# k2uhr.pl
# 
# You need 2 data files: 
#   dec-unicode-wansung.txt
#   dec-unicode-camo.txt
# 
# and you must modify 2 lines bellow to your right full-path:
#
# open(TABLEW, "dec-unicode-wansung.txt") 
# open(TABLEC, "dec-unicode-camo.txt") 
#
# USAGE: 
# perl k2uhr <file name>
# and you can get output file "file.out".
# 
# Input file has euc-kr(ks code) characters in it,
# and output file "file.out" has 
# decimal Unicode hangul (charcter entity reference??),
# euc-kr, and HR. Each line is tripled.


# Englich (need to wrapped in "[" and "]")
$alphabet = "-*abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ";
# camo data
@hr_camo = ("k", "kk", "ks", "n", "nj", "nh", "t", "tt",
            "r", "rk", "rm", "rp", "rs", "rth", "rph", "rh",
            "m", "p", "pp", "ps", "s", "ss", "0", "j",
            "jj", "ch", "kh", "th", "ph", "h", "a", "ae",
            "ya", "yae", "eo", "e", "yeo", "ye", "o", "wa",
            "wae", "oe", "yo", "u", "wo", "we", "wi", "yu",
            "eu", "eui", "i");
# cheos-so-ri data
@hr_first = ("k", "kk", "n", "t", "tt", "r", "m", "p",
             "pp", "s", "ss", "", "j", "jj", "ch", "kh",
             "th", "ph", "h");
# ka-un-tes-so-ri data
@hr_second = ("a", "ae", "ya", "yae", "eo", "e", "yeo", "ye",
              "o", "wa", "wae", "oe", "yo", "u", "wo", "we",
              "wi", "yu", "eu", "eui", "i");
# kkeus-so-ri data
@hr_third = ("", "k", "kk", "ks", "n",  "nj",  "nh", "t",
             "l", "lk", "lm", "lp", "ls", "lth", "lph", "lh",
             "m", "p", "ps", "s", "ss", "ng", "j", "ch",
             "kh", "th", "ph", "h");
$kugiri_moji = "-";
$tandoku_moji = "*";
$hr_hangul_off = "[";
$hr_hangul_on = "]";
$flag_hangul = 1;               # not English mode
$flag_hr = 0;                   # the last char in hr_buffer is HR

$char_nature = "";              # 0:English 1:symbols 2:Hangul
$u_str = "";                    # to Unicode number
$hr_str = "";                   # to HR
$hr_buffer = "";                # to HR (buffer)

# to make data: from ks hangul char to Unicode number
%dec_u = ();
open(TABLEW, "dec-unicode-wansung.txt") # PLEASE MODIFY
    || die "Can't open dec-unicode-wansung.txt: $!\n";
while(<TABLEW>){
    ($a, $b) = split(' ', $_);
    $hex_code = &get_hex_code( $b );
    $dec_u{$hex_code} = $a;
}
close TABLEW;

open(TABLEC, "dec-unicode-camo.txt") # PLEASE MODIFY
    || die "Can't open dec-unicode-camo.txt: $!\n";
while(<TABLEC>){
    ($a, $b) = split(' ', $_);
    $hex_code = &get_hex_code( $b );
    $dec_u{$hex_code} = $a;
}
close TABLEC;

