(tye)Re: Base36 numbers: speed and golf

[tye]

in reply to Base36 numbers: speed and golf

Of course, I can't say whether this is the fastest but it was designed to be fast.

#!/usr/bin/perl -w
use strict;
my %next;
BEGIN {
    @next{'0'..'9','A'..'Z'}= ('1'..'9','A'..'Z','0');
}
sub inc36 {
    my $add= "";
    my $dig;
    while(  not $dig= $next{chop($_[0])}  ) {
        $add .= $dig;
        if(  ! length($_[0])  ) {
            $dig= '1';
            last;
        }
    }
    $_[0] .= $dig . $add;
}
while(  <DATA>  ) {
    for(  split " "  ) {
        print "$_ + 1 = ";
        inc36( $_ );
        print $_,$/;
    }
}
__END__
0 1 9 A Y Z 10 19 1Z 9Z AZ ZZ YZZ XYZZY XYZZZ

which outputs:

0 + 1 = 1
1 + 1 = 2
9 + 1 = A
A + 1 = B
Y + 1 = Z
Z + 1 = 10
10 + 1 = 11
19 + 1 = 1A
1Z + 1 = 20
9Z + 1 = A0
AZ + 1 = B0
ZZ + 1 = 100
YZZ + 1 = Z00
XYZZY + 1 = XYZZZ
XYZZZ + 1 = XZ000

        - tye (but my friends call me "Tye")

[jmcnamara]

Well done tye. This is an elegant solution and the fastest by a mile.

There is a lot of good code produced here at PerlMonks but I wonder how much of it is actually studied or even looked at. At the time of writing two other solutions that contain bugs have higher reps than this node.

--
John.

[demerphq]

Well, this was interesting. I wrote a version of your code that used an array and ord instead of the hash. It was 10% slower in my tests which quite suprised me. I really thought the difference in lookup time between an array and a hash would outway the cost of the ord function, but not so.

Thanks, very educational.

Yves / DeMerphq
--
When to use Prototypes?
Advanced Sorting - GRT - Guttman Rosler Transform