(.+) /;$h{$'}{$"}=1;$h{$1}{$'}=$1ne$'}{sub z{map$h{$_}{$_[0]}&&($u{$_}++&&die,$y{$_}++||z($_)&&print,$u{$_}--),%h}z$"
123.50	Jonathan Stimmel	Mon Jul 8 00:54:24 2002	-ln / /;$`ne$'?$r{$'}{$`}:$x=$r{$`}||={}}while(@n=grep!%{$r{$_}}&&print,keys%r or%r&&exit 1){map delete@$_{@n},\%r,values%r
123.52	Wladimir Palant	Wed Jul 3 23:59:05 2002	-0 $_=<>;@a=/\S+/g;s/^(\S+) \1$//mg;for$a(@a){s/^\Q$a\E (.+)/push@a,$1;''/mge,$a{$a}++||print"$a\n"if!/ \Q$a\E$/m}/\S/&&die
123.53	Jasper McCrea	Fri Jul 5 11:47:56 2002	-lp0 @e{@a=split}=1;$o=join$\,keys%e;1/!eval'grep/(\S+) (\S+)/&$o=~s/^(\Q$2\E)$((.| )*)^(\Q$1\E)$/$4$2$1/m,/.*/g;'x@a;$_=$o
			Although a baconiser does sound like a good idea.
123.57	Andrew Savige	Sun Jul 7 04:27:16 2002	-aln @n{@F}+=!s/^(.+) \1$//&!$:{$_}++}{0-$n{$x=$_}or++$z,print,map/ /&$`eq$x&&--$n{$'},%:,"$x $x"for(@m=keys%n)x@m;$z-@m&&&
			This is a really ugly solution. I hope, however, it makes `/ just a wee bit scared. :-)
123.57	Jasper McCrea	Wed Jul 3 14:45:33 2002	-lp0 @e{/\S+/g}=1;$o=join" ",keys%e;$_=eval'grep/(\S+) (\S+)/&$o=~s/^(\Q$2\E)$((.| )*)^(\Q$1\E)$/$4$2$1/m,/.*/g;'x99?die:$o
			hurrah! I join the ranks of those whose score is less than twice Ton's!
123.60	Aj	Mon Jul 8 04:59:27 2002	-l map{/ (.*)/,($o{$`}+=0)<$o{$1}or$`eq$1or$o{$1}=$o{$`}+1,o/($o{$1}<$.)}(<>)x++$.;for$i(0..$.){map$o{$_}^$i||print,keys%o}
124.49	Chris Dolan	Wed Jul 3 21:40:28 2002	-nal sub a{$c{my$z=pop}++&&die;$z eq$_||a($_)for@{$a{$z}};$c{$z}--;$o{$z}++||print$z}push@{$a{$F[1]}},@F}map{%c=a$_}keys%a;{
			Breakthrough! At last!!!
124.52	Wladimir Palant	Wed Jul 3 23:20:06 2002	-0 $_=<>;@a=/\S+/mg;s/^(\S+) \1$//mg;for$a(@a){s/^\Q$a\E (.+)/push@a,$1;''/mge,$a{$a}++||print"$a\n"if!/ \Q$a\E$/m}/\S/&&die
124.56	Stephen Turner	Wed Jul 3 20:59:54 2002	-l map{(grep{/ \Q$a\E$/&!/^\Q$a /}@0)||(print$a)&&map{s/^\Q$a //;s/^\Q$a\E$//}@0if($a)=/(\S+)/}@0for@0=<>,@0;"@0"=~/\S/&&die
			Completely different method, but not much better.
124.56	Eugene van der Pijll	Tue Jul 2 01:24:36 2002	-lap0 @g{@F}=a;$_ x=@o=keys%g;$2eq$1||$g{$2}>$g{$1}||a/(($g{$2}=$g{$1}+1)<@o)while/(.*) (.*)/g}for(sort{$g{$a}<=>$g{$b}}@o){
124.57	Jasper McCrea	Thu Jul 4 15:43:40 2002	-lp0 @e{/\S+/g}=1;$o=join$\,@l=keys%e;1/!eval'grep/(\S+) (\S+)/&$o=~s/^(\Q$2\E)$((.| )*)^(\Q$1\E)$/$4$2$1/m,/.*/g;'x@l;$_=$o
			curse you referees, and your insane love of non-hard-coded limits! p.s. I am not convinced that 'baconise' is a word.
125.52	Honza Pazdziora	Wed Jul 3 12:05:52 2002	-alp ($,,$b)=@F;$l{$b}||=[];$,ne$b&&push@{$l{$,}},$b}while(%l){(($_)=grep{$*=$_;!grep$*eq$_,map@$_,%l}keys%l)?delete$l{$_}:&g
125.52	Jon Coppeard	Sat Jul 6 22:07:45 2002	-n / /;$p{$a=$`.$/}+=0;$' eq$a or$p{$'}++,push@$a,$'}{delete$p{$_},print,map--$p{$_},@$_ while($_)=grep!$p{$_},keys%p;%p&&die
125.53	Scott Hall	Sun Jul 7 14:08:09 2002	-nl / /;$f{$_}++||push@{$a{$'}},$`eq$'?():$`}sub v{$b{$_}++?die:v($_),--$b{$_}for@{$a{$_}};$p{$_}++||print}for(keys%a){my%b;v
125.53	Wladimir Palant	Wed Jul 3 22:52:53 2002	-0 $_=<>;@a=/\S+/mg;s/^(\S+) \1$//mg;for$a(@a){/ \Q$a\E$/m||(s/^\Q$a\E (.+)/push@a,$1;''/mge,$a{$a}++||print"$a\n")}/\S/&&die
126.45	David Lowe	Mon Jul 1 21:58:20 2002	-lna push@g,[@w{@F}=@F]}{(map{print(delete$w{$w}),redo if$w=$_,!grep{$$_[0]ne$w&&exists$w{$$_[0]}&&$w eq$$_[1]}@g}keys%w)&&die
126.47	Wladimir Palant	Wed Jul 3 17:26:59 2002	-nal push@a,[@F]if$F[0]ne$F[1];@a{@F}=(1)x2}{a while delete@a{grep{$a=$_;print if!grep$$_[1]eq$a&$a{$$_[0]},@a}keys%a};%a&&die
126.48	Juho Snellman	Mon Jul 1 15:11:55 2002	@a=<>;{map{s/^(.+) \1$/$1/;($a,$b)=split;if(/\S/&&!grep/ \Q$a\E$/,@a){ print"$a ";s/^\Q$a\E\s//for@a;redo}}@a}map{/\S/&&die}@a
			Regexp metacharacters should work now.
126.48	pom	Mon Jul 1 12:03:47 2002	-ln / /;++$b{$'}{$`}if$`ne$';$b{$`}||={}}{map delete@{%$_}{@d},values%b while delete@b{@d=grep!%{$b{$_}}&&print,keys%b};exit%b
			Again, by a hair!
126.49	Michael Thelen	Mon Jul 1 17:09:32 2002	-ln / /,$`ne$'&&$b{$'}{$`}++;$b{$`}{0}}while(%b){delete@b{@k=grep!%{$b{$_}}&&print,keys%b}or die;delete@{$b{$_}}{@k}for keys%b
126.49	Albert Dvornik	Fri Jul 5 21:17:30 2002	-ln0 $l=$_;sub X{my$x=$_;$v{$x}++&&return;@r=map$_ eq$x?():X(),$l=~/^(.*) \Q$x\E$/mg;$v{$x}>1&&die;@r,$x}print for map X,split
			A brand new search-based approach. Can this get me closer to The Amazing Mr. Hospel? >=)
127.50	Honza Pazdziora	Wed Jul 3 11:23:32 2002	-alp ($a,$b)=@F;$l{$b}||=[];push@{$l{$a}},$b if$a ne$b}while(%l){(($_)=grep{$*=$_;!grep$*eq$_,map@$_,%l}keys%l)?delete$l{$_}:&g
127.52	Martin Carlsen	Tue Jul 2 08:25:06 2002	-p0 $n++>8*y///c&&die while s/^((\S+) \S+ )(.*^(?!\2 )\S+ \2 )/$3$1/sm||s/( \S+)( .*\1 )/$2/s||s/ / /||s/^(\S+ )(.*^)\1/$1$2/ms
127.53	Geoff Hubbard	Mon Jul 8 01:35:27 2002	-ln $r{$1}{$`}=/ (.*)/}{sub v{$c{$"=$_}=2;$_ ne$"and$c{$_}?$c{$_}^2||die:&v for keys%{$r{$_}};$c{$_}=print}$c{$_}||v for keys%r
128.44	David Lowe	Mon Jul 1 21:30:08 2002	-lna push@g,[@w{@F}=@F]}O:{for$w(keys%w){print(delete$w{$w}),redo O if!grep{$$_[0]ne$w&&exists$w{$$_[0]}&&$w eq$$_[1]}@g}%w&&die
128.48	Martin Carlsen	Mon Jul 1 18:44:55 2002	-nl / /;\$p{$`};$`eq$'||++$p{$'}^push@{$s{$`}},$'}{1while map{delete$p{$_};--$p{$_}for@{$s{$_}};print}grep!$p{$_},keys%p;%p&&die
128.53	Andrew Savige	Sun Jul 7 02:03:51 2002	-aln @n{@F}+=!s/^(.+) \1$//&!$:{$_}++}{@l=grep!$n{$_},@m=keys%n;//,print,push@l,grep!--$n{$_},map/^\Q$'\E (.+)/,%:for@l;@l-@m&&&
			Alas, still on the same horse. I've tried a couple of different approaches, but could not make them shorter. :-( Still, as is often the case, trying a different approach often yields a tactical trick that can then be applied to a previous solution.
128.61	Aj	Sun Jul 7 19:09:39 2002	-l map{/ (.*)/,($o{$`}+=0)<($o{$1}+=0)or$`eq$1or$o{$1}=$o{$`}+1,o/($o{$1}<$.)}(<>)x++$.;for$i(0..$.){map$o{$_}^$i||print,keys%o}
128.62	Terje K	Fri Jul 5 05:31:30 2002	-nl /(.+) /;$h{$'}{$"}=1;$h{$1}{$'}=$1ne$'}{sub z{my$x=pop;map$h{$_}{$x}&&{$u{$_}++&&die,$y{$_}++||z($_)&&print,$u{$_}--},%h}z$"
129.48	Wladimir Palant	Wed Jul 3 16:27:04 2002	-nal push@a,[@F]if$F[0]ne$F[1];@a{@F}=(1,1)}{a while grep{$a=$_;delete$a{$a},print if!grep$$_[1]eq$a&$a{$$_[0]},@a}keys%a;%a&&die
129.49	Martin Carlsen	Mon Jul 1 17:28:43 2002	-nl / /;\$p{$`};$`eq$'||++$p{$'}&&push@{$s{$`}},$'}{1while map{delete$p{$_};--$p{$_}for@{$s{$_}};print}grep!$p{$_},keys%p;%p&&die
129.51	Jon Coppeard	Tue Jul 2 21:09:09 2002	-n / /;$p{$a=$`.$/}+=0;$a ne$'&&$p{$'}++,push@$a,$'}{@n=grep++$n&&!$p{$_},keys%p;--$n,print,push@n,grep!--$p{$_},@$_ for@n;exit$n
129.52	Wladimir Palant	Wed Jul 3 22:10:38 2002	-0 $_=<>;@a=/^\S+/mg;s/^(\S+) \1$//mg;for$a(@a){/ \Q$a\E$/m&&next;$a{$a}++||print"$a\n";s/^\Q$a\E (.+)/push@a,$1;''/mge}/\S/&&die
130.44	Albert Dvornik	Wed Jul 3 20:23:46 2002	-l @l=<>;A:while(grep$_,@l){for$k(map split,@l){!grep/ \Q$k\E$/&&!s/^\Q$k\E //,@l and(map(s/^\Q$k\E\s//,@l),print($k),next A)}die}
130.47	TheodoreYoung	Fri Jul 5 16:38:26 2002	-lan END{map{print;delete$C{$_}}grep{!grep{$C{$_}}keys%{$C{$_}}}keys%C or die while%C}($l,$r)=@F;$C{$l}{\n}=$l ne$r?$C{$r}{$l}=1:1
130.48	Andrew Savige	Sun Jul 7 01:54:53 2002	-aln / /;$p{$_}++or@n{@F}+=$`ne$'&&!!push@{$s{$`}},$'}{@l=grep!$n{$_},@m=keys%n;print,push@l,grep!--$n{$_},@{$s{$_}}for@l;@l-@m&&&
			Don't laugh! Not only does it have a better tie-breaker, it is faster too!
130.49	Stephen Turner	Wed Jul 3 20:59:23 2002	-nla ($",$b)=@F;push@{$v{$"}},$;;$"eq$b||push@{$v{$b}},$"}{map{map{print,delete$v{$_}if!grep{$v{$_}}@{$v{$_}}}keys%v}0..$.;%v&&die
			Not very good, but I like the expression grep{$v{$_}}@{$v{$_}}
130.52	Peter Haworth	Fri Jul 5 18:15:09 2002	-nla push@v,($,,$r)=@F;$,eq$r||push@{$e{$r}},$,}sub v{$v{$_}=2;$v{$_}?$v{$_}&2&&die:&v for@{$e{$_}};$v{$_}=print}{$v{$_}||&v for@v
130.53	Andrew Savige	Sat Jul 6 05:37:24 2002	-aln @n{@F}+=!s/^(.+) \1$//&!$:{$_}++}{@l=grep!$n{$_},@m=keys%n;/.+/,print,push@l,grep!--$n{$_},map/^\Q$&\E (.+)/,%:for@l;@l-@m&&&
130.54	Yanick Champoux	Thu Jul 4 01:04:55 2002	-lpa push@a,[@F[$_..1]]for$F[0]eq$F[1],1}while(@a=grep!grep($x eq$_,@$_),@a){$_=(($x)=map{($x=@$_[0])x!grep$$_[1]eq$x,@a}@a)?$x:&:
			BWAH! I had forgotten about this über-dirty trick!
130.58	Aj	Sun Jul 7 13:52:36 2002	-l $f=0,map{/ (.*)/,($o{$`}+=0)<($o{$1}+=0)or$`eq$1or$f=$o{$1}=$o{$`}+1}@0for(@0=<>,$,=$/);1/!$f,print sort{$o{$a}<=>$o{$b}}keys%o
131.49	Michael Thelen	Mon Jul 1 15:34:55 2002	-ln / /,$`ne$'&&$b{$'}{$`}++;$b{$`}{0}}while(%b){delete@b{@k=grep!%{$b{$_}},keys%b}or die;delete@{$b{$_}}{@k}for keys%b;print for@k
			More minor improvements... I need something less cumbersome than a HoH, methinks.
131.50	pom	Mon Jul 1 11:08:13 2002	-ln / /;++$b{$'}{$`}if$`ne$';$b{$`}||={}}{while(delete@b{@d=grep{!%{$b{$_}}&&print}keys%b}){delete@{%{$b{$_}}}{@d}for keys%b}exit%b
			By a hair...
131.50	Jon Coppeard	Tue Jul 2 15:51:43 2002	-n / /;$p{$'}++,push@$a,$' if($a=$`.$/)ne$';$p{$a}+=0}{@n=grep++$n&&!$p{$_},keys%p;--$n,print,push@n,grep!--$p{$_},@$_ for@n;exit$n
131.51	Jonathan Stimmel	Sun Jul 7 23:48:53 2002	-ln / /;$`ne$'?$r{$'}{$`}:$_=$r{$`}||={}}$,=$\;while(@n=grep!%{$r{$_}},keys%r or exit(%r>0)){map delete@$_{@n},\%r,values%r;print@n
131.53	Yanick Champoux	Thu Jul 4 01:03:14 2002	-lpa push@a,[@F[$_..1]]for$F[0]eq$F[1],1}while(@a=grep!grep($x eq$_,@$_),@a){$_=(($x)=map{($x=@$_[0])x!grep$$_[1]eq$x,@a}@a)?$x:die
			Did I mentioned already how hilarious I find that grep!grep construct?
131.54	Geoff Hubbard	Sun Jul 7 23:53:31 2002	-ln $r{$2}{$1}=/(.*) (.*)/}{sub v{$c{$"=$_}=2;$_ ne$"and$c{$_}?$c{$_}^2||die:&v for keys%{$r{$_}};$c{$_}=print}$c{$_}||v for keys%r
131.62	Terje K	Thu Jul 4 22:25:31 2002	-nl /(.+) /;$h{$'}{$"}=1;$h{$1}{$'}=1if$1ne$'}{sub z{my$x=pop;map$h{$_}{$x}&&{$u{$_}++&&die,$y{$_}++||z($_)&&print,$u{$_}--},%h}z$"
132.45	David Lowe	Mon Jul 1 21:25:23 2002	-lna push@g,[@w{@F}=@F]}O:{for$w(keys%w){print(delete$w{$w}),redo O if!grep{$$_[0]ne$$_[1]&&exists$w{$$_[0]}&&$w eq$$_[1]}@g}%w&&die
132.51	Honza Pazdziora	Tue Jul 2 16:36:35 2002	-alp $l{$F[1]}||=[];push@{$l{$F[0]}},$F[1]if$F[0]ne$F[1]}while(%l){($_)=grep{$*=$_;!grep{$*eq$_}map@$_,%l}keys%l or die;delete$l{$_}
132.52	Michael Robinson	Sat Jul 6 04:21:29 2002	-nl @n{/(.+) (.+)/}|=n;$n{$2}+=$1ne$2;push@{$g{$1}},$2}{(map{$n{$_}--for@{$g{$_}};delete$n{$_};print}grep!$n{$_},keys%n)?redo:exit%n
133.47	Amir Karger	Wed Jul 3 19:27:56 2002	-nal ($l,$r)=@F; $p{$l}{$r}=1;$p{$r}{$r}=0 } { for(keys%p){ for$n(keys%p){ grep{$p{$_}{$n}}keys%p or delete$p{$n},print$n } } %p&&die
			If I can't get the fancy algorithm to work, I guess I'll try fixing the old one. It does put me ahead a few spots. But it seems clear I need yet another another new algorithm.
133.49	Wladimir Palant	Wed Jul 3 16:07:04 2002	-nal push@a,[@F]if$F[0]ne$F[1];@a{@F}=(1,1)}{a while grep{$a=$_;delete$a{$_},print$a if!grep$_->[1]eq$a&$a{$_->[0]},@a}keys%a;%a&&die
133.51	pom	Mon Jul 1 10:11:59 2002	-ln / /;++$b{$'}{$`}if$`ne$';$b{$`}||={}}{while(%b){delete@b{@d=grep{!%{$b{$_}}&&print}keys%b}||die;delete@{%{$b{$_}}}{@d}for keys%b}
			Must... write... thesis... Must... not... play... golf... argl
133.52	Yanick Champoux	Thu Jul 4 00:41:40 2002	-lna push@a,[@F[$_..1]]for$F[0]eq$F[1],1}while(@a=grep!grep($x eq$_,@$_),@a){(($x)=map{($x=$$_[0])x!grep$$_[1]eq$x,@a}@a)?print$x:die
			gooble gooble gooble
133.52	Scott Hall	Sun Jul 7 13:39:11 2002	-nla ($;,$a)=@F;$f{$_}++||push@{$a{$a}},$;eq$a?():$;}sub v{$b{$_}++?die:v($_),--$b{$_}for@{$a{$_}};$p{$_}++||print}for(keys%a){my%b;v
134.53	Peter Haworth	Fri Jul 5 17:35:02 2002	-nla @v{($,,$r)=@F}=0;$,eq$r||push@{$e{$r}},$,}sub v{$v{$_}=2;$v{$_}?$v{$_}&2&&die:&v for@{$e{$_}};$v{$_}=print}{$v{$_}||&v for keys%v
			Only a few strokes off. Need new algorithm
135.48	Wladimir Palant	Wed Jul 3 14:24:23 2002	-anl push@a,[@F]if$F[0]ne$F[1];@a{@F}=(1,1)}{a while grep{$a=$_;delete$a{$_},print$a if!grep{$_->[1]eq$a&$a{$_->[0]}}@a}keys%a;die if%a
135.53	Scott Hall	Sat Jul 6 17:59:11 2002	-nla ($;,$a)=@F;$f{$_}++||push@{$a{$a}},$;eq$a?():$;} sub v{$b{$_}++?die:v($_),--$b{$_}for@{$a{$_}};$p{$_}++||print} for(keys%a){my%b;v
135.54	Michael Robinson	Wed Jul 3 02:47:50 2002	-nl @n{/(.+) (.+)/}||=0;$n{$2}+=$1ne$2;push@{$g{$1}},$2}{(map$n{$_}?():do{map$n{$_}--,@{$g{$_}};delete$n{$_};print},keys%n)?redo:exit%n
			My problem is, when I switch to a different method, it always ends up *longer*.
