#!/usr/bin/perl -w
use strict;

=head1 NAME

uni2utf8

=head1 SYNOPSIS

uni2utf8 [options] number [number...]

=head1 DESCRIPTION

Converts unicode into utf-8.

Supports these options:

=over 4

=item -z

Put a single space every half-byte and a triple space every byte.

=back

=cut

my %options;
$options{z} = 1;

for my $arg (@ARGV) {

	if (length($arg) > 7) {
		print_bin($arg);

	} elsif ($arg =~ m/[abcdefABCDEF]/) {
		print_hex("0x" . $arg);

	} elsif (substr($arg, 0, 2) eq '0x') {
		print_hex($arg);

	} elsif (substr($arg, 0, 1) eq '0') {
		print_oct($arg);

	} else {
		print_dec($arg);
	}
}

exit 0;


########################################################################
############################# SUBROUTINES ##############################
########################################################################

sub print_bin {
	my $result = binary_uni2utf8(@_);

	if ($options{z}) {
		print substr($result, 0, 4), " ", substr($result, 4, 4);
		if (length($result) > 8) {
			print "   ", substr($result, 8, 4), " ", substr($result, 12, 4);
			if (length($result) > 16) {
				print "   ", substr($result, 16, 4), " ", substr($result, 20, 4);
			}
		}
		print "\n";
	} else {
		print $result, "\n";
	}
}

sub print_hex {
	my $result = binary_uni2utf8(sprintf("%016b", oct $_[0]));
	if (length($result) == 8) {
		print sprintf("%02x", oct("0b$result")), "\n";
	} elsif (length($result) == 16) {
		print sprintf("%04x", oct("0b$result")), "\n";
	} else {
		print sprintf("%06x", oct("0b$result")), "\n";
	}
	
}

sub print_oct {
	my $result = binary_uni2utf8(sprintf("%016b", oct $_[0]));
	if (length($result) == 8) {
		print sprintf("%02o", oct("0b$result")), "\n";
	} elsif (length($result) == 16) {
		print sprintf("%04o", oct("0b$result")), "\n";
	} else {
		print sprintf("%06o", oct("0b$result")), "\n";
	}
}

sub print_dec {
	my $result = binary_uni2utf8(sprintf("%016b", oct $_[0]));
	if (length($result) == 8) {
		print sprintf("%02d", oct("0b$result")), "\n";
	} elsif (length($result) == 16) {
		print sprintf("%04d", oct("0b$result")), "\n";
	} else {
		print sprintf("%06d", oct("0b$result")), "\n";
	}
}

=item binary_uni2utf8

Takes a 16-char string of 0s and 1s in Unicode.
Returns a string of 0s to 1s in UTF-8, either 8, 16, or 24 chars.
This function mangles 32-bit unicode values.

=cut

sub binary_uni2utf8 {
	my $unicode = shift;

	if (substr($unicode, 0, 8) eq '00000000') {
		return "0" . substr($unicode, 9, 7);

	} elsif (substr($unicode, 0, 5) eq '00000') {
		return "110" . substr($unicode, 5, 5) . "10" . substr($unicode, 10, 6);

	} else {
		return "1110" . substr($unicode, 0, 4) . "10" . substr($unicode, 4, 6) . "10" . substr($unicode, 10, 6);
	}
}