package Math::BaseCalc;

use strict;
use Carp;
use vars qw($VERSION);
$VERSION = '1.014';

sub new {
  my ($pack, %opts) = @_;
  my $self = bless {}, $pack;
  $self->{has_dash} = 0; 
  $self->digits($opts{digits});
  return $self;
}

sub digits {
  my $self = shift;
  if (@_) {
    # Set the value


    if (ref $_[0]) {
      $self->{digits} = [ @{ shift() } ];
    } else {
      my $name = shift;
      my %digitsets = $self->_digitsets;
      croak "Unrecognized digit set '$name'" unless exists $digitsets{$name};
      $self->{digits} = $digitsets{$name};
    }
    $self->{has_dash} = grep { $_ eq '-' } @{$self->{digits}};

    $self->{trans} = {};
    # Build the translation table back to numbers
    @{$self->{trans}}{@{$self->{digits}}} = 0..$#{$self->{digits}};

  }
  return @{$self->{digits}};
}


sub _digitsets {
  return (
      'bin' => [0,1],
      'hex' => [0..9,'a'..'f'],
      'HEX' => [0..9,'A'..'F'],
      'oct' => [0..7],
      '64'  => ['A'..'Z','a'..'z',0..9,'+','/'],
      '62'  => [0..9,'a'..'z','A'..'Z'],
     );
}

sub from_base {
  my $self = shift;
  return -1*$self->from_base(substr($_[0],1)) if !$self->{has_dash} && $_[0] =~ /^-/; # Handle negative numbers
  my $str = shift;
  my $dignum = @{$self->{digits}};

  # Deal with stuff after the decimal point
  my $add_in = 0;
  if ($str =~ s/\.(.+)//) {
    $add_in = $self->from_base(reverse $1)/$dignum**length($1);
  }

  $str = reverse $str;
  my $result = 0;
  my $trans = $self->{trans};
  while (length $str) {
    ## no critic
    return undef unless exists $trans->{substr($str,0,1)};
    # For large numbers, force result to be an integer (not a float)
    $result = int($result*$dignum + $trans->{chop $str});
  }

  # The bizarre-looking next line is necessary for proper handling of very large numbers
  return $add_in ? $result + $add_in : $result;
}

sub to_base {
  my ($self,$num) = @_;
  return '-'.$self->to_base(-1*$num) if $num<0; # Handle negative numbers

  my $dignum = @{$self->{digits}};

  my $result = '';
  while ($num>0) {
    substr($result,0,0) = $self->{digits}[ $num % $dignum ];
    use integer;
    $num /= $dignum;
    #$num = (($num - ($num % $dignum))/$dignum);  # An alternative to the above
  }
  return length $result ? $result : $self->{digits}[0];
}


1;
__END__


=head1 NAME

Math::BaseCalc - Convert numbers between various bases

=head1 VERSION

version 1.017

=head1 SYNOPSIS

  use Math::BaseCalc;

  my $calc = new Math::BaseCalc(digits => [0,1]); #Binary
  my $bin_string = $calc->to_base(465); # Convert 465 to binary

  $calc->digits('oct'); # Octal
  my $number = $calc->from_base('1574'); # Convert octal 1574 to decimal

=head1 DESCRIPTION

This module facilitates the conversion of numbers between various
number bases.  You may define your own digit sets, or use any of
several predefined digit sets.

The to_base() and from_base() methods convert between Perl numbers and
strings which represent these numbers in other bases.  For instance,
if you're using the binary digit set [0,1], $calc->to_base(5) will
return the string "101".  $calc->from_base("101") will return the
number 5.

To convert between, say, base 7 and base 36, use the 2-step process
of first converting to a Perl number, then to the desired base for the
result:

 $calc7  = new Math::BaseCalc(digits=>[0..6]);
 $calc36 = new Math::BaseCalc(digits=>[0..9,'a'..'z']);

 $in_base_36 = $calc36->to_base( $calc7->from_base('3506') );

If you just need to handle regular octal & hexdecimal strings, you
probably don't need this module.  See the sprintf(), oct(), and hex()
Perl functions.

=head1 METHODS

=over 4

=item * new Math::BaseCalc

=item * new Math::BaseCalc(digits=>...)

Create a new base calculator.  You may specify the digit set to use,
by either giving the digits in a list reference (in increasing order,
with the 'zero' character first in the list) or by specifying the name
of one of the predefined digit sets (see the digit() method below).

If your digit set includes the character C<->, then a dash at the
beginning of a number will no longer signify a negative number.

=item * $calc->to_base(NUMBER)

Converts a number to a string representing that number in the
associated base.

If C<NUMBER> is a C<Math::BigInt> object, C<to_base()> will still work
fine and give you an exact result string.

=item * $calc->from_base(STRING)

Converts a string representing a number in the associated base to a
Perl integer.  The behavior when fed strings with characters not in
$calc's digit set is currently undefined.

If C<STRING> converts to a number too large for perl's integer
representation, beware that the result may be auto-converted to a
floating-point representation and thus only be an approximation.

=item * $calc->digits

=item * $calc->digits(...)

Get/set the current digit set of the calculator.  With no arguments,
simply returns a list of the characters that make up the current digit
set.  To change the current digit set, pass a list reference
containing the new digits, or the name of a predefined digit set.
Currently the predefined digit sets are:

       bin => [0,1],
       hex => [0..9,'a'..'f'],
       HEX => [0..9,'A'..'F'],
       oct => [0..7],
       64  => ['A'..'Z','a'..'z',0..9,'+','/'],
       62  => [0..9,'a'..'z','A'..'Z'],

 Examples:
  $calc->digits('bin');
  $calc->digits([0..7]);
  $calc->digits([qw(w a l d o)]);

If any of your "digits" has more than one character, the behavior is
currently undefined.

=back

=head1 QUESTIONS

The '64' digit set is meant to be useful for Base64 encoding.  I took
it from the MIME::Base64.pm module.  Does it look right?  It's sure in
a strange order.

=head1 AUTHOR

Ken Williams, ken@forum.swarthmore.edu

=head1 COPYRIGHT

This is free software in the colloquial nice-guy sense of the word.
Copyright (c) 1999, Ken Williams.  You may redistribute and/or modify
it under the same terms as Perl itself.

=head1 SEE ALSO

perl(1).

=cut