135.61	Terje K	Thu Jul 4 21:25:33 2002	-anl ($;,$~)=@F;$h{$~}{$"}=1;$h{$;}{$~}=1if$;ne$~}{sub z{my$x=pop;map$h{$_}{$x}&&{$u{$_}++&&die,$y{$_}++||z($_)&&print,$u{$_}--},%h}z$"
136.47	Michael Thelen	Mon Jul 1 15:07:11 2002	-ln / /,$`ne$'&&$b{$'}{$`}++;$b{$`}{0}}while(%b){delete@b{@k=grep{!keys%{$b{$_}}}keys%b}or die;delete@{$b{$_}}{@k}for keys%b;print for@k
136.48	Jasper McCrea	Wed Jul 3 11:39:38 2002	-ln0 @e{/\S+/g}=1;@a=keys%e;eval'$s=0;for$i(0..@a){for$j($i+1..@a){$s=@a[$i,$j]=@a[$j,$i]if/^\Q$a[$j] $a[$i] /m}}'x@a;$s?die:print for@a
			I have another solution in the pipeline, but those damn metacharacters are a right royal pain!!
136.51	Yanick Champoux	Wed Jul 3 23:38:08 2002	-lna push@a,[@F[$_..1]]for$F[0]eq$F[1],1}while(@a=grep!grep($_ eq$x,@$_),@a){ (($x)=grep{$x=$_;!grep$$_[1]eq$x,@a}map@$_,@a)?print$x:die
			What I wouldn't give to borrow Ton's brain for a few hours...
137.46	Andrew Savige	Wed Jul 3 21:08:18 2002	-aln $p{$_}++or@n{($:,$;)=@F}+=$:ne$;&&!!push@{$s{$:}},$;}{@l=grep!$n{$_},keys%n;print,push@l,grep!--$n{$_},@{$s{$_}}for@l;@l-keys%n&&die
			caught `/ again!
137.50	Yanick Champoux	Wed Jul 3 15:49:40 2002	-lna push@a,[@F[$_..1]]for$F[0]eq$F[1],1}{($x)=grep{$x=$_;!grep$$_[1]eq$x,@a}map@$_,@a or die;print$x;redo if@a=grep!grep($_ eq$x,@$_),@a
			Sec, okay. |3en, it's hard. But. /-\. will. *not* beat. me.
137.52	Honza Pazdziora	Tue Jul 2 16:12:48 2002	-alp @h{@F}=@F;push@{$l{$F[0]}},$F[1]if$F[0]ne$F[1]}while(%h){($_)=grep{$*=$_;!grep{$*eq$_}map{@$_}%l}%h or die;delete$l{$_};delete$h{$_}
137.53	Benoît Chauvet	Mon Jul 1 08:47:17 2002	-ln / /;$p{$'}+=$`ne$';$p{$`}|=0;$n{$`}.=$'.$;;END{{}while grep{$a+=print;map{$p{$_}--}$_,split$;,$n{$_}}grep!$p{$_},@a=keys%p;exit$a-@a}
			First try. Not so difficult than tpr04 !
137.54	Peter Haworth	Fri Jul 5 11:01:48 2002	-nla @v{($,,$r)=@F}=();$,eq$r||push@{$e{$r}},$,}sub v{$v{$_}=1;$v{$_}?$v{$_}&1&&die:&v for@{$e{$_}};print;$v{$_}=2}{$v{$_}||&v for keys%v
			Here's algorithm #3. I love google
137.58	Terje K	Thu Jul 4 20:58:11 2002	-anl ($a,$b)=@F;$h{$b}{$"}=1;$h{$a}{$b}=1if$a ne$b}{sub z{my$x=pop;map{$h{$_}{$x}&&{$u{$_}++&&die,$y{$_}++||z($_)&&print,$u{$_}--}}%h}z$"
138.48	Jasper McCrea	Tue Jul 2 17:06:05 2002	-ln0 @e{/\S+/g}=1;@a=keys%e;eval'$s=0;for$i(0..@a){for$j($i+1..@a){$s=@a[$i,$j]=@a[$j,$i]if/^\Q$a[$j] $a[$i]\E$/m}}'x@a;$s?die:print for@a
			sorry 'bout that.
139.49	Ala Qumsieh	Thu Jul 4 03:48:52 2002	-ln / /;$`eq$'||$s{$`}{$'}++||$p{$'}++;$p{$`}+=0}while(%p){@a=grep!$p{$_},keys%p or die;map{delete$p{$_};print;$p{$_}--for keys%{$s{$_}}}@a
			Dunno how much more juice I can suck out of this ...
139.50	Jon Coppeard	Mon Jul 1 20:41:20 2002	-nl12 / /;$p{$'}++,push@{$e{$`}},$' if$`ne$';$p{$`}+=0}{@n=grep!$p{$_},keys%p;delete$p{$_},print,push@n,grep!--$p{$_},@{$e{$_}}for@n;exit%p
139.57	Terje K	Thu Jul 4 20:46:17 2002	-anl ($a,$b)=@F;$h{$b}{$"}=1;$h{$a}{$b}=1if$a ne$b}{sub z{my$x=pop;map{if($h{$_}{$x}){$u{$_}++&&die;$y{$_}++||z($_)&&print;$u{$_}--}}%h}z$"
140.47	Michael Thelen	Mon Jul 1 15:02:14 2002	-lna / /&&$`ne$'&&$b{$'}{$`}++;$b{$`}||={}}while(%b){delete@b{@k=grep{!keys%{$b{$_}}}keys%b}or die;delete@{$b{$_}}{@k}for keys%b;print for@k
140.48	Amir Karger	Wed Jul 3 13:15:24 2002	-nal ($l,$r)=@F; $p{$l}{$r}=$p{$r}{$r}=1 } { for(keys%p){ for$n(keys%p){ grep{$_ ne$n&&$p{$_}{$n}}keys%p or delete$p{$n},print$n } } %p&&die
			Use -na with }{ trick. I fear this algorithm is nearing its end, though...
140.49	David Lowe	Mon Jul 1 20:14:32 2002	-lna ($l,$r)=@F;$l eq$r||++$g{$r}{$l};$g{$l}||={}}{map{if(!%{$g{$_}}){print;for$k(keys%g){delete$g{$k}{$_}}delete$g{$_};redo}}keys%g;%g&&die
140.52	Scott Hall	Sat Jul 6 17:20:28 2002	-nla $f{$_}++&&next;($;,$a)=@F;push@{$a{$a}},$;eq$a?():$;} sub v{$b{$_}++?die:v($_),--$b{$_}for@{$a{$_}};$p{$_}++||print} for(keys%a){my%b;v
141.46	Chris Dolan	Wed Jul 3 21:25:00 2002	-pal sub a{$c{my$z=pop}++&&die;$z eq$_||a($_)for@{$a{$z}};$c{$z}--;push@o,$z if!grep$_ eq$z,@o}push@{$a{$F[1]}},@F}map{%c=a$_}keys%a;for(@o){
			The new test scripts are killing me...
142.46	Andrew Savige	Wed Jul 3 15:23:03 2002	-aln $p{$_}++or@n{($:,$;)=@F}+=$:ne$;&&(push(@{$s{$:}},$;),1)}{ @l=grep!$n{$_},keys%n;print,push@l,grep!--$n{$_},@{$s{$_}}for@l;@l-keys%n&&die
142.51	Yanick Champoux	Wed Jul 3 15:13:55 2002	-lna push@a,[@F[$_..1]]for$F[0]eq$F[1]..1}while((@a,$_)=grep!grep($_ eq$x,@$_),@a){($x)=grep{$x=$_;!grep$$_[1]eq$x,@a}map@$_,@a or die;print$x
			So much for so little.
143.46	Andrew Savige	Wed Jul 3 14:29:18 2002	-aln $p{$_}++or@n{($:,$y)=@F}+=$:ne$y&&(push(@{$s{$:}},$y),1)}{ @l=grep!$n{$_},keys%n;print,push@l,grep!--$n{$_},@{$s{$_}}for@l; @l-keys%n&&die
			This is utter bilge and I have no idea why it works but I've finally caught `/ by a nostril. :)))
143.53	Yanick Champoux	Wed Jul 3 02:20:27 2002	-lna push@a,[$F[1]],[@F[0..$F[0]ne$F[1]]]}while((@a,$_)=grep!grep($_ eq$y,@$_),@a){for$x(map@$_,@a){grep$_->[1]eq$x,@a or$y=$_=$x}/./?print:die
			For that '0 1' test case that blarged my score, I hate you of all my heart, and curse your name until the one hundred and twenty-seventh generation (inclusive).
143.57	Michael Robinson	Mon Jul 1 23:43:00 2002	-n @n{/(.+) (.+)/}||=0;$1ne$2&&$n{$2}++;push@{$g{$1}},$2}{(map{$n{$_}?():do{map$n{$_}--,@{$g{$_}};delete$n{$_};print"$_ "}}keys%n)?redo:%n&&die
			Just printing in the loop does save a few bytes.
144.39	Bruce Gray	Thu Jul 4 19:36:32 2002	-lan ($l,$r)=@F;$s{$r}{$l}=1if$l ne$r;$s{$l}||={};END{while(($l)=grep!%{$s{$_}},keys%s){print$l;delete$s{$_}{$l}for keys%s;delete$s{$l}}%s&&die}
145.44	Dr. Mu	Sun Jul 7 22:33:11 2002	-l -0 ($s=$\.<>)=~s/^(\S+) \1$/\1/gm;$s=~s/(\s)\Q$t[0]\E(?=\s)/$1/gm&&print$t[0]while@t=grep$s!~/\S \Q$_\E$/m,$s=~/^ *(\S+)(?:\s)/gm;exit$s=~/\S/
145.47	Honza Pazdziora	Tue Jul 2 15:53:45 2002	-alp @h{@F}=@F;push@{$l{$F[0]}},$F[1]if$F[0]ne$F[1]}while(keys%h){($_)=grep{$*=$_;!grep{$*eq$_}map{@$_}%l}keys%h or die;delete$l{$_};delete$h{$_}
145.56	Eugene van der Pijll	Tue Jul 2 00:04:37 2002	-lan00 $g{$_}+=$|--for@F;$g{$_}--for/^(.*) \1$/gm;for($y=@o=keys%g;$y-=@n;){for$x(@n=grep!$g{$_},@o){$g{$_}--||print for/^\Q$x\E (.*)/gm,$x}1/@n}
			You don't often see a for(;;) loop in golf. It's slowly getting better, but I'll have to find a better algorithm.
146.44	Philippe Bricout	Sun Jul 7 22:03:36 2002	@h{(split)}++for@t=<>;$v=join$/,keys%h,'';{($a,$b)=split,$c+=$v=~s/^\Q$b\E$(.*)^\Q$a\E$/$a${1}$b/msfor@t;$c?$c=0:last;++$h{$v}>1&&die;redo}print$v
147.44	Dr. Mu	Sun Jul 7 18:41:52 2002	-l -0 ($s=$\.<>)=~s/^(\S+) \1$/\1/gm;while(@t=grep$s!~/\S \Q$_\E$/m,$s=~/^ *(\S+)(?:\s)/gm){$s=~s/(\s)\Q$t[0]\E(?=\s)/$1/gm;print$t[0]}exit$s=~/\S/
147.45	Michael Thelen	Mon Jul 1 06:08:45 2002	-lna ($j,$k)=@F;$b{$k}{$j}=1if$j ne$k;$b{$j}||={}}while(%b){delete@b{@k=grep{!keys%{$b{$_}}}keys%b}or die;delete@{$b{$_}}{@k}for keys%b;print for@k
			The solution still looks big and gross, but it's time for bed.
147.49	Ross Younger	Fri Jul 5 13:08:28 2002	-alp @U{($a,$b)=@F}=0;$R{$b}{$a}=1if$a ne$b}@U=keys%U;while(@U){@U=sort{%{$R{$a}}<=>%{$R{$b}}}@U;%{$R{$a=shift@U}}&&die;delete$R{$_}{$a}for@U;$_=$a
			Yikes, changed my algorithm and only saved 7 strokes :-/
147.52	Wladimir Palant	Wed Jul 3 14:04:06 2002	-anl push@a,[@F]if$F[0]ne$F[1];@_{@F}=(1,1)}{$b=1,map{$c=$_;if(!grep{$_->[1]eq$c&$_{$_->[0]}}@a){delete$_{$c},$b=0,print$c}}keys%_ until$b;die if%_
148.56	Eugene van der Pijll	Mon Jul 1 23:54:51 2002	-lan00 $g{$_}+=$|--for@F;$g{$_}--for/^(.*) \1$/gm;for($y=@o=keys%g;$y-=@n;){1/(@n=grep!$g{$_},@o);for$x(@n){$g{$_}--||print for/^\Q$x\E (.*)/gm,$x}}
149.46	Jonathan Stimmel	Sat Jul 6 00:37:38 2002	-ln / /;$` ne $'&&$r{$`}{$'}++;$r{$'}||={}}$,=$\;while(%r){my%n;@n{keys%r}++;map delete@n{keys%$_},values%r;%n?print grep delete$r{$_},keys%n:exit 1;
149.55	Ala Qumsieh	Tue Jul 2 19:31:24 2002	-ln /^(.*) (.*)$/;$1eq$2||$_{$1}{$2}++||$e{$2}++;$e{$1}+=0}{@a=grep!$e{$_},keys%e or die;map{print;delete$e{$_};map$e{$_}--,keys%{$_{$_}}}@a;%e&&redo
			not much ... but better ..
150.46	Jon Coppeard	Mon Jul 1 19:16:11 2002	-nl12 ($a,$b)=split;$p{$b}++,push@{$e{$a}},$b if$a ne$b;$p{$a}+=0}{@n=grep!$p{$_},keys%p;delete$p{$_},print,push@n,grep!--$p{$_},@{$e{$_}}for@n;exit%p
150.55	Eugene van der Pijll	Mon Jul 1 23:29:33 2002	-lan00 $g{$_}+=$|--for@F;$g{$_}--for/^(.*) \1$/gm;while(grep$g{$_}>=0,@F){1/(@n=grep!$g{$_},keys%g);for$x(@n){$g{$_}--||print for/^\Q$x\E (.*)/gm,$x}}
151.46	Damien Neil	Tue Jul 2 23:18:19 2002	-lna $M{$F[0]}||={};@{$M{$F[1]}}{@F}=1;END{while(keys%M>0){($n)=grep{keys%{$M{$_}}<=1}keys%M or die;print$n;delete$M{$n};map{delete$_->{$n}}values%M;}}
			Initial what-the-hey solution, just to get something in there.
152.47	Amir Karger	Fri Jul 5 02:18:04 2002	-n0 s/^(\S+) \1$/$1/gm; $a=1; while ($a--) { for $node (split) { if (!/ \Q$node\E\n/) { s/^\Q$node\E\s//gm; print "$node\n"; $a=1; last } } } die if $_
			This is It! The One True Algorithm! This version is still somewhat readable; let's work on that.
152.50	Josef Drexler	Thu Jul 4 19:33:12 2002	-lna @v{($@,$b)=@F}=0;push@{$e{$@}},$@ne$b?$b:()}sub d{my($x,$o)=@_;$r{$x}++&&die;$o+=1+d($_)for@{$e{$x}};$r{$x}--;$o}map print,sort{d($b)- d$a}keys%v;{
152.52	Amir Karger	Wed Jul 3 13:17:54 2002	-pal @a{($l,$r)=@F}=@F; $p{$r}{$l}=1if$l ne$r } while(($n)=grep!%{$p{$_}},%a){ delete$p{$_}{$n}for%p; delete$a{$n}; print$n } for(keys%a){%{$p{$_}}&&die
			YAAWWPGMN (Yet another algorithm which will probably get me nowhere.)
153.44	Michael Thelen	Mon Jul 1 05:40:04 2002	-lna ($j,$k)=@F;$b{$k}{$j}=1if$j ne$k;$b{$j}||={}}while(%b){@k=grep{!keys%{$b{$_}}}keys%b;@k||die;delete@b{@k};delete@{$b{$_}}{@k} for keys%b;print for@k
			Not bad for 40 minutes' work, I suppose. Lots of easy improvements, and probably many more to be made.
153.49	David Lowe	Mon Jul 1 19:09:18 2002	-lna ($l,$r)=@F;$l eq$r||++$g{$r}{$l};$g{$l}or$g{$l}={}}O:if(%g){for(keys%g){if(!%{$g{$_}}){print;for$k(keys%g){delete$g{$k}{$_}}delete$g{$_};goto O}}die
			First pass
153.50	Philippe Bricout	Sat Jul 6 12:34:18 2002	@t=map{chop;/ /;@h{$',$`}++;"\$v=~s/^\Q$'\E\$(.*)^\Q$`\E\$/$`\${1}$'/ms"}<>;$v=join$/,keys%h,'';{$b+=eval for@t;$b?$b=0:last;++$h{$v}>1&&die;redo}print$v
153.55	Eugene van der Pijll	Mon Jul 1 22:58:41 2002	-lan00 $g{$_}+=$|--for@F;$g{$_}--for/^(.*) \1$/gm;while(grep$g{$_}>=0,@@=keys%g){@n=grep!$g{$_},@@or&x;for$x(@n){print$x;$g{$_}--for/^\Q$x\E (.*)/gm,$x}}
			Undefined subroutine &main::x called...
154.48	Andrew Savige	Wed Jul 3 07:47:49 2002	-aln $p{$_}++or($:,$;)=@F,$n{$:}+=0,$:eq$;or++$n{$;},push@{$s{$:}},$;}{ @l=grep!$n{$_},keys%n;print,map--$n{$_}||push(@l,$_),@{$s{$_}}for@l; exit@l-keys%n
			Using exit like this may be unsound; I will leave it for referees to decide. Note that exit(256) returns 0 to the operating system.