# MAIN PROGRAM
open(OUTFILE, '>file.out');
while(<>){                      # read each line
    $u_str = "";
    $hr_str = "";
    $hr_buffer = "";
    $flag_hangul = 1;
    $flag_hr = 0;
    chomp;                      # chop the last "\n"
    $ks_str = $_;               # $ks_str is no-conversion-str (so ks code)
    $in_str = $ks_str;
    ($char_nature, $head_char, $in_str) = &get_first_char($in_str);
    while($char_nature ne "ERROR") {    
        if($char_nature == 0) { # case: English char
            $flag_hr = 0;
            $u_str = $u_str . $head_char;
            if($flag_hangul){
                $flag_hangul = 0;
                $hr_str = $hr_str . $hr_buffer;
                $hr_buffer = "";
            }
            $hr_buffer = $hr_buffer . $head_char;
        } elsif($char_nature == 1) { # case: Symbol
            $flag_hr = 0;
            $u_str = $u_str . $head_char;
            $hr_buffer = $hr_buffer . $head_char;
        } else {                # case: Hangul
            $u_code = &to_u($head_char);
            if($u_code eq "ERROR") { # when fail to convert to Unicode
                $u_str = $u_str . $head_char;
                $hr_buffer = $hr_buffer . $head_char;
            } else {            # when succeed to  convert to Unicode
                $u_str = $u_str . '&#' . $u_code . ';';
                unless($flag_hangul) {
                    $flag_hangul = 1;
                    $hr_str = $hr_str . $hr_hangul_off . $hr_buffer . $hr_hangul_on;
                    $hr_buffer = "";
                }
                if($flag_hr) {
                    $hr_buffer = $hr_buffer . $kugiri_moji;
                }
                $hr_buffer = $hr_buffer . &to_hr($u_code);
                $flag_hr = 1;
            }
        }
        ($char_nature, $head_char, $in_str) = &get_first_char($in_str);
    }
    # flash the buffer
    if($flag_hangul) {
        $hr_str = $hr_str . $hr_buffer;
    } else {
        $hr_str = $hr_str . $hr_hangul_off . $hr_buffer . $hr_hangul_on;
    }
    # print out
    print OUTFILE $u_str;
    print OUTFILE "\n";
    print OUTFILE $ks_str;
    print OUTFILE "\n";
    print OUTFILE $hr_str;
    print OUTFILE "\n";
    print OUTFILE "\n";
}
close(OUTFILE);

sub get_hex_code {              # from 2byte char to its code(hex)
    local(@code);
    local(@code) = unpack("C2", $_[0]);
    sprintf("%X%X",@code);
}

sub get_first_char {            # from str to its head char etc.
    local($source_str) = $_[0];
    local($l) = length($source_str);
    if($l > 0){                 # not null str
        local($head_char) = substr($source_str, 0, 1);
        local(@char_num) = unpack("c", $head_char);
        local($minusp) = substr($char_num[0], 0, 1);
        if($minusp ne '-'){     # 7bit char
            if(index($alphabet, $head_char) != -1){ # case: English char
                (0, $head_char, substr($source_str, 1));
            } else {            # case: Symbol
                (1, $head_char, substr($source_str, 1));
            }
        } else {                # 8bit char
            if($l > 1){         # 2byte char (Hangul)
                $head_char = substr($source_str, 0, 2);
                (2, $head_char, substr($source_str, 2));
            } else {            # 1byte char (not Hangul)
                ("ERROR");
            }
        }
    } else {                    # null str
        ("ERROR");
    }
}
        
sub to3bul {                    # from wansung Unicode to 3bul code
    local($dec) = $_[0];
    $dec = $dec - 44032;
    local($tmp) = 0;
    local($third_code) = $dec % 28;
    $tmp = ($dec - $third_code) / 28;
    local($second_code) = $tmp % 21;
    local($first_code) = ($tmp - $second_code) / 21;
    ($first_code, $second_code, $third_code);
}

sub to_camo {                   # from camo Unicode to camo code
    $_[0] - 12593;
}

sub to_u {                      # from ks char to its (decimal) Unicode
    local($hex) = &get_hex_code($_[0]);
    if($dec_u{$hex} ne "") {
        $dec_u{$hex};
    } else {
        "ERROR";
    }
}
    
sub to_hr {                     # from (decimal) Unicode  Hangul to HR
    local($code) = $_[0];
    if($code >= 12593 && $code <= 12643) {
        $tandoku_moji . $hr_camo[&to_camo($code)];
    } elsif ($code >= 44032 && $code <= 55203) {
        local(@wansung_code) = to3bul($code);
        $hr_first[$wansung_code[0]] . $hr_second[$wansung_code[1]] . $hr_third[$wansung_code[2]];
    } else {
        "ERROR";
    }
}