Also, what he didn't mention is that it stemmed from an argument between PHP and Perl (don't bother preaching, i know the advantages and disadvantages of each, and use both ;)).

Anyway, this was the PHP solution I came up with.

#!/usr/bin/php4 -q
<?
for($i=1;$i<10;$i++){for($n=1;$n<10;$n++){echo max(abs($i-(10-$argv[2]
+)),abs($n-$argv[1]));}echo"\n";}
?>
[download]

Not counting the #! line or the <?'s, I was the winner by about 4 characters I think (keeping in mind there was a time limit on the competition).

Since we basically had the same idea for the algorithm I guess all it really came down to was the length of php/perl's syntax. I have also since seen much much smaller perl answers but I think I would be very hard pressed to make the php solution any smaller.

- James

#!/usr/bin/perl
@q=@ARGV;for($y=9;$y--;){print abs$q[1]-$y>abs$q[0]-$_?abs$q[1]-$y:abs
+($q[0]-$_)for(1..9);print$/}
[download]

.02

cLive ;-)

$x=shift;$y=10-shift;for$a(1..9){for(1..9){$c=abs$y-$a;$d=abs$x-$_;pri
+nt$c>$d?$c:$d}print"\n"}
[download]

-Mark

And getting rid of the for loops appears to take another 10 strokes off.

$x=shift;$z=10-shift;map{$t=abs$z-$_;print map($t>abs$x-$_?$t:abs$x-$_
+,1..9),$/;}0..9
[download]

You can get 4 more with shift vs pop. + 1 for a useless ;.

$z=10-pop;$x=pop;map{$t=abs$z-$_;print map($t>abs$x-$_?$t:abs$x-$_,1..
+9),$/}1..9
[download]

(I also fixed a bug.)

($a,$b)=@ARGV;for($y=9;$y>0;$y--){$c=abs$b-$y;print$/,map{$d=abs$a-$_;$c>$d?$c:$d}(1..9)}

without cheating it's a few characters longer.

Let's knock another 4 characters off :)

($a,$b)=@ARGV;for($y=9;$y--;){$c=abs$b-$y;print$/,map{$d=abs$a-$_;$c>$
+d?$c:$d}(1..9)}
[download]

.02

cLive ;-)

Ahh i really like that one, i was thinking of doing it a similar way but i couldn't get it producing the right output, i also thought of using sort($a-$_,$b-$y)[0]; to get the greatest but couldn't get that working.

Neil Archibald
- /dev/IT -

$i=9;do{print((map{(sort map{abs}$_-$ARGV[0],$i-$ARGV[1])[-1]}1..9),"\
+n")}while--$i
[download]

Russ

Okay, now 77 (if you don't miss the trailing newline):

$i=9;do{print"\n",map{(sort abs$_-$ARGV[0],abs$i-$ARGV[1])[-1]}1..9}wh
+ile--$i
[download]

Russ

74 :)

$i=9;do{print$/,map{(sort abs$_-$ARGV[0],abs$i-$ARGV[1])[1]}1..9}while
+--$i
[download]

$!/usr/bin/perl
@A=map{abs}@ARGV;for($y=9;$y>0;$y--){print$A[1]-$y>$A[0]-$_?$A[1]-$y:(
+$A[0]-$_)for(1..9);print$/}
[download]

Basically all I did was eliminate the redundant use of 'abs', and shortened the name of ARGV. Oh, and used $/ instead of "\n" to save two keystrokes. It's about 15 keystrokes shorter, I believe.

UPDATE:

Many apologies for bungling my untested attempt. As you pointed out, the abs wasn't exactly redundant. I just read precidence wrong. Actually, it was redundant, but its parameters weren't. In the spirit of "do it once and once only" I broke the common functionality out into a subroutine. And as before, I used $/ instead of "\n". The following IS tested, works, and saves you ten keystrokes (for 98 total, plus the shebang line).

sub a{abs$ARGV[shift]-pop}for($y=9;$y;$y--){print a(1,$y)>a(0,$_)?a 1,
+$y:a(0,$_)for 1..9;print$/}
[download]

Dave

"If I had my life to do over again, I'd be a plumber." -- Albert Einstein

I just tested that, and it doesn't seem to work. The abs calls wern't really redundant because you needed to test the greater value displaced from 0, so when you have two negative values -5 < -4 but you still want -5 returned. :)

UPDATE: Well done with the update, it works nicely when i tested it, i liked the use of the sub too, i only wish the game of golf had a longer time limit so i could have played around with some of these ideas for longer,

Neil Archibald
- /dev/IT -

$_=('x'x9 .$/)x9;
substr($_,10*(9-$ARGV[1])+$ARGV[0]-1,1)=0;
for$a(0..8){for$b(0,8,9,10){s/(?<=$a.{$b})x|x(?=.{$b}$a)/$a+1/esg}}
print;
[download]

--
[ e d @ h a l l e y . c c ]