154.53	Ross Younger	Tue Jul 2 22:11:28 2002	-alp @U{@F}=0;push@{$L{$F[0]}},pop@F}sub v{$U{my$a=$_}=2;$_ eq$a||1<$U{$_}&&die||$U{$_}||&v for@{$L{$a}};--$U{$a};@O=($a,@O)}$U{$_}||v for keys%U;for(@O){
155.48	Andrew Savige	Wed Jul 3 07:45:34 2002	-aln $p{$_}++or($:,$;)=@F,$n{$:}+=0,$:eq$;or++$n{$;},push@{$s{$:}},$;}{ @l=grep!$n{$_},keys%n;print,map--$n{$_}||push(@l,$_),@{$s{$_}}for@l; @l-keys%n&&die
			Shuffling deck chairs on the Titanic. And yet I am enjoying it in a perverse kind of way.
155.49	FatPhil	Wed Jul 3 15:21:29 2002	-n / (\S+)/;$p{$1}.=" $` "if$`ne$1;$p{$`}x=1}for(;;@f=()){map{@f||$p{$_}||print+@f=$_,$/}keys%p;@f?delete$p{$f[0]}:each%p?die:last;map s/ \Q@f //g,values%p
			$f -> @f. Phew, half of my drop clawed back.
156.42	Yanick Champoux	Tue Jul 2 15:46:47 2002	-pa push@{$s{$F[0]}},$F[1]if$F[0]ne$F[1];$s{$F[1]}||=[]}{while(($x)=grep!@{$s{$_}},keys%s){$\="$x ".$\;map{@$_=grep$_ ne$x,@$_}values%s;delete$s{$x}}%s&&die
			*whistling that 'Barber of Seville' tune*
156.57	Michael Robinson	Mon Jul 1 11:15:44 2002	-n @n{/(.+) (.+)/}||=0;$1ne$2&&$n{$2}++,push@{$g{$1}},$2}{$c=0;map{$n{$_}||do{$c=$o.="$_ ";map$n{$_}--,@{$g{$_}};delete$n{$_}}}keys%n;$c?redo:%n?die:print$o
157.45	Jonathan Stimmel	Fri Jul 5 05:57:29 2002	-lna ($l,$r)=@F;$l ne $r&&$r{$l}{$r}++;$r{$r}||={}}$,=$\;while(%r){my%n;@n{keys%r}++;map delete@n{keys%$_},values%r;%n?print grep delete$r{$_},keys%n:exit 1;
157.49	Brad Jones	Tue Jul 2 04:58:59 2002	-nl ($",$b)=split;$"ne$b&&$l{$b}->{$"}++;$l{$"}||={};END{while(%l){(($c)=grep{!%{$l{$_}}}keys%l)||die;print$c;delete$l{$c};for(keys%l){delete$l{$_}->{$c};}}}
157.51	Michael Wrenn	Sun Jul 7 05:45:41 2002	-lan @u{@F}=@_=(@_,$_)}{@u=keys%u;while(!@g&(++$h<1E4||die$h)){for(@_){@c{@u}=0..@u;@g=($x,$y)=@c{+split};if($x>$y){@u[@g]=@u[$y,$x];@g=();last}}}print for@u
			Movin' on up to the back of the pack! Now I need to find another new algorithm!Got any suggestions? Damn those time-wasting hashes! They were way too much fun! Great Hole Rick & company!
157.51	Wladimir Palant	Wed Jul 3 13:42:44 2002	-anl push@a,[@F]if$F[0]ne$F[1];@_{@F}=(1,1)}{$b=1,map{$c=1;foreach$d(@a){$c=0if$d->[1]eq$_&$_{$d->[0]}}delete$_{$_},$b=0,print$_ if$c}keys%_ until$b;die if%_
158.47	Amir Karger	Tue Jul 2 21:29:43 2002	-l while (<>) { ($l,$r)=split; $p{$l}{$r}=$p{$r}{$r}=1 } for(keys%p) { for$n(keys%p) { grep{$_ ne$n&&$p{$_}{$n}} keys%p or delete $p{$n},print$n } } die if %p
			Ha! I've caught up to Andrew! Too bad this algorithm is going nowhere.
158.47	Daniel Cutter	Thu Jul 4 21:38:30 2002	-na $a{$_}++for($a,$b)=@F;push@r,qr/( \Q$b\E\n)(.*)( \Q$a\E\n)/s;END{$e.=" $_\n"for keys%a;for(@r){$e=~s/$_/$3$2$1/for@r}$e=~$_&&die for@r;$e=~s/ //g;print$e}
159.48	FatPhil	Wed Jul 3 14:51:36 2002	-n / (\S+)/;$p{$1}.=" $` "if$`ne$1;$p{$`}x=1}for(;;$f=''){map{''ne$f||$p{$_}||print$f=$_,$/}keys%p;''ne$f?delete$p{$f}:each%p?die:last;map s/ \Q$f //g,values%p
			Hoorah, getting longer!
159.51	Prakash Kailasa	Mon Jul 8 03:59:11 2002	-lan push@{$h{$F[0]}},$F[1];$h{$F[1]}||=[]}{sub d{my($w,$x)=@_;$c{$w}||=1+do{map{$d++>$.&&&,,$x+=d($_)if$_ ne$w}@{$h{$w}};$x}}map{print}sort{d($b)<=>d$a}keys%h
159.53	Philippe 'BooK' Bruhat	Thu Jul 4 12:23:30 2002	-lna END{do{for(@a=grep{!$a{$_}}keys%a){$a{$_}--for$a=~/$;\Q$_\E (\S+)^I/g;delete$a{$_};print}%a&&!@a&&die}while%a}$a{pop@F}+=$_ for($F[0]ne$F[1],0);$a.="$;$_^I"
			A whole new approach, using a classical algorithm. Gained a dozen strokes, but still far behind.
160.48	Daniel Cutter	Thu Jul 4 21:24:09 2002	-na $a{$_}++for($a,$b)=@F;push@r,qr/( \Q$b\E\n)(.*)( \Q$a\E\n)/s;END{$e.=" $_\n"for keys%a;for(@r){$e=~s/$_/$3$2$1/for@r}$e=~/$_/&&die for@r;$e=~s/ //g;print$e}
160.49	Prakash Kailasa	Mon Jul 8 03:34:03 2002	-lan push@{$h{$F[0]}},$F[1];$h{$F[1]}||=[]}{sub d{my$w=pop;my$x=1;$c{$w}||=do{map{$d++>$.&&&,,$x+=d($_)if$_ ne$w}@{$h{$w}};$x}}map{print}sort{d($b)<=>d$a}keys%h
			just 100 more than The Alien :-)
160.52	Eugene van der Pijll	Mon Jul 1 22:44:58 2002	-ln00 $g{$_}+=$|--for/\S+/g;$g{$_}--for/^(.*) \1$/gm;while(grep$g{$_}>=0,keys%g){@n=grep!$g{$_},keys%g or die;for$x(@n){print$x;$g{$_}--for/^\Q$x\E (.*)/gm,$x}}
			Grrrr... ugly solution... much too long... can't see anything shorter... really don't like this course... grrrr.
161.47	Brad Jones	Tue Jul 2 04:07:55 2002	-nl ($",$b)=split;$"ne$b&&$l{$b}->{$"}++;$l{$"}||={};END{while(%l){(($c)=grep{!keys%{$l{$_}}}keys%l)||die;print$c;delete$l{$c};for(keys%l){delete$l{$_}->{$c};}}}
162.45	Peter Haworth	Thu Jul 4 17:49:37 2002	-nla $p{$_}+=0for($,,$r)=@F;$,eq$r||++$p{$r}+push@{$s{$,}},$r}for(@q=grep!$p{$_},keys%p;@q;push@q,grep!--$p{$_},@{$s{ $_}}){print$_=pop@q}die if grep$_,values%p;{
162.46	Jonathan Stimmel	Fri Jul 5 04:40:43 2002	-lna ($l,$r)=@F;$l ne $r&&$r{$l}{$r}++;$r{$r}||={}}$,=$\;while(%r){my%n;@n{keys%r}++;map delete@n{keys %$_},values %r;%n||exit 1;print grep delete$r{$_},keys(%n);
164.41	Yanick Champoux	Tue Jul 2 01:07:46 2002	-na END{while(($x)=grep!@{$s{$_}},keys%s){$\="$x ".$\;map{@$_=grep$_ ne$x,@$_}values%s;delete$s{$x}}%s?die:print}push@{$s{$F[0]}},$F[1]if$F[0]ne$F[1];$s{$F[1]}||=[]
			Okay, enough for tonight.
164.48	Brad Jones	Tue Jul 2 02:28:54 2002	-nl ($",$b)=split;($"ne$b)&&$l{$b}->{$"}++;$l{$"}||={};END{while(%l){(($c)=grep{!keys%{$l{$_}}}keys%l)||die;print$c;delete$l{$c};for(keys%l){delete $l{$_}->{$c};}}}
165.58	Michael Robinson	Mon Jul 1 09:20:37 2002	-n /(.+) (.+)/;$1ne$2&&$n{$2}++,push@{$g{$1}},$2;@n{$1,$2}||=()}{$c=0;map{$n{$_}||do{map$n{$_}--,@{$g{$_}};$c=$o.="$_ ";delete$n{$_}}}keys%n;$c&&redo;%n&&die;print$o
			Not very good translation of tsort(1) from BSD, with minor tweaks.
166.52	Philippe Bricout	Thu Jul 4 22:01:43 2002	@t=map{s/(.*) (.*) /\$v=~s\/^\Q$2\E\$(.*)^\Q${1}\E\$\/$1\${1}$2\/ms/;@h{$1,$2}++;$_}<>;$v=join" ",keys%h,'';{$b+=eval for@t;$b||last;++$h{$v}>1&&die;$b=0;redo}print$v
167.46	Marcelo E. Magallon	Sat Jul 6 12:32:46 2002	-ln @p{($a,$b)=split}||={};next if$a eq$b;$p{$b}{$a}=1}{L:while(@k=keys%p){for(@k){next if keys%{$p{$_}};for$k(@k){delete$p{$k}{$_};}delete$p{$_};print;next L}exit 1;}
			What am I doing in the alien league? I'm submitting this shameful solution only because I don't want to let another TPR course fly by.
167.53	Philippe 'BooK' Bruhat	Thu Jul 4 14:37:07 2002	-lna END{@a=keys%a;T:$b=~/\Q@a/&&die;map{($c,$d)=@$_;@a{@a}=0..@a;$b.="^I@a",@a[@a{@$_}]=@a[@a{$d,$c}],goto T if$a{$c}>$a{$d}}@r;$"=$/;print"@a";exit}push@r,[@a{@F}=@F]
			s/for/map/ to shave a stroke. Test 27 takes time and memory, but it passes.
168.41	Yanick Champoux	Tue Jul 2 00:55:00 2002	-na END{N:for$x(grep!@{$s{$_}},keys%s){$\="$x ".$\;map{@$_=grep$_ ne$x,@$_}values%s;delete$s{$x};goto N}%s?die:print}push@{$s{$F[0]}},$F[1]if$F[0]ne$F[1];$s{$F[1]}||=[]
			Not very efficient, but I find it cute, in an unwholesome manner.
168.52	Ala Qumsieh	Tue Jul 2 18:57:59 2002	-lna $F[0]eq$F[1]||$_{$F[0]}{$F[1]}++||$e{$F[1]}++;$e{$F[0]}+=0}while(@b=keys%e){if (@a=grep!$e{$_},@b){for (@a){print;delete$e{$_};$e{$_}--for keys%{$_{$_}}}}else{die}
			At least this one works. Major changes coming up ..
168.53	Philippe 'BooK' Bruhat	Thu Jul 4 13:55:37 2002	-lna END{@a=keys%a;T:$b=~/\Q@a/&&die;for(@r){($c,$d)=@$_;@a{@a}=0..@a;$b.="^I@a",@a[@a{@$_}]=@a[@a{$d,$c}],goto T if$a{$c}>$a{$d}}$"=$/;print"@a";exit}push@r,[@a{@F}=@F]
			Well after all, this is a sort. So I used a classic approach to sorting: swap elements that are in the wrong order. And keep a list of everything we tried. If we try the same thing twice, then there's a loop. For the record, test 27 requires 3587 swaps, and the string that stores all the possibilities we tried is 6675407 bytes long.
169.47	Andrew Savige	Tue Jul 2 11:18:27 2002	-aln $p{$_}++or($:,$;)=@F,$n{$:}+=0,$:eq$;or++$n{$;},push@{$s{$:}},$;}{ @l=grep!$n{$_},keys%n;while(@l){--$n{$_}||push@l,$_ for@{$s{$_=pop@l}};print} map$_&&die,values%n
			Don't you dare laugh! Just because all I did was golf on PPT tsort.pl a bit. Oh well, it's a start. But I fear I won't have the time to hold off `/ this time and will wind up being stabbed in the back while changing a light bulb.
169.50	Josef Drexler	Thu Jul 4 20:22:12 2002	-lna @v{($@,$b)=@F}=0;$e{$@}{$b}++if$@ne$b}sub d{$r{$_}++&&die,$o{$_}||=1+(sort{$b-$a}map d($_),keys%{$e{$_}})[0],$r{$_}--for@_;@o{@_}}map print,sort{d($b)- d$a}keys%v;{
			this one's a lot faster...
173.52	Wladimir Palant	Wed Jul 3 13:03:50 2002	-anl push@a,[@F]if$F[0]ne$F[1];$_{$F[0]}=$_{$F[1]}=1;END{$b=1,map{$c=1;foreach$d(@a){$c=0if($d->[1]eq$_&&$_{$d->[0]})}delete$_{$_},$b=0,print$_ if$c}keys%_ until$b;die if%_}
173.57	Philippe 'BooK' Bruhat	Wed Jul 3 13:47:14 2002	-lna ($0,$b)=@F;$0ne$b&&!$0{$0}{$b}&&$r{$b}++;$0{$0}{$b}=$0{$b}||={};END{while(@a=keys%0){(@_=grep{!$r{$_}}@a)||die;for(@_){$r{$_}--for$_,keys%{$0{$_}};print;delete$0{$_}}}}
			Removed the sub that's called once.
176.43	Yanick Champoux	Tue Jul 2 00:12:45 2002	-na END{N:if(@z=keys%s){for$x(@z){@{$s{$x}}&&next;$\="$x ".$\;$_=[grep$_ ne$x,@$_]for values%s;delete$s{$x};goto N}die}print}push@{$s{$F[0]}},$F[1]if$F[0]ne$F[1];$s{$F[1]}||=[]
			Of course, using a string and regexes is probably the way to go. But arrays want love, too!
176.46	Bruce Gray	Mon Jul 1 17:59:27 2002	-lan my($l,$r)=@F;$n{$l}+=0;next if$l eq$r;++$n{$r};push@{$s{$l}},$r;END{@_=grep{!$n{$_}}keys%n;while(@_){$_=pop@_;print;push@_,grep!--$n{$_},@{$s{$_}}}$n{$_}&&die for keys %n}
178.52	Prakash Kailasa	Mon Jul 8 02:14:10 2002	-lan $i{$_}++||push@{$h{$F[1]}},$F[0];$h{$F[0]}||=[]}{sub d{my($w,$d,$x)=@_;$c{$w}||=do{map{$d>$.&&die,$x+=1+d($_,++$d)if$_ ne$w}@{$h{$w}};$x}}map{print}sort{d($a)<=>d($b)}keys%h
			I refuse to end this on sandtrap. This one's longer but should zip through the tests.
179.52	Philippe 'BooK' Bruhat	Wed Jul 3 12:18:08 2002	-lna ($0,$b)=@F;$0ne$b&&!$0{$0}{$b}&&$r{$b}++;$0{$0}{$b}=$0{$b}||={};sub a{@_||die'';for(@_){$r{$_}--for$_,keys%{$0{$_}};print;delete$0{$_}}}END{a grep{!$r{$_}}keys%0while keys%0}
			Removed the index variable from a for loop.
181.00	Michael Thelen	Mon Jul 1 05:27:52 2002	-l while(<>){($j,$k)=split;$j ne$k&&($b{$k}{$j}=1);$b{$j}||={}}while(%b){@k=grep{keys%{$b{$_}}==0}keys%b;exit(-1)if!@k;delete @b{@k};for$k(keys%b){delete$b{$k}{$_}for@k}print for@k}
			Boy, it's not at all pretty, but it's a first shot.
183.51	Philippe 'BooK' Bruhat	Wed Jul 3 12:13:11 2002	-lna ($0,$b)=@F;$0ne$b&&!$0{$0}{$b}&&$r{$b}++;$0{$0}{$b}=$0{$b}||={};sub a{@_||die'';for$a(@_){$r{$_}--for$a,keys%{$0{$a}};print$a;delete$0{$a}}}END{a grep{!$r{$_}}keys%0while keys%0}
			Replaced $a by $0 for consiseness.
184.48	Daniel Cutter	Thu Jul 4 19:54:16 2002	-an $a{$_}++for@F;($a,$b)=@F;push@r,qr/( \Q$b\E )(.*)( \Q$a\E )/ if$a ne$b;END{$e.=" $_ "for keys%a;for(1..keys%a){$e=~s/$_/$3$2$1/for@r}$e=~/$_/&&die for@r;print"$_\n"for split' ',$e}
185.49	Amir Karger	Wed Jul 3 18:47:05 2002	-nal ($l,$r)=@F; $p{$l}+=0; $p{$r}++if$r ne$l&&!$c{$l}{$r}++ } sub z{grep{!$p{$_}}@a=keys%p} @o=z; while(defined($_=pop@o)){ print; push@o,grep{!--$p{$_}}keys%{$c{$_}} } die if z!=@a; {
			Hm. This new algorithm was supposed to gain me more ground...
185.49	Philippe 'BooK' Bruhat	Wed Jul 3 12:07:59 2002	-lna ($a,$b)=@F;$a ne$b&&!$a{$a}{$b}&&$r{$b}++;$a{$a}{$b}=$a{$b}||={};sub a{@_||die'';for$a(@_){$r{$_}--for$a,keys%{$a{$a}};print$a;delete$a{$a}}}END{a grep{!$r{$_}}keys%a while keys%a}
			Removed @F for consiseness.
190.46	Boyd Nation	Sun Jul 7 04:04:00 2002	-l while(<>){chomp;($a,$b)=split;if($a ne$b){push(@{$s{$a}},$b);$p{$b}++}if(!$p{$a}){$p{$a}=0}}&H;sub H{for$v(keys%p){if(!$p{$v}){print$v;for$w(@{$s{$v}}){$p{$w}--}delete($p{$v});&H}}exit%p}
191.43	Yanick Champoux	Mon Jul 1 15:23:46 2002	-lna END{N:while(@z=keys%s){for$x(@z){if(!@{$s{$x}}){@y=($x,@y);$s{$_}=[grep$_ ne$x,@{$s{$_}}]for@z;delete$s{$x};next N}}exit 1}print for@y}push@{$s{$F[0]}},$F[1]if$F[0]ne$F[1];$s{$F[1]}||=[]
			Just testing the ground...
192.50	Brad Jones	Tue Jul 2 00:32:29 2002	-nl ($",$b)=split;($"ne$b)&&push@{$l{$b}},$";$l{$"}||=[];END{while(%l){$o=0;for$c(keys%l){if(!@{$l{$c}}){print$c;$o++;delete$l{$c};for(keys%l){$l{$_}=[grep{$_ ne$c}@{$l{$_}}];}}}!$o&&exit 1;}}
193.49	Brad Jones	Tue Jul 2 00:28:15 2002	-nl ($a,$b)=split;($a ne$b)&&push@{$l{$b}},$a;$l{$a}||=[];END{while(%l){$o=0;for$c(keys%l){if(!@{$l{$c}}){print$c;$o++;delete$l{$c};for(keys%l){$l{$_}=[grep{$_ ne$c}@{$l{$_}}];}}}!$o&&exit 1;}}
197.42	BigrTex	Wed Jul 3 18:57:09 2002	-lna ($l,$r)=@F;exists$p{$_}?next:x;$v{$l}+=0;if($l ne$r){++$v{$r};push@{$s{$l}},$r}}@t=grep{!$v{$_}}keys%v;while(@t){print$z=pop@t;map{push@t,$_ unless--$v{$_}}@{$s{$z}}}if(grep{$v{$_}}keys%v){die
			More fun with perldoc perlrun
197.51	Philippe 'BooK' Bruhat	Wed Jul 3 12:04:29 2002	-lna $F[0]ne$F[1]&&!$a{$F[0]}{$F[1]}&&$r{$F[1]}++;$a{$F[0]}{$F[1]}=$a{$F[1]}||={};sub a{@_||die'';for$a(@_){$r{$_}--for$a,keys%{$a{$a}};print$a;delete$a{$a}}}END{a grep{!$r{$_}}keys%a while keys%a}
			With the correct algorithm, it works much better... And I am below 200, too!
198.48	Daniel Cutter	Thu Jul 4 14:41:33 2002	-an $a{$_}++for@F;($a,$b)=@F;push@r,qr/( \Q$b\E )(.*)( \Q$a\E )/ if$a ne$b;END{$e.=" $_ "for keys%a;for(1..@r*keys%a){$e=~s/$_/$3$2$1/for@r}for(@r){die if$e=~s/$_/$3$2$1/}print"$_\n"for split' ',$e}
199.41	BigrTex	Wed Jul 3 18:54:05 2002	-ln ($l,$r)=split;exists$p{$_}?next:x;$v{$l}+=0;if($l ne$r){++$v{$r};push@{$s{$l}},$r}}@t=grep{!$v{$_}}keys%v;while(@t){print$z=pop@t;map{push@t,$_ unless--$v{$_}}@{$s{$z}}}if(grep{$v{$_}}keys%v){die
			for becomes map, and I get under 200 strokes...
199.42	Erik Beatty	Wed Jul 3 16:30:53 2002	while(<>){($p,$s)=split;$h{$s}.="\"$p\""if$p ne$s;$h{$p}.="";}$c=@k=keys%h;while(@k){$k=shift@k;if($h{$k}eq""){$c=@k;print"$k\n";map{$h{$_}=~s#"\Q$k\E"##g}@k;}else{push@k,$k;last if$c--<0;}}exit($c);
			its just not human to go below the 100 mark.. how in the world did they do it? ;)
201.41	BigrTex	Wed Jul 3 18:11:31 2002	-ln ($l,$r)=split;exists$p{$_}?next:x;$v{$l}+=0;if($l ne$r){++$v{$r};push@{$s{$l}},$r}}@t=grep{!$v{$_}}keys%v;while(@t){print$z=pop@t;for(@{$s{$z}}){push@t,$_ unless--$v{$_}}}if(grep{$v{$_}}keys%v){die
			And for that matter, although you can't just concat $l and $r to index into $p, you could use $_...
203.39	Yanick Champoux	Mon Jul 1 14:59:47 2002	-lna push@{$s{$F[0]}},$F[1]if$F[0]ne$F[1];$s{$F[1]}||=[];END{$,=$\;N:while(keys%s){for$x(keys%s){if(!@{$s{$x}}){unshift@y,$x;delete$s{$x};@{$s{$_}}=grep$_ ne$x,@{$s{$_}}for keys%s;next N}}exit 1}print@y}
			It's a start...
205.41	BigrTex	Wed Jul 3 18:10:28 2002	-ln ($l,$r)=split;exists$p{$l}{$r}?next:x;$v{$l}+=0;if($l ne$r){++$v{$r};push@{$s{$l}},$r}}@t=grep{!$v{$_}}keys%v;while(@t){print$z=pop@t;for(@{$s{$z}}){push@t,$_ unless--$v{$_}}}if(grep{$v{$_}}keys%v){die
			exists should create $p{$l}{$r}, so you don't need to ++ it...
207.50	Wladimir Palant	Wed Jul 3 10:42:54 2002	-anl push(@{$a{$F[1]}},$F[0])if$F[0]ne$F[1];push(@{$a{$F[0]}});END{$b=1;while($b){$b=0;map{$c=0;foreach$e(@{$a{$_}}){$c=1if!$d{$e};}if(!$d{$_}&&!$c){$d{$_}=1;print$_;$b=1;}}keys%a;}exit -1if keys%d!=keys%a;}
208.49	Matthias Schoder	Fri Jul 5 14:11:55 2002	-lan012 ($x,$y)=@F;$X{$x}{$y}=$Y{$y}{$x}=1if$x ne$y;map{$N{$_}++}@F;END{exit map{map{if(!$Y{$_}){print;for$y(keys%Y){delete$Y{$y}{$_};delete$Y{$y}if!%{$Y{$y}}}delete$X{$_};delete$N{$_};exit if!%N}}keys%N}%N}
210.49	Matthias Schoder	Fri Jul 5 14:02:40 2002	-lan012 ($x,$y)=@F;$X{$x}{$y}=$Y{$y}{$x}=1if$x ne$y;map{$N{$_}++}@F;END{exit map{map{if(!$Y{$_}){print;for$y(keys%Y){delete$Y{$y}{$_};delete$Y{$y}if!%{$Y{$y}}}delete$X{$_};delete$N{$_};exit if!%N}}keys%N}0..%N}
212.48	Brad Jones	Tue Jul 2 00:08:02 2002	-nl ($a,$b)=split;($a ne$b)&&push@{$l{$b}},$a;$l{$a}||=[];END{while(%l){@j=sort{@{$l{$a}}<=>@{$l{$b}}}keys%l;$c=shift @j;(@{$l{$c}})&&exit(1);print$c;delete$l{$c};foreach$d(@j){$l{$d}=[grep{$_ ne$c}@{$l{$d}}];}}}
212.49	Matthias Schoder	Fri Jul 5 13:00:18 2002	-lan012 ($x,$y)=@F;$X{$x}{$y}=$Y{$y}{$x}=1if$x ne$y;map{$N{$_}++}@F;END{map{map{if(!$Y{$_}){print;for$y(keys%Y){delete$Y{$y}{$_};delete$Y{$y}if!%{$Y{$y}}}delete$X{$_};delete$N{$_};exit if!%N}}keys%N}0..%N;exit 1}
212.54	Josef Drexler	Wed Jul 3 22:32:38 2002	-lna $v{$F[1]}||=0; ($v{$F[0]},$e{$F[1]}{$F[0]})=1if$F[0]ne$F[1] } @%=keys%v; sub d{ $r{$_}++&&die; $o{$_}||=1+(sort{$b-$a}map{d()}keys%{$e{$_}})[0]; $r{$_}--; $o{$_} } d for@%; map print,sort{$o{$a}-$o{$b}}@%; {
			It's a start... it works. I left the newlines in to make it more readable. I'll golf it later...
215.50	Matthias Schoder	Thu Jul 4 08:51:51 2002	-lan012 ($x,$y)=@F;$X{$x}{$y}=$Y{$y}{$x}=1if$x ne$y;map{$N{$_}++}@F;END{for(0..%N){for(keys%N){if(!$Y{$_}){print;for$y(keys%Y){delete$Y{$y}{$_};delete$Y{$y}if!%{$Y{$y}}}delete$X{$_};delete$N{$_};exit if!%N}}}exit 1}
216.42	BigrTex	Wed Jul 3 17:57:35 2002	-ln ($l,$r)=split;exists$p{$l}{$r}?next:$p{$l}{$r}++;$v{$l}+=0;if($l ne$r){++$v{$r};push@{$s{$l}},$r}}@t=grep{!$v{$_}}keys%v;while(@t){print$z=pop@t;for(@{$s{$z}}){push@t,$_ unless--$v{$_}}}if(grep{$v{$_}}keys%v){die
			A couple more strokes, since things aren't working at work... :(
219.42	BigrTex	Wed Jul 3 16:15:01 2002	-ln ($l,$r)=split;exists$p{$l}{$r}?next:$p{$l}{$r}++;$v{$l}+=0;if($l ne$r){++$v{$r};push@{$s{$l}},$r}}@t=grep{!$v{$_}}keys%v;while(@t){$_=pop@t;print;for$c(@{$s{$_}}){push@t,$c unless--$v{$c}}}if(grep{$v{$_}}keys%v){die
			*sighs* Dead last and trimming strokes slowly, not nearly quick enough. I don't know if you'll see any more submissions from me because of the long holiday weekend...
229.50	Wladimir Palant	Wed Jul 3 10:17:31 2002	while(<>){chomp;split(/ /);push(@{$a{$_[1]}},$_[0])if$_[0]ne$_[1];push(@{$a{$_[0]}});}$b=1;while($b){$b=0;map{$c=0;foreach$e(@{$a{$_}}){$c=1if!$d{$e};}if(!$d{$_}&&!$c){$d{$_}=1;print"$_\n";$b=1;}}keys%a;}exit -1if keys%d!=keys%a;
			First try
235.43	BigrTex	Tue Jul 2 19:21:52 2002	-l while(<>){($l,$r)=@l=split;exists$p{$l}{$r}?next:$p{$l}{$r}++;$r{$l}+=0;if($l ne$r){++$r{$r};push@{$s{$l}},$r}}@t=grep{!$r{$_}}keys%r;while(@t){$_=pop@t;print;for$c(@{$s{$_}}){push@t,$c unless--$r{$c}}}if(grep{$r{$_}}keys%r){exit 1}
			Another 65 strokes... variable names > 1 char are bad... :)
239.51	Michael Wrenn	Sat Jul 6 15:35:13 2002	-lna z($~,($^,$~)=@F);map{for$^(k($_)){z($_,$^,$~)}}k($~);map{for$^(keys%k){$~eq$^&&die$!or&z($_,$_,$^)if$k{$^}{$~}}}k($~)}map{print}k(@p=sort{$c{$a}<=>$c{$b}}keys%k),@p;{sub z{$k{$a=pop}{+pop}||=++$c{$a}if$^ne$_[0]}sub k{keys%{$k{$_[0]}}}
			That's all I have with hashes! This will give Kristen a moving target.
245.55	Philippe Bricout	Wed Jul 3 21:24:56 2002	-l for (<>) { ^I($a,$b)=split; ^I@h{$a,$b}=(); ^Inext if $a eq $b; ^I$i{$a,$b}=1 } @t=keys %h; if (keys %i){ ^I$c=@t*@t; ^IO:for $p(0..$#t){ ^I^Idie if !$c--; ^I^Ifor $j($p+1..$#t) { ^I^I^I@t[$p,$j]=@t[$j,$p],redo O if $i{$t[$j],$t[$p]} ^I^I} ^I} } print for @t
			Les premiers seront les derniers et lycée de Versailles.
252.52	Michael Wrenn	Thu Jul 4 22:10:35 2002	-lna &q($~,($^,$~)=@F);map{for$^(&k($_)){&q($~,$^,$_)}}&k($~);map{for$^(keys%k){$~eq$^&&die$!or&q($^,$_,$_)if$k{$^}{$~}}}&k($~)}@p=sort{$c{$a}<=>$c{$b}}keys%k;print for &k($p[0]),@p;{sub q{$k{$a=$_[0]}{$_[1]}||=++$c{$a}if$^ne pop}sub k{keys%{$k{+pop}}}
			Can I go back to Beginner now? Do I at least get my choice of wooden spoons? At least I'm under 4A (4 x Alien)!
261.41	Amir Karger	Tue Jul 2 21:05:21 2002	while (<>) { ($l, $r) = split; $pairs{$l}{$r} = 1; $pairs{$r}{$r} = 1; } for (keys %pairs) { for $node (keys %pairs) { next if (grep {$pairs{$_}{$node}} keys %pairs) - !!$pairs{$node}{$node}; print "$node\n"; delete $pairs{$node}; } } die if %pairs;
			A bit shorter without the niceties. But is this shortenable by 150 or so? I doubt it.
265.46	Amir Karger	Tue Jul 2 16:14:06 2002	$str = join"",<>; 1 while $str =~ s/^(\S+) \1$/$1/gm; while ((($parent)=grep {!($str=~/ \Q$_\E$/m)} $str =~ /^(\S+) /gm)) { $str =~ s/^\Q$parent\E( (\S+)|\n)/$2/gme and print "$parent\n"; } for($str=~/(\S+)/g){print "$_\n" unless $a{$_}++} exit 1 if $str =~/ /;
			This is the easy part...
267.47	Amir Karger	Wed Jul 3 18:41:05 2002	-nal ($p, $c) = @F; $parent{$p}+=0; $parent{$c}++ if $c ne $p && !$child{$p}{$c}++ } sub gr{grep{!$parent{$_}}@all=keys%parent} @orphan = gr; while (defined($_ = pop @orphan)) { print; push @orphan, grep{!--$parent{$_}} keys %{$child{$_}} } die if gr!=@all; {
			Ready or not, here I come!
289.49	Michael Wrenn	Sat Jul 6 15:37:29 2002	-l while(<>){ chomp; push(@l,$_); } for(@l){($s,$t)=split;$u{$s}=$u{$t}=1} @u=keys%u; $g=1; while($g&&(++$h<50*@l||die$h)){ for(@l){ $g=0; ($j,$k)=split; $x=z($j)-1; $y=z($k)-1; if($x>$y){@u[$x,$y]=@u[$y,$x];$g=1;last} } } print for@u; sub z {$c=0;for(@u){if($_[0] eq$u[$c++]){return$c}}0}
			Before I do the fun part, does this pass? I tries hashes, regex and splice ... now swapping. Where am I supposed to find that magic algo?
301.38	BigrTex	Tue Jul 2 19:12:37 2002	-l while(<>){($l,$r)=@l=split;exists$pairs{$l}{$r}?next:$pairs{$l}{$r}++;$npred{$l}+=0;if($l ne$r){++$npred{$r};push@{$succ{$l}},$r}}@list=grep{!$npred{$_}}keys%npred;while(@list){$_=pop@list;print;for$child(@{$succ{$_}}){push@list,$child unless--$npred{$child}}}if(grep{$npred{$_}}keys%npred){exit 1}
			trimmed off 73 strokes (20%)... I'm on the board which is always nice, but I can't tell if I'll still be in the money on Sunday...
334.47	Amir Karger	Wed Jul 3 18:32:39 2002	-nal ($p, $c) = @F; $parent{$p}+=0; next if $c eq $p; $parent{$c}++ if !$child{$p}{$c}++ } sub gr{grep{!$parent{$_}}@all=keys%parent} @orphan = gr; while (defined($node = pop @orphan)) { print "$node"; for $kid (keys %{$child{$node}}) { $parent{$kid}--; ^Ipush @orphan, $kid if !$parent{$kid} } } die if gr!=@all; {
			Yet another another algorithm. This one stolen indirectly from Knuth.
374.42	BigrTex	Tue Jul 2 18:56:35 2002	-l while(<>){ my($l,$r)=my@l=split; next if defined $pairs{$l}{$r}; $pairs{$l}{$r}++; $npred{$l}+=0; if($l ne $r){ ++$npred{$r}; push @{$succ{$l}},$r}} my @list=grep{!$npred{$_}} keys %npred; while (@list) { $_ = pop @list; print; foreach my $child (@{$succ{$_}}){ unshift @list, $child unless --$npred{$child}; } } if (grep {$npred{$_}} keys %npred) { exit 1; }
			It passed the 1.6 testcase... It's a lot shorter than a lot of the adjustments that I tried to make to accomodate those silly a->a loops... It's based on code from http://www.perl.com/language/ppt/src/tsort/tcsort.html
380.46	Kristen Thelen	Thu Jul 4 07:13:18 2002	sub visit{my$z=$_[0];return if@{$l{$z}}[0]eq 2;exit 1if@{$l{$z}}[0]eq 1;@{$l{$z}}[0]=1;$c=0;foreach(@{$l{$z}}){visit($_)if$c ne 0;$c++}@{$l{$z}}[0]=2;unshift(@o,$z);}while(<>){($f,$s)=split;push@{$l{$f}},0if!@{$l{$f}};push@{$l{$s}},0if!@{$l{$s}};if($f ne$s){$g=0;$c=0;for (@{$l{$f}}){$g=1if($c ne 0)&&($_ eq$s);$c++}push @{$l{$f}}, $s if!$g;}}visit($_)for keys%l;print "$_ "for@o;
380.46	Kristen Thelen	Thu Jul 4 15:47:25 2002	sub visit{my$z=$_[0];return if@{$l{$z}}[0]eq 2;exit 1if@{$l{$z}}[0]eq 1;@{$l{$z}}[0]=1;$c=0;foreach(@{$l{$z}}){visit($_)if$c ne 0;$c++}@{$l{$z}}[0]=2;unshift(@o,$z);}while(<>){($f,$s)=split;push@{$l{$f}},0if!@{$l{$f}};push@{$l{$s}},0if!@{$l{$s}};if($f ne$s){$g=0;$c=0;for (@{$l{$f}}){$g=1if($c ne 0)&&($_ eq$s);$c++}push @{$l{$f}}, $s if!$g;}}visit($_)for keys%l;print "$_ "for@o;
390.45	Peter Haworth	Thu Jul 4 17:18:23 2002	-pla ($l, $r) = @F; $npred {$_} += 0for@F; next if $l eq $r; $pairs{$l}{$r}++; ++$npred{$r}; push @{$succ{$l}}, $r; } @list = grep !$npred{$_}, keys %npred; while (@list) { $_ = pop @list; push @order,$_; foreach my $child (@{$succ{$_}}) { push @list, $child unless --$npred{$child}; } } die if grep $npred{$_}, keys %npred; $_=join$/,@order; {
			There I was thinking I needed recursion, when all I needed was to steal from tchrist's ppt :-) Lets see how easy this is to shorten
455.43	Amir Karger	Tue Jul 2 16:08:31 2002	-w use strict; my $str = join"",<>; my $parent; my $count; # remove self-rules 1 while $str =~ s/^(\S+) \1$/$1/gm; # While there's a parent with no parents, print it and get rid of it while ((($parent)=grep {!($str=~/ \Q$_\E$/m)} $str =~ /^(\S+) /gm)) { $str =~ s/^\Q$parent\E( (\S+)|\n)/$2||""/gme and print "$parent\n"; $count++; } # Print remaining children my %a =map {$_,1} $str=~/(\S+)/g; print join("\n",keys %a),"\n"; exit 1 if $str =~/ /;
			And so it begins...
551.48	Kristen Thelen	Thu Jul 4 06:33:59 2002	my @l, @o;sub visit{if (@{$l{$_[0]}}[0] eq 2){return;}if (@{$l{$_[0]}}[0] eq 1){exit 1;}@{$l{$_[0]}}[0]=1;$count=0;foreach $v (@{$l{$_[0]}}){if ($count ne 0){visit($v);}$count++;}@{$l{$_[0]}}[0]=2;unshift(@o, $_[0]);return;}while (<>) {($f, $s) = split;if (!@{$l{$f}}){push @{$l{$f}}, 0; }if (!@{$l{$s}}){push @{$l{$s}}, 0; }if ($f ne $s){$found = 0;$count =0;foreach $item (@{$l{$f}}){if (($count ne 0) && ($item eq $s)){$found = 1;}$count++;}if (!$found){push @{$l{$f}}, $s; }}}foreach $key (keys %l){visit($key);}foreach $i (@o){print "$i ";}
632.37	Amir Karger	Tue Jul 2 21:04:39 2002	-w use strict; my %pairs; while (<>) { my ($l, $r) = split; # If it has children, set them. Otherwise, make a self-link $pairs{$l}{$r} = 1; $pairs{$r}{$r} = 1; } # Try n times for (keys %pairs) { # Find nodes without parents (self-link doesn't count as a parent) for my $node (keys %pairs) { ^I# Node has more parents than just itself. Scalar grep is # of parents next if (grep {$pairs{$_}{$node}} keys %pairs) - !!$pairs{$node}{$node}; ^Iprint "$node\n"; ^Idelete $pairs{$node}; # Node is no longer a parent } } # If anything is left, we couldn't get rid of all nodes: cycle! die "cycle" if %pairs;
			Totally different algorithm, stolen from a web CS class. Shortenable?
3786.99	André Savigne	Sun Jul 7 08:26:09 2002	eval eval '"'. ('['^',').('`'| '(').('`'|')').("\`"| ',').('`'|'%').'('.('<'). '>'.')'.'\\'.'{'.('!'^('+')). '\\'.'$'.('`'|')').'='.('`'|')' ).('`'|'.').('`'|'$').('`'|'%').( '['^'#').'\\'.'$'.'_'.','."'".('{'^ '[')."'".';'.('!'^'+').'\\'.'$'.('['^ '#').'='.('['^'(').('['^'.').('`'|'"'). ('['^'(').('['^'/').('['^')').'\\'.('$'). '_'.','.('^'^('`'|'.')).','.'\\'.'$'.("\`"| ')').';'.('!'^'+').'\\'.'$'.('['^'"').('=').( '['^'(').('['^'.').('`'|'"').('['^'(').('['^'/' ).('['^')').'\\'.'$'.'_'.','.'\\'.'$'.('`'|"\)"). '+'.('^'^('`'|'/')).';'.('`'|'#').('`'|'(').('`'| '/').('['^'+').'\\'.'$'.('['^'"').';'.("\!"^ '+'). '\\'.'@'.('`'^'&').'='.'('.'\\'.'$'.(('[')^ '#') .','.'\\'.'$'.('['^'"').')'.';'.('!'^('+')). '\\'. '$'.('['^'+').'\\'.'{'.'\\'.'$'.'_'.'\\'.'}'. ('+'). '+'.('`'|'/').('['^')').'\\'.'@'.('`'|'.').'\\'.'{'.''. '\\'.'@'.('`'^'&').'\\'.'}'.'+'.'='.'\\'.'$'.('['^'#'). ('{'^'[').('`'|'.').('`'|'%' ).'\\'.'$'.('['^'"').'&'.'&' .'!'.'!'.('['^'+').('['^'.' ).('['^'(').('`'|'(').'\\'. '@'.'\\'.'{'.'\\'.'$'.('['^ '(').'\\'.'{'.'\\'.'$'.('[' ^'#').'\\'.'}'.'\\'.('}'). ','.'\\'.'$'.('['^('"')).( '!'^'+').'\\'.'}'.(('!')^ '+').'\\'.'@'.('`'|"\,"). '='.('`'|"'").('['^')').( '`'|'%').('['^'+').('!'). '\\'.'$'.('`'|"\.").'\\'. '{'.'\\'.'$'.'_'.'\\'.'}' .','.('`'|'+').('`'|'%'). ('['^'"').('['^'(').'%'.( '`'|'.').';'.('!'^('+')).( '['^'+').('['^')').(('`')| ')').('`'|'.').('['^('/')). '('.'\\'.'$'.'_'."\.".'\\'. '$'.'/'.')'.','.('[' ^'+') .('[' ^'.').('['^'(').('`' |'(').'\\'.'@'.('`' |',' ).(( ',')).('`'|("'")).( '['^')').('`'|'%') . ( '['^'+').'!'."\-". '-'.'\\'.'$'.('`'| '.').'\\'.'{'.'\\' .'$'.'_'.'\\'.'}'. ','.'\\'.'@'.'\\'. '{'.'\\'.'$'.('['^ '(').'\\'.'{'.'\\' .'$'.'_'.'\\'.('}'). '\\'.'}'.('`'|'&').( '`'|'/').('['^"\)"). '\\'.'@'.('`'|"\,"). ';'.('!'^'+') .'\\' .'@'. ('`'|',').'-' .('`'|'+').( '`'| '%') .('['^'"').( '['^('(')). ( '%').('`' | '.').('&'). '&'.(('`')| '%').('['^'#') .('`'|')'). ('['^"\/"). '('.('^'^('`'|'/' )).')'.('!' ^'+').'"';$: ='.'^'~';$~='@'|'('; $^=')'^"\["; $/='`'|'.';$_ ='('^'}';$,='`'|'!';$\ =')'^"\}";$:= '.'^('~');$~= '@'|'(';$^=')'^('[');$/= '`'|('.');$_= ('(')^ "\}"; $,='`'|'!';$\=')'^"\}";$:= "\."^ '~';$~ ='@'| '('; $^=')'^"\["; $/= '`'| "\."; ($_) = '('^"\}";$,= '`' | '!'; $\= (( ')'))^"\}"; $:= '.' ^+ '~' ;( ($~))= '@' |+ ( '(' ) ;$^= ')' ^ (( ( ( ( ( ( ( ( ( ( ( ( ( ( ( '[' ))) ) ) ) ) ) ) ) )) ) ) ) ) ; $/ = ( ( ( '`')) ) | (( '.'));$_='(' ^ ( '}');$,=('`')| ( '!');$\="\)"^ '}' ;$:='.'^'~'; $~='@'|'(';
			Pictorial representation of golf contest between Ton Hospel and `/anick. Colorful but not very suspenseful.

Artistic/Unorthodox

Score	Golfer	Submit Time	Code
119.51	Eugene van der Pijll	Tue Jul 2 20:17:32 2002	-wlap0 $SIG{__WARN__}=[@g{@F}=$x=$_]}sub a{map{(1)x("@_"eq$_||a($_))}$x=~/(.*) \Q@_ /g,@_}for(sort{a($a)<=>a$b}keys%g){
			As the refs didn't like my memory eating recursion trick, I've added a signal handler. This solution needs the -w.
167.53	Philippe 'BooK' Bruhat	Thu Jul 4 15:33:41 2002	-lna END{@a=keys%a;T:$b=~/\Q@a/&&die;map{($c,$d)=@$_;@a{@a}=0..@a;$b.="^I@a",@a[@a{@$_}]=@a[@a{$d,$c}],goto T if$a{$c}>$a{$d}}@r;$"=$/;print"@a";exit}push@r,[@a{@F}=@F]
			There was probably a typo in the previous submission, since this once is ok on all tests for me. Except that it breaks the 1 minute barrier on test 27, naturally! ;-)
192.42	Philippe 'BooK' Bruhat	Wed Jul 3 15:53:31 2002	-lna @n{@F}=push@r,$_;$"=$/;sub p{my (@p,%s)=@{+pop};if(@p==keys%n){ print"@p";exit}@s{@p}=1..@p;for (@r){($a,$b)=split;return if$s{ $a}>$s{$b}}map{p([$_,@p])}grep{ !$s{$_}}keys%n}END{p[];die}
			Yet another unorthodox entry, where we try to find a solution using a brute force system. This brought test 22 out of memory, which seems to indicate that I need to rework this a bit.
220.43	Philippe 'BooK' Bruhat	Wed Jul 3 15:02:22 2002	-lna @n{@F}=push@r,$_;$"=$/;sub p{my @a;return[@_]if@_==1; for$0(@_){push@a,map{[$0,@ $_]}p(grep{$0ne$_}@_)}@a}; END{T:for(p(keys%n)){@s{@s }=0..(@s=@$_);for(@r){($a, $b)=split;next T if$s{$a}> $s{$b}}print"@s";exit}die}
			This is an improvement over the previous unorthodox entry! Now we don't try n^n possibilities, but only n! (factorial n). We will need less memory (but still a lot, since we test all), but a loooot of time to test all the possibilities. Maybe it's less unorthodox, since we don't use glob anymore.(by the way, the glob entry would not work with all chars, like *)
784.98	Max Tie	Sun Jul 7 18:40:38 2002	eval pack'b*','^I ^I ^I ^I^I^I ^I ^I ^I ^I^I ^I^I^I^I ^I ^I ^I^I^I^I^I^I ^I ^I^I^I^I ^I ^I ^I ^I ^I ^I^I^I^I^I ^I ^I ^I^I^I^I ^I ^I ^I^I^I^I ^I^I ^I^I^I ^I^I ^I^I ^I^I^I ^I ^I ^I^I^I^I^I ^I ^I ^I^I^I^I ^I^I^I^I ^I^I^I^I^I ^I^I ^I^I^I ^I^I ^I^I^I ^I^I^I^I ^I ^I^I^I^I ^I ^I ^I ^I^I^I ^I ^I^I ^I ^I ^I^I ^I ^I ^I ^I ^I ^I ^I^I^I ^I ^I ^I^I ^I ^I ^I^I^I^I ^I ^I ^I ^I ^I^I ^I^I^I^I ^I ^I ^I^I ^I^I ^I^I^I ^I^I ^I^I ^I^I^I ^I^I^I^I ^I ^I ^I^I^I ^I ^I ^I ^I ^I ^I ^I^I ^I ^I ^I ^I^I^I^I ^I ^I^I^I^I^I ^I^I^I^I^I ^I^I^I ^I ^I^I^I ^I ^I ^I^I ^I^I^I ^I^I ^I ^I^I^I ^I ^I ^I ^I ^I^I ^I ^I ^I ^I ^I ^I^I^I ^I^I^I ^I ^I^I ^I ^I ^I^I ^I^I ^I^I ^I ^I ^I^I ^I ^I^I ^I^I^I ^I^I^I^I ^I ^I^I^I^I ^I ^I^I^I ^I ^I ^I ^I ^I ^I ^I^I ^I ^I^I^I ^I ^I ^I ^I ^I^I^I ^I ^I^I ^I^I^I ^I^I^I^I ^I ^I^I^I^I ^I ^I ^I^I ^I^I ^I^I^I ^I^I ^I^I ^I ^I^I^I^I ^I ^I^I^I ^I ^I^I ^I ^I ^I^I ^I ^I ^I^I^I^I ^I ^I^I^I ^I^I ^I^I ^I^I^I ^I ^I ^I ^I ^I ^I^I^I^I ^I^I^I^I^I^I ^I^I^I^I^I^I ^I ^I '
			actually it's from my brother Min.

Rejected

Score	Golfer	Submit Time	Code
11.29	Wladimir Palant	Fri Jul 5 16:11:51 2002	exec'tsort'
			Unix systems only ;)
37.44	Martin Carlsen	Mon Jul 1 15:46:44 2002	print$_=`tsort<in.tmp 2>&1`;exit(/ /)
			This is of course cheating and works only on Unix. (but I thought I'd submit it anyway :-)
42.40	Mtv Europe	Wed Jul 3 18:20:59 2002	print"Hello, world!\n" # for Ton's fear!!!
			sorry for this dumb code, referees, i'm just trying to make trail in history that i'm in Paris now.
67.52	Ton Hospel	Mon Jul 1 23:31:51 2002	-ln0 /(?!\G) \Q$& /||print($&)+s/^(\Q$&\E\s)+/ /mgwhile/\S+/g;1/!//
			fails if the same singleton can appear multiple times in the input. Needs rule clarification
75.48	Prakash Kailasa	Mon Jul 1 19:18:06 2002	-lan $s=~s/\Q$F[1]\E |$/$_ /}{$s{$_}++&&die,print for$s=~/(\S+)(?: +\1)* /g
			just a little tweaking.
75.50	Prakash Kailasa	Tue Jul 2 04:41:11 2002	-lan $s=~s/ \Q$F[1]\E |$/ $_ /}{print,$s{$_}++&&&;for$s=~/(\S+)(?: +\1)* /g
76.58	Prakash Kailasa	Mon Jul 1 18:56:00 2002	-lan $x=~s/\Q$F[1]\E |$/$_ /}{map{$h{$_}++&&die;print}$x=~/(\S+)(?: +\1)* /g
			Wow! I can't believe myself. Beating Ton "The Alien" Hospel? Who'd have thunk? Let's see how long it lasts.
77.48	Prakash Kailasa	Mon Jul 1 21:03:57 2002	-lan $s=~s/ \Q$F[1]\E |$/ $_ /}{$s{$_}++&&die,print for$s=~/(\S+)(?: +\1)* /g
			This passes the new test program (v1.4) at the cost of two strokes :-( I am surprised that no one beat me to punch yet. I guess I'll enjoy the lead while I can.
77.51	Juho Snellman	Thu Jul 4 20:00:59 2002	-ln0 $_ x=2;s/(^| )(\Q$1\E\s)+/$1&&&1/megwhile/^(\S+)\s(?!\C* \1 )/m&&print$1
			Replace die() with a call to an undefined sub.
78.49	Juho Snellman	Thu Jul 4 18:06:53 2002	-ln0 $_ x=2;s-(^| )(\Q$1\E\s)+-$1&&die-megwhile/^(\S+)\s(?!\C* \1 )/m&&print$1
			I doubt that there's much left to gain by reorganizing the loop a third time.
78.50	Ton Hospel	Mon Jul 1 16:44:13 2002	-Xln0 /^(?!\Q$&\E ).* \Q$& /m||print$&|s/^(\Q$&\E\s)+/ /mgwhile/^\S+/gm;exit//
			This shows a problem with the test program
79.49	Juho Snellman	Thu Jul 4 17:56:18 2002	-ln0 $_ x=2;(print$1),s-(^| )(\Q$1\E\s)+-$1&&die-megwhile/^(\S+)\s(?!\C* \1 )/m
			It seems that the carefully crafted loop based on s///e and redo was rather suboptimal...
80.48	pom	Wed Jul 3 07:59:47 2002	-ln0 s/\b(.+) \1/$1/|/ \Q$&\E$/m||print($&)^s/^\Q$&\E\s//gm while/\S+/gm;$_&&die
			Hum, this one is a little shaky, I'm not sure it will pass all the tests... Anyway, I got a solution with 81 strokes if this one is not correct.
81.45	Juho Snellman	Thu Jul 4 17:43:57 2002	-ln0 s%^(\S+)\s(?!\C* \1 )%print$1;s-(^| )(\Q$1\E\s)+-$1&&die-meg;redo%emfor$_ x2
			Replaced the "$_ x=2;{foo}" with "foo for$_ x2".
81.51	Ton Hospel	Mon Jul 1 16:25:34 2002	-Xln0 /^(?!\Q$&\E ).* \Q$& /m||print"$&"|!s/^(\Q$&\E\s)+/ /mgwhile/^\S+/gm;exit//
82.47	Marko Nippula	Wed Jul 3 07:52:52 2002	-lna0 \sort{$z="\Q$a",/^(?!$z ).* $z /m||s/(\G|\s)$z\s/ /g&&print$a}(@F)x9;&#if/./
			Chopped two chars.
82.47	Juho Snellman	Tue Jul 2 20:01:36 2002	-ln0 $_ x=2;{s%^(\S+)\s(?!\C* \1 )%print$1;s-(^| )(\Q$1\E\s)+-$1&&die-meg;redo%em}
			Smarter test for loops, nicked from an earlier solution. Too bad, I really liked that sqrt...
83.45	Juho Snellman	Tue Jul 2 19:37:44 2002	-ln0 $_ x=2;{s%^(\S+)\s(?!\C* \1 )%sqrt$;{$1}--;print$1;s|^(\Q$1\E\s)*||mg;redo%em}
			Oops, remove an useless .*
83.47	Marko Nippula	Thu Jul 4 13:36:25 2002	-lna0 \sort{$z="\Q$a",/^(?!$z ).* $z /m||s/(\G|\s)$z\s/ /g&&print$a}(@F)x@F;&#if/./
83.52	Ton Hospel	Mon Jul 1 16:23:33 2002	-Xln0 /^(?!(\Q$&\E) ).* \Q$& /m||print"$&"|!s/^(\Q$&\E\s)+/ /mgwhile/^\S+/gm;exit//
84.45	Juho Snellman	Mon Jul 1 21:39:59 2002	-ln0 s/^(\S+)\s(?!.*((?!\1)\S+) \1 )/print$1;s%(^| )(\Q$1\E\s)+%$1&&die%meg;redo/sem
			Embed everything inside the main substitution. We aren't even a day into the contest, and already I don't understand what the code actually does.
84.47	Marko Nippula	Tue Jul 2 18:01:37 2002	-lna0 &#for(sort{$z="\Q$a",/^(?!$z ).* $z /m||s/(\G|\s)$z\s/ /g&&print$a}(@F)x9)x/./
			Totally new solution.
85.46	Juho Snellman	Tue Jul 2 18:42:01 2002	-ln0 $_ x=2;{s%^(\S+)\s(?!\C*.+ \1 )%sqrt$;{$1}--;print$1;s|^(\Q$1\E\s)*||mg;redo%em}
			I don't know why this works. Probably I don't quite understand the exact semantics of zero-width negative look-ahead. I'm not going to complain, though.
86.51	Ton Hospel	Mon Jul 1 16:08:47 2002	-Xln0 $;=$&and/^(?!\Q$;\E ).* \Q$; /m||s/^(\Q$;\E\s)+/ /mg*print$;while/^\S+/gm;exit//
88.51	Ton Hospel	Mon Jul 1 15:35:48 2002	-Xln0 $;=$+and/^(?!\Q$;\E ).* \Q$; /m||s/^(\Q$;\E\s)+/ /mg*print$;while/^(\S+)/gm;exit//
90.53	Ton Hospel	Mon Jul 1 15:33:18 2002	-ln0 $;=$+and/^(?!\Q$;\E ).* \Q$; /m||s/^(\Q$;\E\s)+/ /mg*print$;while/^(\S+)/gm;//&&die}{
91.53	Ton Hospel	Mon Jul 1 15:29:00 2002	-ln0 $;=$+and/^(?!\Q$;\E ).* \Q$; /m||s/^(\Q$;\E\s)+/ /mg*print$;while/^(\S+)/gm;/./&&die}{
91.54	pom	Tue Jul 2 15:47:50 2002	-0777n s/^(.+) \1$/$1/gm;while(/^\S+/gm){if(!/ $&$/m){print"$&\n";s/^$&[ \n]//gm}}/\S/&&die
			Woohoo! Under 100! And there is still room for improvement...
91.55	Juho Snellman	Tue Jul 2 09:46:57 2002	-p0 1/(1e5-$z++);s|^((\S+) (\S+) )(.*\S+ \2 )|$4$1|msg&&redo;y/ / /;s/^(\S+ )(?=.*^\1)//gsm
			Just testings the waters, before I spend any more time on this idea. This program fails for some inputs of over 100000 lines. If this entry isn't accepted, a clarification on the maximum input size would be appreciated.
92.45	Juho Snellman	Tue Jul 2 00:48:07 2002	-ln0 $_ x=2;s%^(\S+)\s(?!.*((?!\1)\S+) \1 )%sqrt$n{$1}--;print$1;s|^(\Q$1\E\s)*||mg;redo%sem
			Heh, wouldn't have thought that sqrt would be useful on this hole :-)
93.48	Juho Snellman	Mon Jul 1 23:26:05 2002	-ln0 $_.=$_;s/^(\S+)\s(?!.*((?!\1)\S+) \1 )/$n{$1}++&&die;print$1;s|^(\Q$1\E\s)*||mg;redo/sem
			Works even with "a b, b c, d a", unlike the 84.
93.56	Prakash Kailasa	Mon Jul 1 18:49:53 2002	-lan $x=~s/\Q$F[1]\E |$/$_ /}{$x=~s/(\S+)( +\1)+ /$1 /g;map{$h{$_}++&&die;print}@x=$x=~/\S+/g
97.46	Juho Snellman	Mon Jul 1 20:13:28 2002	-ln0 s/$/\t $`/;s/^(\S+)\s(?!.*((?!\1)\S+) \1 ).*\t/print$1;s|^(\Q$1\E\s)*||mg;redo/sem;/\S/&&die
			Join the sub-100 club using some regex merging.
99.46	Jukka Suomela	Thu Jul 4 16:02:31 2002	-alp my($x,$y)=map{bless$x{$_}||=[$_]}@F;$x-$y&&push@$x,$y}DESTROY{$d?dump:print$_[0][0]}{%x=0;$d=q
101.44	Juho Snellman	Mon Jul 1 18:49:08 2002	-ln0 s/^(.+) \1$/$1/gm;s/$/\t $`/;s/^(\S+)\s(?!.* \1 ).*\t/print$1;s|^\Q$1\E\s||mg;redo/sem;/\S/&&die
			Yesss my preciousss..... We takessss back the second place...
102.48	Ton Hospel	Mon Jul 1 14:46:53 2002	-ln0 $;=$+and/ \Q$; /||s/^\Q$;\E\s/ /mg*print$;while/^(\S+)/gm;print$+while s/^(\S+) \1$//m;/./&&die}{
103.51	Eugene van der Pijll	Tue Jul 2 17:47:11 2002	-lap0 @g{@F}=0;$x=$_}sub a{map{(1)x("@_"eq$_||a($_))}$x=~/(.*) \Q@_ /g,@_}for(sort{a($a)<=>a$b}keys%g){
			Nasty...
104.49	Ton Hospel	Mon Jul 1 14:05:17 2002	-ln0 $;=$+and/ \Q$; /||s/^ ?\Q$;\E\s/ /mg*print$;while/^(\S+)/gm;print$+while s/^(\S+) \1$//m;/./&&die}{
104.50	Chris Dolan	Sun Jul 7 21:32:53 2002	-nal push@{$a{pop@F}},@F}&a for%a;sub a{$.--<0&&die;$z eq$_||&a for@{$a{$z=$_}};$.++;$o{$_}++.ref||print
106.49	David Lowe	Wed Jul 3 23:07:45 2002	-lna @w{@F}=@F;push@g,"@F"}{&%if map{print,delete$w{$w},redo if$w=$_,!grep/ \Q$w\E$/&&$`ne$w&&$w{$`},@g}%w
			grrgble
107.61	Ala Qumsieh	Mon Jul 1 18:58:58 2002	-lpa $x=~/ \Q$F[1]\E (.* )*\Q$F[0] /&&die;$x=~s/ \Q$F[1] / $_ / or$x.=" $_ "}for(grep!$x{$_}++,$x=~/\S+/g){
			Sometimes I feel sooooo stupid!
108.47	Albert Dvornik	Wed Jul 3 22:43:28 2002	-l -034 $_=<>;s/^(.*) \1$/$1/mg;A:while($_){for$k(split){/ \Q$k\E$/m ||s/^\Q$k\E\s//mg+print($k)+next A}die}
			the tiebreak score helps again.
108.49	Juuso Salonen	Wed Jul 3 09:21:14 2002	-ln0 s// /;/(\S+)/g,($a=$1)?/(?<! \Q$a\E) \Q$a /?0:do{while(s/^\Q$a\E\s//m){};print$a;//?3:exit}:die while 3
108.51	Chris Dolan	Fri Jul 5 22:14:42 2002	-nal push@{$a{pop@F}},@F}sub a{$c{$_}++>2**$.&&die;$z eq$_||a()for@{$a{$z=$_}};$o{$_}++.ref||print}for(%a){a
			Slow: takes just over 2 minutes on a PIII 866.
108.51	Juuso Salonen	Wed Jul 3 09:50:30 2002	-ln0 s// /;/(\S+)/g,($a=$1)?/(?<! \Q$a\E) \Q$a /?0:do{while(s/^\Q$a\E\s//m){};print$a;//?3:exit}:1/$/while 1
			Sorry for sending almost the same thing, just hard to debug the code as it passes the test program and all the tests I can think of :)
108.61	Jukka Suomela	Tue Jul 2 19:55:08 2002	-p / (.*)/;bless$x{$_}||={"",($_)x3}for$`,$1;$x{$`}{$1}||=$x{$1}}DESTROY{$0?$\.=$_[0]{""}.$/:dump}{%x=1;$0=0
			This code is NOT valid. It fails on a line consisting of "0 0" only. However, it passes the test script.
110.50	Eugene van der Pijll	Wed Jul 3 17:05:00 2002	-lap0 @g{@F}=$x=$_}sub a{map{(1)x("@_"eq$_||a($_))}$x=~/(.*) \Q@_ /g,@_}alarm 20;for(sort{a($a)<=>a$b}keys%g){
110.61	Ala Qumsieh	Mon Jul 1 18:47:53 2002	-lpa $x=~/ \Q$F[1]\E (.* )*\Q$F[0] /&&exit$=;$x=~s/ \Q$F[1] / $_ / or$x.=" $_ "}for(grep!$x{$_}++,$x=~/\S+/g){
			A little re-arrangement saved me 14 chars!
111.51	David Lowe	Wed Jul 3 09:04:51 2002	-lna push@g,[@w{@F}=@F]}{&%if map{print,delete$w{$w},redo if$w=$_,!grep$$_[0]ne$w&&$w{$$_[0]}&&$$_[1]eq$w,@g}%w
			running out of steam time for a new approach, I think...
111.51	Juuso Salonen	Tue Jul 2 17:36:16 2002	-ln0 s/^(\S+) \1$/$1/gm;/(\S+)/g,($a=$1)?/ \Q$a\E /?0:do{while(s/^\Q$a\E\s//m){};print$a;//?3:exit}:3/$/while 3
111.54	Prakash Kailasa	Mon Jul 1 18:43:20 2002	-lan $x=~s/ \Q$F[1]\E |$/ $_ /}{$x=~s/(\S+)( +\1)+ /$1 /g;map$h{$_}++&&die,@x=$x=~/\S+/g;map{$s{$_}++||print}@x
112.51	David Lowe	Wed Jul 3 05:58:44 2002	-lna push@g,[@w{@F}=@F]}{&f if map{print,delete$w{$w},redo if$w=$_,!grep$$_[0]ne$w&&$w{$$_[0]}&&$$_[1]eq$w,@g}%w
112.52	Martin Carlsen	Mon Jul 1 23:06:36 2002	-p0 $n++>8*y/ //&&die while s/^((\S+) \S+ )(.*^\S+ \2 )/$3$1/sm||s/( \S+)( .*\1 )/$2/s||s/ / /||s/^(\S+ )\1/$1/m
			Yay! Not a single ; in sight
112.52	Martin Carlsen	Mon Jul 1 23:07:18 2002	-p0 $n++>8*y/ //&&die while s/^((\S+) \S+ )(.*^\S+ \2 )/$3$1/sm||s/( \S+)( .*\1 )/$2/s||s/ / /||s/^(\S+ )\1/$1/m
			Yay! Not a single ; in sight
116.64	Ala Qumsieh	Mon Jul 1 18:12:12 2002	-lpa $x=~s/ $F[1] / $_ / or$x.=" $_ ";/^(\S+) \1$/||$x=~/ $F[1] (.* )*$F[0] /&&exit$=}for(grep!$x{$_}++,$x=~/\S+/g){
117.48	Albert Dvornik	Wed Jul 3 22:22:48 2002	-l -034 ($l=<>)=~s/^(.*) \1$/$1/mg;while($l){map+($l!~/ \Q$_\E$/m and$l=~s/^\Q$_\E\s//mg,print,next),split' ',$l;die}
			Hmm, sometimes the tie break score does matter...
117.49	David Lowe	Tue Jul 2 18:18:19 2002	-lna push@g,[@w{@F}=@F]}{(map{print,delete$w{$w},redo if$w=$_,!grep{$$_[0]ne$w&&$w{$$_[0]}&&$$_[1]eq$w}@g}keys%w)&&&f
118.48	David Lowe	Mon Jul 1 23:30:48 2002	-lna push@g,[@w{@F}=@F]}{(map{print,delete$w{$w},redo if$w=$_,!grep{$$_[0]ne$w&&$w{$$_[0]}&&$$_[1]eq$w}@g}keys%w)&&die
118.62	Ala Qumsieh	Mon Jul 1 18:06:25 2002	-lpa $x=~s/ \Q$F[1] / $_ / or$x.=" $_ ";$F[0]ne$F[1]&&$x=~/ $F[1] (.* )*$F[0] /&&exit$=}for(grep!$x{$_}++,$x=~/\S+/g){
119.46	David Lowe	Tue Jul 2 22:28:28 2002	-lna push@g,[@w{@F}=@F]}&a;sub a{&a if map{print,delete$w{$w}if$w=$_,!grep{$$_[0]ne$w&&$w{$$_[0]}&&$$_[1]eq$w}@g}keys%w
			death by stack overflow
119.47	TheodoreYoung	Wed Jul 3 01:48:20 2002	-lan END{print for @A}@H{@A}=reverse 1..@A;pop@F if$F[0]eq$F[1];($X,$Y)=@H{@F};$X&&$Y?$X<$Y&&die:splice@A,@A-$X-$Y,1,@F
			My first time here!
119.49	David Lowe	Mon Jul 1 23:29:00 2002	-lna push@g,[@w{@F}=@F]}{(map{print(delete$w{$w}),redo if$w=$_,!grep{$$_[0]ne$w&&$w{$$_[0]}&&$$_[1]eq$w}@g}keys%w)&&die
121.51	Michael Thelen	Mon Jul 1 21:25:31 2002	-ln0 print$1while s/^(\S+) \1$//m;while(/\S/){$t=0;for$m(/^\S+/mg){/ \Q$m\E$/m||s/^\Q$m\E[ ]//mg&&++$t&&print$m}$t||die}
			Regex solutions kick butt. I think this one can be improved greatly.
121.55	Keith Calvert Ivey	Tue Jul 2 11:40:51 2002	-p0 s/ (.+) /$&$1 /g;s/^(.+) \1 //gm;while(%f=/ (.+)( )/g,($a)=grep!$f{$_},/(.+) /g){s/^\Q$a\E .* //gm;s/$/ $a/}/ /&&die
			Need to submit something after a whole day in last place.
122.56	Wladimir Palant	Sat Jul 6 20:53:21 2002	-lp0 while(/(\S+) (\S+)/g){/^/g,$a{$2}++,a%(200/$a{$2})if($a{$1}+=a)>=$a{$2}&&$1ne$2}$_=join"\n",sort{$a{$a}-$a{$b}}keys%a
			Hm, it passes the test program but it could be a bit too slow :)
123.49	Juuso Salonen	Tue Jul 2 09:12:26 2002	-ln0 s/^/ /while s/^(\S+) \1$/$1/m;/(\S+)/g,($a=$1)?/\Q $a\E$/m?0:do{while(s/^\Q$a\E\s//m){};print$a;//?3:exit}:3/$/while 1
			Curse the special cases :)
124.53	Sec	Tue Jul 2 07:50:36 2002	-lan @t{@F}.=" $F[0] "if$F[0]ne$F[1];@q{@F}=@F}for(keys%q){$t{$_}||$,or$t{$,=$_}+=print for%q;$,=$,?!map{s/ \Q$, //g}%t:die;
			My last submission got rejected. Have to try again...
124.54	Martin Carlsen	Tue Jul 2 09:01:05 2002	-p0 $n++>8*y///c&&die while s/^((\S+) \S+ )(.*^(?!\2 )\S+ \2 )/$3$1/sm||s/( \S+)( .*\1 )/$2/s||s/ / /||s/^((\S+ ).*^)\2/$1/m
124.63	Ala Qumsieh	Mon Jul 1 18:26:04 2002	-lpa $x=~s/ \Q$F[1] / $_ / or$x.=" $_ ";/^(\S+) \1$/||$x=~/ \Q$F[1]\E (.* )*\Q$F[0] /&&exit$=}for(grep!$x{$_}++,$x=~/\S+/g){
			Damn special characters!
125.48	Juho Snellman	Mon Jul 1 14:17:50 2002	@a=<>;{map{s/^(.*) \1$/$1/;($a,$b)=split;if(!grep/ $a$/,@a){print"$a ";@a=grep!/^$a(\s|$)/,@a;$b&&push@a,$b;redo}}@a;@a&&die}
125.54	Yanick Champoux	Fri Jul 5 01:25:02 2002	-lpa push@a,[@F[$_..1]]for$F[0]eq$F[1],1}while(/.+/,@a=grep{!grep$&eq$_,@$_}@a){($_)=map{(($x)=@$_)x!grep$$_[1]eq$x,@a}@a or&
			A Perl golf solution without a regex is not a true Perl golf solution.
126.51	Qingning Huo	Mon Jul 1 11:51:08 2002	-ln0 while (/^\S+/gm) { ($y = $&) =~ s/\./[.]/g; s/^$y\s//gm,s/^$y\s//gm if /$y/ & !/^(\S+) (?!\1$)$y$/m && print$&; } $_&&die
			It passed the test program (1.0)
127.54	Yanick Champoux	Fri Jul 5 00:19:17 2002	-lpa push@a,[@F[$_..1]]for$F[0]eq$F[1],1}while(@a=grep{!grep$x eq$_,@$_}@a){$_=(($x)=map{(($x)=@$_)x!grep$$_[1]eq$x,@a}@a)?$x:&
			I think it's about time I have a satori. Or an epiphany. Or just a good idea (I'm not picky).
127.54	Sec	Tue Jul 2 00:00:27 2002	-lan $t{$F[1]}.=" $F[0] "if$F[0]ne$F[1];@q{@F}=@F}for(keys%q){$t{$_}||$,or$t{$,=$_}+=print for%q;$,=$,?!map{s/ \Q$, //g}%t:die;
			Creative use of ?: and loosing a space shoul put me in the top 10 for now. - Time for bed!
128.54	Yanick Champoux	Thu Jul 4 02:52:04 2002	-lpa push@a,[@F[$_..1]]for$F[0]eq$F[1],1}while(@a=grep!grep($x eq$_,@$_),@a){$_=(($x)=map{(($x)=@$_)x!grep$$_[1]eq$x,@a}@a)?$x:&
			Argh. That's all I have to say: argh.
128.59	Andrew Savige	Fri Jul 5 05:34:07 2002	-aln @n{@F}+=!s/^(.+) \1$//&!$:{$_}++}{0-$n{$x=$_}or++$z,print,map/^\Q$x\E (.+)/&&--$n{$1},%:,"$x $x"for(@a=keys%n)x200;$z-@a&&&
			Finally, I understand what this hard-wired 200 limit business is about! This solution is quite disgusting but I am curious to know if the referees accept such a solution or not.
129.54	Yanick Champoux	Thu Jul 4 01:05:56 2002	-lpa push@a,[@F[$_..1]]for$F[0]eq$F[1],1}while(@a=grep!grep($x eq$_,@$_),@a){$_=(($x)=map{($x=@$_[0])x!grep$$_[1]eq$x,@a}@a)?$x:&
			Why be dirty when we can be dirtier?
129.55	Dave Hoover	Wed Jul 3 17:41:12 2002	-pal @g{@F}=/^\Q$F[1] /||$f{$F[1]}{$F[0]}--}sub Y{"@_"=~/ \Q$_[0] /?die:@_,map{Y($_,@_)}%{$f{$_[0]}}}for(sort{Y($a)- Y$b}keys%g){
			This would fail if the following test were added: [<
131.52	Andrew Savige	Fri Jul 5 07:41:53 2002	-aln @n{@F}+=!s/^(.+) \1$//&!$:{$_}++}{@l=grep!$n{$_},@m=keys%n;$x=$_,print,push@l,grep!--$n{$_},map/^\Q$x\E (.+)/,%:for@l;@l-@m&&&
			What an interesting game! Much of the golfing elite (such as Eugene, Stephen and Mtv) seem to be stuck in the mud, while a few insightful beginners and lesser known golfers have broken through to the 80s and below. I suspect I am using a similar approach to those in the 120's and 130's and that Stephen and Eugene, being brilliant golfers, have golfed this well-known and straightforward approach down to a limit of 124.56. No, to get to 80 will require a change of horse. So, instead of wasting more time golfing this approach, I should go the stream, grab a fresh, wet trout and slap myself about the face with it.
131.52	Sec	Mon Jul 1 23:46:29 2002	-lan $t{$F[1]}.=" $F[0] "if$F[0]ne$F[1];@q{@F}=@F}for(keys%q){$t{$_}||$d or$t{$d=$_}+=print for%q;$d||die;map{s/ \Q$d //g}%t;$d="";
			Use that nice 5.6.1 feature to modify hash values
131.62	Marko Nippula	Mon Jul 1 09:02:52 2002	-pal $g=~s/(^| )\Q$F[1]\E( |$)/$1@F$2/g||($g.=" @F")}{$_=$g;s/(\G| )(\S+)(?= \2( |$))//g;s/^ //;y/ / /;/(^| )(.*) .* \2( |$)/s&&die
			First working solution.
132.50	Andrew Savige	Thu Jul 4 12:50:30 2002	-aln @n{@F}+=!s/^(.+) \1$//&!$:{$_}++}{@l=grep!$n{$_},keys%n;$x=$_,print,push@l,grep!--$n{$_},map/^\Q$x\E (.+)/,%:for@l;@l-keys%n&&&
			Using the call to undefined subroutine ';' (sic) to save 2 strokes over die. For golf historians, this trick was made infamous by Piers Cawley in the inaugural Santa Claus tournament. Gee, `/ is playing well, he might even overtake the great Eugene van der Pijll and Stephen Turner! I can't catch him. :-(
132.56	Dave Hoover	Wed Jul 3 17:37:45 2002	-pal @g{@F}=/^\Q$F[1] /||($f{$F[1]}{$F[0]}='')}sub Y{"@_"=~/ \Q$_[0] /?die:@_,map{Y($_,@_)}%{$f{$_[0]}}}for(sort{Y($a)- Y$b}keys%g){
			'die' is a lot better than 'exit 1'!
132.56	Dave Hoover	Thu Jul 4 00:25:12 2002	-pal @g{@F}=/^\Q$F[1] /||($f{$F[1]}{$F[0]}='')}sub Y{"@_"=~/ \Q$_[0] /?die:@_,map{Y($_,@_)}%{$f{$_[0]}}}for(sort{Y($a)- Y$b}keys%g){
			If this does not pass, please tell me why (provide test that failed).
133.52	Damien Neil	Wed Jul 3 01:15:21 2002	-lna for(@F){$_=$h{$_}||=bless[$_]} push@{$F[0]},$F[1]if($F[0]!=$F[1]); DESTROY{print$_[0][0];grep(!$_,@{$_[0]})&&exit 1;} END{%h=()}
133.57	Alexander Onokhov	Tue Jul 2 10:46:46 2002	-l map{/ (.+)/;$h{$`}={$1,1,map%$_,@h{$`,$1}};$h{$1}||=1}(<>)x99;map{print}sort{$h{$a}{$b}?($h{$b}{$a}?exit 1:-1):$h{$b}{$a}&1}keys%h
134.49	Prakash Kailasa	Mon Jul 1 18:40:12 2002	-lan $x=~s/ \Q$F[1]\E |$/ $_ /}{$x=~s/(\S+)( +\1)+ /$1 /g;map{die if $h{$_};$h{$_}++}@x=$x=~/\S+/g;map{print unless $s{$_};$s{$_}++}@x
135.47	dragonchild	Tue Jul 2 18:22:35 2002	-lna @k{($d,$k)=@F}=1;push@$k,$d if$d ne$k;END{@x=keys%k;for$y(@x){$d++>0xFFF&&die 1; map$d{$_}||push(@x,$y)&&next,@$y;$d{$y}=print$y}}
135.55	Eugene van der Pijll	Tue Jul 2 00:58:10 2002	-lap0 @g{@F}=0;$_ x=@F;$2eq$1or$g{$2}>$g{$1}or$g{$2}=$g{$1}+1while/(.*) (.*)/g}1/!grep$g{$_}>@o,@o=keys%g;for(sort{$g{$a}<=>$g{$b}}@o){
			Eureka!
136.51	Chris Dolan	Tue Jul 2 04:11:31 2002	-pal sub a{($x=$n{my$z=pop})>9&&die;$z eq$_||$n{$_}>$x||a($n{$_}=$x+1,$_)for@{$a{$z}}}$a{$F[1]}=[@F],a@F}for(sort{$n{$b}-$n{$a}}keys%a){
136.56	Dave Hoover	Wed Jul 3 17:16:33 2002	-pal @g{@F}=/^\Q$F[1] /||($f{$F[1]}{$F[0]}='')}sub Y{"@_"=~/ \Q$_[0] /&&exit 1;@_,map{Y($_,@_)}%{$f{$_[0]}}}for(sort{Y($a)- Y$b}keys%g){
			Ahh, back ahead of `/anick, at least for a little while.
137.48	Jasper McCrea	Tue Jul 2 16:47:32 2002	-ln0 @e{/\S+/g}=1;@a=keys%e;eval'$s=0;for$i(0..@a){for$j($i+1..@a){$s=@a[$i,$j]=@a[$j,$i]if/^\Q$a[$j] $a[$i]\E$/m}}'x9;$s?die:print for@a
			ref: my 140 passed 1.5 on my machine, and so does this. If this fails, can you let me know where?
137.54	Sec	Mon Jul 1 23:23:11 2002	-lan $t{$F[1]}.=" $F[0] "if$F[0]ne$F[1];@q{@F}=@F}for(keys%q){$t{$_}||$d||($t{$d=$_}+=print)for%q;$d||die;map{$t{$_}=~s/ $d //g}%t;$d="";
			Yet another way to write the if/for.
138.48	Bruce Gray	Tue Jul 2 20:28:24 2002	-lan @a{@F}=($l,$r)=@F;$h{$l}{$r}=$l ne$r;END{@_=keys%a;while($k=shift@_){$s{"@_"}++&&die;grep($h{$_}{$k},%h)?push@_,$k:($h{$k}=print$k)}}
139.46	Jon Coppeard	Mon Jul 1 20:11:06 2002	-nl12 ($a,$b)=split;$a eq$b or$p{$b}++,push@$a,$b;$p{$a}+=0}{@~=grep!$p{$_},keys%p;delete$p{$_},print,push@~,grep!--$p{$_},@$_ for@~;exit%p
139.51	Yanick Champoux	Wed Jul 3 01:48:20 2002	-lna push@a,[$F[1]],[@F[0..$F[0]ne$F[1]]]}while((@a,$y)=grep!grep($_ eq$y,@$_),@a){for$x(map@$_,@a){grep$_->[1]eq$x,@a or$y=$x}print$y||die
			Okay, enough for tonight...
139.54	Sec	Mon Jul 1 23:38:08 2002	-lan $t{$F[1]}.=" $F[0] "if$F[0]ne$F[1];@q{@F}=@F}for(keys%q){$t{$_}||$d||($t{$d=$_}+=print)for%q;$d||die;map{$t{$_}=~s/ \Q$d //g}%t;$d="";
			A new testcase steals two more characters from me.
140.48	Jasper McCrea	Tue Jul 2 12:32:40 2002	-ln0 @e{/\S+/g}=1;@a=keys%e;eval'$s=0;for$i(0..@a){for$j($i+1..@a){map{$s=@a[$i,$j]=@a[$j,$i]if"@a[$j,$i]"eq$_}/.*/g}}'x9;$s?die:print for@a
			I don't remember changing anything, but this is two shorter
140.51	Martin Carlsen	Mon Jul 1 15:54:42 2002	-pl sub i{($n{$_}=my$n=1+pop)>$.&&die;i($n)for@{$g{$_}}}/ /;$n=\$n{$`};push@{$g{$`}},$_=$';$`eq$'||i$$n}{$_=join$\,sort{$n{$a}-$n{$b}}keys%n
140.53	Jasper McCrea	Mon Jul 1 16:34:17 2002	-p0 @e{/\S+/g}=1;@a=keys%e;eval'$s=0;for$i(0..@a){for$j($i+1..@a){$s=@a[$i,$j]=($1,$2)if/^($a[$j]) ($a[$i])$/m}};'x9;$s&&die;$"=" ";$_="@a "
			86! Does not compute! Bzzzt! Does not compute!
141.51	Damien Neil	Wed Jul 3 01:01:23 2002	-lna for(@F){$h{$_}||=bless[$_]}next if($F[0]eq$F[1]);push@{$h{$F[0]}},$h{$F[1]};DESTROY{print$_[0][0];grep(!$_,@{$_[0]})&&exit 1;}END{%h=()}
141.54	Terje K	Thu Jul 4 11:57:05 2002	-anl ($a,$b)=@F;$h{$b}{$"}||=1;$h{$a}{$b}=1if$a ne$b}{sub z{my$x=pop;for(keys%h){if($h{$_}{$x}){$u{$_}++>1e5&&die;z($_);$u{$_}<2&&print}}}z$"
141.55	Dave Hoover	Wed Jul 3 17:00:36 2002	-pal @g{@F}=/^\Q$F[1] /||($f{$F[1]}{$F[0]}='')}sub Y{($y,@c)=@_;"@c"=~/ \Q$y /&&exit 1;@_,map{Y($_,@_)}%{$f{$y}}}for(sort{Y($a)- Y$b}keys%g){
142.48	Jasper McCrea	Mon Jul 1 17:12:05 2002	-ln0 @e{/\S+/g}=1;@a=keys%e;eval'$s=0;for$i(0..@a){for$j($i+1..@a){$s=@a[$i,$j]=@a[$j,$i]if grep"@a[$j,$i]"eq$_,/.*/g}};'x9;$s?die:print for@a
			hopefully this will get me out of the sand trap.
142.49	Chris Dolan	Tue Jul 2 16:57:17 2002	-pal sub a{($x=$n{my$z=pop})>99&&die;$z eq$_||$n{$_}>$x||a($n{$_}=$x+1,$_)for@{$a{$z}}}push@{$a{$F[1]}},@F;a@F}for(sort{$n{$b}-$n{$a}}keys%a){
			Reversion of my solution to pass test v1.6. I was wondering if this would happen...
142.51	Sec	Mon Jul 1 23:12:10 2002	-lan $t{$F[1]}.=" $F[0] "if$F[0]ne$F[1];@q{@F}=@F}for(keys%q){$t{$_}||($t{$d=$_}+=print)and last for%q;$d||die;map{$t{$_}=~s/ $d //g}%t;$d="";
			Reformatting the for/if gives a few more chars
143.50	Dave Hoover	Wed Jul 3 14:43:56 2002	-pal @g{@F}=($u,$v)=@F;$v eq$u||push@{$f{$v}},$u}sub Y{($y,@c)=@_;"@c"=~/ \Q$y /&&exit 1;@_,map{Y($_,@_)}@{$f{$y}}}for(sort{Y($a)- Y$b}keys%g){
			Ah, finally up with `/anick and /-\ndrew.
145.52	Sec	Mon Jul 1 23:06:30 2002	-lan $t{$F[1]}.=" $F[0] "if$F[0]ne$F[1];@q{@F}=@F}for(keys%q){for(%q){if(!$t{$_}){print;$t{$d=$_}=1;last}}$d||die;map{$t{$_}=~s/ $d //g}%t;$d="";
			New try: we store strings inside those hashes. (So much work for only three points gain)
146.47	Bruce Gray	Mon Jul 1 22:10:17 2002	-lan @a{@F}=($l,$r)=@F;$h{$l}{$r}=$l ne$r;END{@_=keys%a;while($k=shift@_){$s{"@_"}++&&die;push(@_,$k),next if grep$h{$_}{$k},%h;print$k;$h{$k}=1}}
146.50	Martin Carlsen	Mon Jul 1 13:26:40 2002	-pl sub i{my$n=pop;$n{$b=$_}=++$n;$n>$.?die:i($n)for@{$g{$b}}}/ /;$n=\$n{$`};push@{$g{$`}},$_=$';$`eq$'||i$$n}{$_=join$\,sort{$n{$a}-$n{$b}}keys%n
146.55	Terje K	Thu Jul 4 13:49:36 2002	-anl ($a,$b)=@F;$h{$b}{$"}||=1;$h{$a}{$b}=1if$a ne$b}{sub z{my$x=pop;map{if($h{$_}{$x}){$u{$_}++>9&&die;z($_);$y{$_}++||print;$u{$_}--}}keys%h}z$"
147.49	Chris Dolan	Tue Jul 2 20:14:48 2002	-pal sub a{$c{my$z=pop}++&&die;$z eq$_||$n{$_}>$n{$z}||a($n{$_}=$n{$z}+1,$_)for@{$a{$z}}}%c=push@{$a{$F[1]}},@F;a@F}for(sort{$n{$b}-$n{$a}}keys%a){
			Unlike my earlier (shorter) solutions, this one should work for any number of nodes up to the machine limit.
147.49	Dave Hoover	Wed Jul 3 14:06:45 2002	-pal @g{@F}=($u,$v)=@F;$v eq$u||push@{$f{$v}},$u}sub Y{my($y,@c)=@_;"@c"=~/ \Q$y /&&exit 1;(@c,map{Y($_,@_)}@{$f{$y}})}for(sort{Y($a)- Y$b}keys%g){
148.49	Sec	Mon Jul 1 21:11:02 2002	-lan $t{$F[1]}{$F[0]}=1if$F[0]ne$F[1];$t{$F[0]}{a}}for(keys%t){for(keys%t){!%{$t{$_}}&&print&&delete$t{$d=$_}&&last}map{delete$$_{$d||die}}%t;$d="";
			Some more strokes shaved with s/values//;
150.48	FatPhil	Mon Jul 1 18:31:57 2002	-n / (\S+)/;$p{$1}.=" $` "if$`ne$1;$p{$`}.=''}for(;;$f=''){map{$f||$p{$_}||print$f=$_,$/}keys%p;$f?delete$p{$f}:each%p?die:last;map s/ $f //g,values%p
			Erm, if the last one was rejected, then this one will be too. Waiting to find out why....
150.50	Dave Hoover	Wed Jul 3 13:53:01 2002	-pal @g{@F}=($u,$v)=@F;$v eq$u||push@{$f{$v}},$u}sub Y{my($y,@c)=@_;"@c"=~/ \Q$y /&&exit 1;(@c,map{Y($_,@_)}@{$f{$y}})}for(sort{Y($a)<=>Y($b)}keys%g){
			Better, but still looking for that trout...
151.48	FatPhil	Wed Jul 3 13:07:52 2002	-n / (\S+)/;$p{$1}.=" $` "if$`ne$1;$p{$`}x=1}for(;;$f=''){map{$f||$p{$_}||print$f=$_,$/}keys%p;$f?delete$p{$f}:each%p?die:last;map s/ \Q$f //g,values%p
			I think I saved a character somewhere. This is not _at_ its dead end.
152.48	FatPhil	Mon Jul 1 18:56:46 2002	-n / (\S+)/;$p{$1}.=" $` "if$`ne$1;$p{$`}.=''}for(;;$f=''){map{$f||$p{$_}||print$f=$_,$/}keys%p;$f?delete$p{$f}:each%p?die:last;map s/ \Q$f //g,values%p
			Phew! Never used \Q before. Quick recovery from my 150.45 rejection.
152.50	Martin Carlsen	Mon Jul 1 13:06:39 2002	-pl sub i{my$n=pop;$n{$b=$_}=++$n;$n>$.?die:i($n)for@{$g{$b}};}/ /;$n=\$n{$`};$`eq$'&&next;push@{$g{$`}},$_=$';i$$n}{$_=join$\,sort{$n{$a}-$n{$b}}keys%n
152.53	Yanick Champoux	Wed Jul 3 01:14:15 2002	-lna push@a,[$F[1]],[@F[0..$F[0]ne$F[1]]]}{my$y;for$x(@a){grep$x->[0]eq$_->[1],@a or$y=$x->[0]}print$y||die;redo if@a=grep{$_->[1]ne$y and$_->[0]ne$y}@a
			Totally different approach, same crummy score...
152.56	Damien Neil	Wed Jul 3 01:14:35 2002	-lna for(@F){$_=$h{$_}||=bless[$_]} push@{$F[0]},$F[1]if($F[0]!=$F[1]); DESTROY{ print$_[0][0]; grep(!$_,@{$_[0]})&&exit 1; } END{%h=()}
153.45	Bruce Gray	Mon Jul 1 21:43:43 2002	-lan @a{@F}=($l,$r)=@F;$h{$l}{$r}=1if$l ne$r;END{@_=keys%a;while($k=shift@_){$s{"@_"}++&&die;push(@_,$k),next if grep$_->{$k},values%h;print$k;$h{$k}=1}}
153.50	tinita	Tue Jul 2 09:23:06 2002	-lan push@t,($x,$y)=@F;$x ne$y&&(++$r{$x,$y}&&$r{$y,$x}&&die,$o{$x}{$y}=++$_{$y});sub t{$s{$_}++||print for@_;t(keys%{$o{$_}})for@_}END{t grep!$_{$_},@t}
154.47	Sec	Mon Jul 1 20:56:07 2002	-lan $t{$F[1]}{$F[0]}=1if$F[0]ne$F[1];$t{$F[0]}{a}}for(keys%t){for(keys%t){!%{$t{$_}}&&print&&delete$t{$d=$_}&&last}map{delete$$_{$d||die}}values%t;$d="";
			I had completely forgotten about 'values'
155.45	Bruce Gray	Mon Jul 1 21:36:10 2002	-lan ($l,$r)=@F;@a{@F}=1;$h{$l}{$r}=1if$l ne$r;END{@_=keys%a;while($k=shift@_){$s{"@_"}++&&die;push(@_,$k),next if grep$_->{$k},values%h;print$k;$h{$k}=1}}
155.50	FatPhil	Mon Jul 1 18:14:17 2002	-n / (\S+)/;$p{$1}.=" $` "if$`ne$1;$p{$`}.=''}for(;;$f=''){map{$f||$p{$_}||print$f=$_,$/}keys%p;$f?delete$p{$f}:each%p?die:last;map$p{$_}=~s/ $f //g,keys%p
			I've got it to the stage where I don't understand it. Is that a good or a bad thing?
156.47	Sec	Mon Jul 1 19:24:53 2002	-lan $F[0]ne$F[1]and$t{$F[1]}{$F[0]}=1;$t{$F[0]}{a}}for(keys%t){for(keys%t){!%{$t{$_}}&&print&&delete$t{$d=$_}&&last}map{delete$t{$_}{$d||die}}keys%t;$d="";
			loose some "if". &&,|| and "and" are shorter
156.50	Terje K	Mon Jul 1 18:55:24 2002	-lna $h{$F[0]}||=0;$h{$F[1]}||=$F[0]if$F[1]ne$F[0]}{@t=0;while(%h){@m=();for(keys%h){for$t(@t){if($h{$_}eq$t){print;push@m,$_;delete$h{$_}}}}@m||last;@t=@m}
158.51	Chris Dolan	Mon Jul 1 21:50:12 2002	-pal (grep/^$F[0]$/&&$F[1]ne$_,@{$a{$F[1]}})?die:push@{$a{$F[0]}},$F[1]}sub t{$_[0],map t($_),@{delete$a{$_[0]}}}unshift@t,grep!$c{$_}++,t(%a)while%a;for(@t){
			oops, submitted the wrong thing last time
158.52	Chris Dolan	Mon Jul 1 21:23:18 2002	-pal (grep/^$F[0]$/&&$F[1]ne$_,@{$a{$F[1]}})?die:push@{$a{$F[0]}},$F[1]}sub +{$_[0],map +($_),@{delete$a{$_[0]}}}unshift@t,grep!$c{$_}++,+(%a)while%a;for(@t){
158.64	Qingning Huo	Tue Jul 2 11:41:54 2002	#!perl -ln0 $0=$&,s/^(\Q$0\E) \1$/$1/m,/ \Q$0\E$/m||s/^\Q$0\E\s//gm&print$0while/^\S+/gm;$_&&die
158.64	Qingning Huo	Tue Jul 2 11:42:31 2002	#!perl -ln0 $0=$&,s/^(\Q$0\E) \1$/$1/m,/ \Q$0\E$/m||s/^\Q$0\E\s//gm&print$0while/^\S+/gm;$_&&die
159.44	Bruce Gray	Mon Jul 1 21:24:32 2002	-lan ($l,$r)=@F;@a{@F}=1;$h{$l}{$r}=1if$l ne$r;END{@_=keys%a;while($k=shift@_){$s{"@_"}++&&die;push(@_,$k),next if grep$_->{$k},values%h;print$k;delete$h{$k}}}
161.51	Chris Dolan	Mon Jul 1 16:27:30 2002	-pal push@{$a{$F[0]}},$F[1];die if grep{/^$F[0]$/&&!/^$F[1]$/}@{$a{$F[1]}}}sub t{$_[0],map{t($_)}@{delete$a{$_[0]}}}unshift@t,grep!$c{$_}++,t(%a)while%a;for(@t){
			whittling...
162.46	Qingning Huo	Mon Jul 1 10:28:14 2002	-ln0 $n=map$a{$_}++,split;while($n--){for$x(keys%a){($y=$x)=~s/\./[.]/g;s/^$y .+ //gm,delete$a{$x}if/$y/&&!/^(\S+) (?!\1$)$y$/m&&print$x}}print for keys%a;$_&&die
162.48	Bruce Gray	Mon Jul 1 19:20:47 2002	-ln /(\S+) +(.+)/;$n{$1}+=0;++$n{$2},push@{$s{$1}},$2if$1ne$2;END{@_=grep!$n{$_},keys%n;print&&push@_,grep!--$n{$_},@{$s{$_}}while$_=pop@_;$n{$_}&&die for keys%n}
163.50	FatPhil	Mon Jul 1 17:31:29 2002	for(<>){/ (\S+)/;$p{$1}.=" $` "if($`ne$1);$p{$`}.=''}for(;;$f=''){map{$f||$p{$_}||print$f=$_,$/}keys%p;$f?delete$p{$f}:each%p?die:last;map$p{$_}=~s/ $f //g,keys%p}
			First hack: 500 inc debug, 200 without Using everything I know aboutperl I've squeezed it to 163. YOU GUYS ARE NOT HUMAN! :-D
164.47	Sec	Mon Jul 1 18:40:48 2002	-lan $t{$F[1]}{$F[0]}=1if($F[0]ne$F[1]);$t{$F[0]}{a}}for(keys%t){for(keys%t){if(!%{$t{$_}}){print;delete$t{$d=$_};last}}die if!$d;map{delete$t{$_}{$d}}keys%t;$d="";
			A few more easy optimisations: for->map, omit 'scalar' or 'keys' here and there.
164.53	Martin Carlsen	Mon Jul 1 12:32:05 2002	-pl sub i{my($b,$n)=@_;($n{$b}=++$n)>$.&&die;i($_,$n)for@{$g{$b}}}/ /;$n=$n{$`}=$n{$`};$`eq$'&&next;push@{$g{$`}},$';i($',$n)}{$_=join$\,sort{$n{$a}<=>$n{$b}}keys%n
165.45	Qingning Huo	Mon Jul 1 09:53:25 2002	-ln0 $n=map$a{$_}++,split;s/(\S+) \1 //gm;while($n--){for$x(keys%a){($y=$x)=~s/\./[.]/g;s/^$y .+ //gm,delete$a{$x}if/$y/&&!/ $y$/m&&print$x}}print for keys%a;$_&&die
165.46	Dave Hoover	Wed Jul 3 12:02:45 2002	-pal @g{@F}=($u,$v)=@F;$v eq$u||push@{$f{$v}},$u}sub Y{my($y,@c)=@_;"@c"=~/ \Q$y /&&exit 1;push@c,$y;push@c,map{Y($_,@c)}@{$f{$y}};@c}for(sort{Y($a)<=>Y($b)}keys%g){
			Hack, Hack, Hack
165.51	Prakash Kailasa	Mon Jul 8 01:29:41 2002	-lan $i{$_}++||push@{$h{$F[1]}},$F[0];$h{$F[0]}||=[]}{sub d{my($w,$d,$x)=@_;map{$d>$.&&die;$x+=1+d($_,++$d)if$_ ne$w}@{$h{$w}};$x}map{print}sort{d($a)<=>d($b)}keys%h
			It passes the test program but takes more than 6 minutes my Duron 600MHz box. Will it get me out of the sandtrap (and barely onto the board)?
166.51	Terje K	Mon Jul 1 18:44:46 2002	-lna $h{$F[0]}||=0;$h{$F[1]}||=$F[0]if$F[1]ne$F[0]}{@t=0;while(%h){@m=();$o=1;for(keys%h){for$t(@t){if($h{$_}eq$t){print;push@m,$_;delete$h{$_};$o=0}}}$o&&last;@t=@m}
			first effort
168.49	Chris Dolan	Mon Jul 1 16:20:07 2002	-pal push@{$a{$F[0]}},$F[1];exit 1if grep{/^$F[0]$/&&!/^$F[1]$/}@{$a{$F[1]}}}sub t{$_[0],map{t($_)}@{delete$a{$_[0]}}}unshift@t,grep!$c{$_}++,t(keys%a)while+%a;for(@t){
			Oops, had a copy-n-paste error in the previous entry
169.48	Trevor Leffler	Fri Jul 5 18:49:48 2002	-lan ($a,$b)=@F;push@{$h{$a}},$a ne$b?$s{$b}=$b:();END{@v=grep{!$s{$_}}keys%h or exit 1;while(@v){for(@v){$p{$_}&&exit 1;print;@b{@{$h{$_}}}=$p{$_}++}@v=keys%b;%b=() }}
			Pretty much the "standard algorithm" that you find in the Net. At least it's sub-200...
170.50	Chris Dolan	Mon Jul 1 16:19:21 2002	-pal push@{$a{$F[0]}},$F[1];exit 1if grep{/^$F[0]$/&&!/^$F[1]$/}@{$a{$F[1]}}}sub t{$_ [0],map{t($_)}@{delete$a{$_[0]}}}unshift@t,grep!$c{$_}++,t(keys%a)while+%a;for(@ t){
179.54	tinita	Thu Jul 4 10:20:25 2002	-lan sub k{keys%{$_[0]}}m/ /;$`eq$'?$o{$`}||={}:$o{$`}{$'}++;END{sub _{$g{$_}--||die,$s{$_}++||print for@_;_(k($o{$_}))for@_}@g{map k($_),%o}=k\%o;(@_=grep!$g{$_}++,k\%o)||die;&_}
180.52	Dave Hoover	Wed Jul 3 10:51:25 2002	-pal @g{@F}=($u,$v)=@F;$v eq$u||push@{$f{$v}},$u}sub Y{my($y,$i,$c)=@_;$c=~/ \Q$y /&&exit 1;$c.=" $y";$i++;(map{$i+=Y($_,$i,$c)}@{$f{$y}})[-1]||$i|0}for(sort{Y($a)<=>Y($b)}keys%g){
183.52	Martin Carlsen	Mon Jul 1 12:10:24 2002	-nl sub i{my($b,$n)=@_;($n{$b}=++$n)>$.&&die;for(@{$g{$b}}){i($_,$n)}}/ /;if($`eq$'){$n{$`}=0;next}push@{$g{$`}},$';i($',$n{$`}=$n{$`});END{print join"\n",sort{$n{$a}<=>$n{$b}}keys%n}
			Look, no -pn!
186.43	Matthias Schoder	Wed Jul 3 14:31:29 2002	-lan057 @N=grep{!$N{$_}++}@F;%X=@F;END{while(@N){exit 1if%X&&$i++>@N;$n=shift@N;delete$X{$n}if$X{$n}eq$n;%Y=reverse%X;if(!$Y{$n}){print$n;delete$X{$n};$i=0;%Y=reverse%X}else{push@N,$n}}}
186.53	Dave Hoover	Wed Jul 3 10:32:43 2002	-pal @g{@F}=($u,$v)=@F;$v eq$u||push@{$f{$v}},$u}sub Y{my($y,$i,$c)=@_;$c=~/ \Q$y /&&exit 1;$c.=" $y";$i++;$f{$y}?(map{$i+=Y($_,$i,$c)}@{$f{$y}})[-1]:$i|0}for(sort{Y($a)<=>Y($b)}keys%g){
			Pass the trout, I need a breakthrough.
189.44	Sec	Mon Jul 1 18:30:01 2002	-lan $t{$F[1]}{$F[0]}=1if($F[0]ne$F[1]);$t{$F[0]}{a}}for(1..scalar keys%t){for(keys%t){if(!scalar keys%{$t{$_}}){print;delete$t{$_};$d=$_;last}}die if!$d;for(keys%t){delete$t{$_}{$d}}$d="";
			Try my file obfuscator for obvious gains in whitespace.
191.49	Dave Hoover	Wed Jul 3 01:28:06 2002	-pal @g{@F}=($u,$v)=@F;$u eq$v||push@{$f{$v}},$u}sub Y{my($y,$i,@c)=@_;exit 1 if grep{$_ eq$y}@c;push@c,$y;$i++;$f{$y}?(map{$i+=Y($_,$i,@c)}@{$f{$y}})[-1]:$i|0}for(sort{Y($a)<=>Y($b)}keys%g){
			Whew, that's one ugly mess! Let the golfing begin!
191.54	RRBuonanno	Wed Jul 3 13:20:37 2002	-nal ($x,$y)=@F;($a,$b)=@o{$x,$y};$b&&$b<$a&&die;@o{$x,$y}=$x eq$y?(0,0):($a=$a||9*$.,$b||$a+1);$b&&map{$o{$_}+=$o{$x}if$x ne$_&&$o{$_}>=$b}keys%o;END{$,=$\;print sort{$o{$a}<=>$o{$b}}keys%o}
			...at least it's an entry.
193.38	Keith Calvert Ivey	Mon Jul 1 12:55:49 2002	-lna0 @s{@F}=();$s{$2}{$1}++while/(.+) (?!\1 )(.+)/g;for(keys%s){for$h(values%s){$h==$_ or@{$h}{keys%{$s{$_}}}=@F for keys%$h}$s{$_}{$_}&&die}print for sort{keys%{$s{$a}}<=>keys%{$s{$b}}}keys%s
			First version that passed. I'm submitting it despite its messiness and despite being last on the board. Maybe the score will encourage other high submissions.
200.43	Matthias Schoder	Wed Jul 3 15:38:59 2002	-lan057 @N=grep{!$N{$_}++}@F;%X=@F;%Y=reverse@F;END{while(%X){exit 1if%X&&$i++>@N;$n=shift@N;delete$X{$n}if$X{$n}eq$n;if(!$Y{$n}){print$n;delete$X{$n};$i=0;%Y=reverse%X}else{push@N,$n}}map{print$_}@N}
205.50	Philippe 'BooK' Bruhat	Tue Jul 2 13:08:35 2002	-lna $a{$F[0]}{$F[1]}++;$F[0]ne$F[1]?$r{$F[1]}++:next;exit 1if$a{$F[1]}{$F[0]};sub a{my%b;do{$d{$_}++||print;@b{keys%{$a{$_}}}=();delete$b{$_}}for@_;$i++>99&&exit 1;a(keys%b)if%b}END{a grep{!$r{$_}}keys%a}
			This one is my first working attempt. Quite lame, particularly the $i++>99&&exit 1 part, that is used to detect cycles....
207.49	Chris Dolan	Mon Jul 1 15:57:39 2002	-nal $n{$F[1]}++;push@{$a{$F[0]}},$F[1];exit 1 if grep{/^$F[0]$/&&$F[0] ne$F[1]}@{$a{$F[1]}}}sub t{$_[0],map{t($_)}@{delete$a{$_[0]}}}{unshift@t,grep!$c{$_}++,t(sort{$n{$a}-$n{$b}}keys%a)while+%a;print for@t
			HARD problem!!!
214.49	Peter Haworth	Wed Jul 3 12:46:51 2002	-nla ++$s{$_}for@F;$F[0]eq$F[1]||push@r,[@F]}&p(keys%s);sub p{$,=$/,print(@o),exit 0if!@_;for(@_){my$x=shift;$c{$o[$o++]=$x}=1,&p,$c{$x}=0,$o--if!grep{($,,$;)=@$_;$,eq$x&&$c{$;}||$;eq$x&&!$c{$,}}@r;push@_,$x}}die;{
			This will probably land in the sand trap, too. This takes forever on the last test, but at least it uses constant memory this time
214.52	Matthias Schoder	Fri Jul 5 12:44:19 2002	-lan012 ($.,$?)=@F;$X{$.}{$?}=$Y{$?}{$.}=1if$.ne$?;map{$N{$_}++}@F;END{for(0..%N){for(keys%N){if(!$Y{$_}){print;for$?(keys%Y){delete$Y{$?}{$_};delete$Y{$?}if!%{$Y{$?}}}delete$X{$_};delete$N{$_};exit if!%N}}}exit 1}
216.49	Brad Jones	Tue Jul 2 00:07:05 2002	-nl ($a,$b)=split;($a ne$b)&&push@{$l{$b}},$a;$l{$a}||=[];END{while(%l){@j=sort{@{$\ l{$a}}<=>@{$l{$b}}}keys%l;$c=shift @j;(@{$l{$c}})&&exit(1);print$c;delete$l{$c}\ ;foreach$d(@j){$l{$d}=[grep{$_ ne$c}@{$l{$d}}];}}}
227.44	Philippe 'BooK' Bruhat	Wed Jul 3 14:09:34 2002	-lna @n{map{"\Q$_"}@F}=push@r,$_;END{$"=",";@t=glob "{@{[keys%n]}}$;"x keys%n;T:for(@t){$_=$;.$_;/ $;([!-~]+$;).*\1/x&&next;@s{@s}=0..(@s=split$; );for(@r){($a,$b)=split;next T if$s{$a}>$s{$b} }print join$/,@s[1..$#s];exit}die}
			This one is very greedy on memory... It asks glob to precompute all the possibilities (and even more!), and then uses the same algorithm as tpr04b.pl to print the first correct one. Since there are $n**$n possibilities ($n being the number of nodes in the graph), it eats up memory very fast!
243.50	Sec	Mon Jul 1 18:25:49 2002	-lan ^I$t{$F[1]}{$F[0]}=1 if($F[0] ne$F[1]); ^I$t{$F[0]}{a}; } for(1..scalar keys%t){ ^Ifor(keys %t){ ^I^Iif(!scalar keys%{$t{$_}}){ ^I^I^Iprint; ^I^I^Idelete $t{$_}; ^I^I^I$d=$_; ^I^I^Ilast; ^I^I}; ^I}; ^Idie if!$d; ^Ifor(keys %t){ ^I^Idelete $t{$_}{$d}; ^I}; ^I$d="";
			Try the file upload box so formatting is kept sane.
243.56	Philippe 'BooK' Bruhat	Tue Jul 2 17:10:09 2002	-lna $_="\Q$_"for@F;$b=0;for$a(@a){$b||$a=~s/ +($F[1]|$F[0]) +/ $_ /&&++$b;@c=$a=~/ (\S+) $/g;$_ eq$a&&next,s/^ @c /$a/&&($a='')for@a}push@a," $_ " if!$b;s/( [!-~]+)\1 /$1 /,/( [!-~]+ ).*\1/&&exit 1for@a;END{$_="@a";s/^ +| +$//g;y/ / /s;print}
			I guess I can shave some more, but to butcher this one, I'll have to change my algorithm. (Changed $c to @c)
247.56	Philippe 'BooK' Bruhat	Tue Jul 2 17:03:48 2002	-lna $_="\Q$_"for@F;$b=0;for$a(@a){$b||$a=~s/ +($F[1]|$F[0]) +/ $_ /&&++$b;($c)=($a=~/ (\S+) $/g);$_ eq$a&&next,s/^ $c /$a/&&($a='')for@a}push@a," $_ " if!$b;s/( [!-~]+)\1 /$1 /,/( [!-~]+ ).*\1/&&exit 1for@a;END{$_="@a";s/^ +| +$//g;y/ / /s;print}
			Shortening quotemeta, and using the aliasing capabilities of for.
252.55	Philippe 'BooK' Bruhat	Tue Jul 2 16:59:25 2002	-lna @F=map{quotemeta}@F;$b=0;for$a(@a){$b||$a=~s/ +($F[1]|$F[0]) +/ $_ /&&++$b;($c)=($a=~/ (\S+) $/g);$_ eq$a&&next,s/^ $c /$a/&&($a='')for@a}push@a," $_ " if!$b;s/( [!-~]+)\1 /$1 /,/( [!-~]+ ).*\1/&&exit 1for@a;END{$_="@a";s/^ +| +$//g;y/ / /s;print}
			I thought I found a better approach than the previous one... I hope at least that this one won't be rejected. :-)
278.46	Peter Haworth	Wed Jul 3 10:31:40 2002	-an $set{$_}++||push@set,$_ for@F; $F[0]eq$F[1]or$rerules.=qr/(?>(?=(?>.*?^\Q$F[0]\E$).*^\Q$F[1]\E$))/ms; } sub perm{ my@a; for(@_){ my$x=shift; push@a,"$x\n$_"for&perm; push@_,$x; } @a?@a:''; } for(&perm(@set)){ /$rerules/and print(),exit 0 } die; {
			This seems to work, but is probably the least efficient program I have ever written, so I expect the sand trap for my efforts
316.48	Ryan Zachry	Wed Jul 3 04:03:25 2002	-l while(<>){($l,$r)=split;exit(2)if$x{$r}{$l};if($l ne$r){$x{$l}{$r}=1;$s{$r}++}push@{$p{$l}},$r}@n=@o=grep{!$s{$_}}@q=keys%p;map{print if!$o[0]}@q;while(@o){$z=pop@o;print$z;map{exit(3)if$t{$_}>10;push@o,$_ if!--$s{$_}}@{$p{$z}}}while($_=pop@n){$t{$_}++;exit(4)if$t{$_}>300;push@n,@{$p{$_}};exit(0)if$x eq$_;$x=$_}
322.47	Ryan Zachry	Wed Jul 3 11:44:21 2002	-l while(<>){($l,$r)=split;exit(2)if$x{$r}{$l};if($l ne$r){$x{$l}{$r}=1;$s{$r}++}push@{$p{$l}},$r}@n=@o=grep{!$s{$_}}@q=keys%p;map{print if!exists$o[0]}@q;while(@o){$z=pop@o;print$z;map{exit(3)if$t{$_}>10;push@o,$_ if!--$s{$_}}@{$p{$z}}}while($_=pop@n){$t{$_}++;exit(4)if$t{$_}>300;push@n,@{$p{$_}};exit(0)if$x eq$_;$x=$_}
410.54	Sec	Mon Jul 1 18:11:14 2002	-lan $t{$F[1]}{$F[0]}=1 if($F[0] ne$F[1]);; $t{$F[0]}{""}; } { for(1..scalar keys%t){ for$a(keys %t){ # print "Looking at $a: ",scalar keys(%{$t{$a}}); if(!scalar keys%{$t{$a}}){ print $a; delete $t{$a}; $d=$a; last; }; }; die if!$d; for$a(keys %t){ # print "Del $a,$d"; delete $t{$a}{$d}; }; $d=""; };
			My first working solution *uff*
517.48	Kristen Thelen	Thu Jul 4 05:02:22 2002	my @l, @o;sub visit{if (@{$l{$_[0]}}[0] != 0){return;}@{$l{$_[0]}}[0]=1;$count=0;foreach $v (@{$l{$_[0]}}){if ($count ne 0){if ($v eq $_[1]){exit 1;}visit($v, $_[1]);}$count++;}@{$l{$_[0]}}[0]=2;unshift(@o, $_[0]);return;}while (<>) {($f, $s) = split;if (!@{$l{$f}}){push @{$l{$f}}, 0; }if (!@{$l{$s}}){push @{$l{$s}}, 0; }if ($f ne $s){$found = 0;foreach $item (@{$l{$f}}){if ($item eq $s){$found = 1;}}if (!$found){push @{$l{$f}}, $s; }}}foreach $key (keys %l){visit($key, $key);}foreach $i (@o){print "$i ";}
			I'm a beginner in both perl and perl golf, but enjoyed this exercise anyway. Thanks!
555.49	Kristen Thelen	Thu Jul 4 05:30:52 2002	my @l, @o;sub visit{if (@{$l{$_[0]}}[0] != 0){return;}@{$l{$_[0]}}[0]=1;$count=0;foreach $v (@{$l{$_[0]}}){if ($count ne 0){if ($v eq $_[1]){exit 1;}visit($v, $_[1]);}$count++;}@{$l{$_[0]}}[0]=2;unshift(@o, $_[0]);return;}while (<>) {($f, $s) = split;if (!@{$l{$f}}){push @{$l{$f}}, 0; }if (!@{$l{$s}}){push @{$l{$s}}, 0; }if ($f ne $s){$found = 0;$count =0;foreach $item (@{$l{$f}}){if (($count ne 0) && ($item eq $s)){$found = 1;}$count++;}if (!$found){push @{$l{$f}}, $s; }}}foreach $key (keys %l){visit($key, $key);}foreach $i (@o){print "$i ";}
			Here we go again
588.59	Dave Hoover	Mon Jul 8 04:57:18 2002	-pal @g{@F}=/^\Q$F[1] /||($f{$F[0]}{$F[1]}='')} $_ = " $F[0] "; while (%g) { $i =0; for $k(keys %g) { if (s/ \Q$k\E /join'',map {delete $g{$_};" $_ "}( $k, keys %{$f{$k}})/e) { $i++;last; } } if(!$i) { for $k(keys %g) { $l = $k; for $j(keys %{$f{$k}}) { if (s/ \Q$j\E / $k $j /) { delete $g{$k}; $i++;last; } } } } if(!$i) { $_.=" $l "; delete $g{$l} } } for $x(split){ for $k(keys %{$f{$x}}) { / \Q$x\E .*\Q$k\E .*/ or die; } } for(split){
			So sad. Once again, foiled by the 200 line test. Ran out of time...