#!/usr/bin/perl  -w
#
# Copyright (C) 1998, 2013 Jungshik Shin, Paul Hardy
#
# johab2ucs2.pl 
# This script(working as filter) converts  Hangul "Johab encoded  fonts"
# with an unofficial XLFD name "-johab" in BDF format
# to UCS-2 encoded font in a format defined by
# Roman Czyborra <roman@czyborra.com> at 
# http://czyborra.com/unifont/
#
# LICENSE:
#
#    This program is free software: you can redistribute it and/or modify
#    it under the terms of the GNU General Public License as published by
#    the Free Software Foundation, either version 2 of the License, or
#    (at your option) any later version.
#  
#    This program is distributed in the hope that it will be useful,
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the 
#    GNU General Public License for more details.
#  
#    You should have received a copy of the GNU General Public License
#    along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
#
# 'hanterm304font.tar.gz contains about a dozen of 
# "Johab-encoded" fonts. The package is available 
# ftp://ftp.kaist.ac.kr/hangul/terminal/hanterm/hanterm304beta/fonts
# Please, note that this script only works with  fonts whose 
# XLFD name ends with 
#
#  --16-160-75-75-c-160-johab-1
# (and whose file name in the package doesn't include 's' or 'sh' preceding
# '(m|g)16.bdf'. )
#
# There are  four of them : 
#  johabg16.bdf,johabm16.bdf,johabp16.bdf,iyagi16.bdf.  
#
# Fonts in the package with other XLFD names
# (johabs and johabsh) contain glyphs for about 5000 Hanjas and special symbols
# defined in KS C 5601-1987. 

#  Sep. 29, 1998
#  Jungshik Shin <jshin@pantheon.yale.edu>

# A more complete routine which not only covers
# *modern* pre-composed Hangul syllables in UAC00-UD7A3
# but also supports dynamic rendering of
# Hangul syllables(medieval as well as modern)
# using Hangul comibining Jamos  at [U1100-U11FF]
# was made by Deog-tae Kim <dtkim@calab.kaist.ac.kr
# to be used in Java font-properties file. 
# It's available at  http://calab.kaist.ac.kr/~dtkim/java/

# 2 May 2008: changes by Paul Hardy (unifoundry <at> unfoundry.com):
#
#    - In tconBase, "459" index corrected to "449".
#    - Modified subroutine get_ind to always return 0 for final
#      if no final consonant is in the composite syllable.
#      Previously it always added $tconMap[$m] to the final
#      consonant location even if there was no final consonant.
#    - Index arrays were extended to cover all of Johab encoded
#      Hangul, even though not all glyphs are used to generate
#      the Unicode Hangul Syllables range.
#    - Added comments on the letters in the letter arrays


# Conversion routine from Hangul Jamo index to glyph index 
# of Hangul "Johab-encoded" fonts  as used by 
# Hangul xterm, hanterm.
# The following routine is based on Hanterm by Song, Jaekyung
# available at ftp://ftp.kaist.ac.kr/hangul/terminal/hanterm

# Leading Consonant index values:
#
#    Modern Letters:                     Archaic Letters (no Romanization):
#
#       0 G  (choseong kiyeok)              19 (choseong kapyeounpieup) 
#       1 GG (choseong ssangkiyeok)         20 (choseong pieup-kiyeok)
#       2 N  (choseong nieun)               21 (choseong sios-kiyeok)
#       3 D  (choseong tikeut)              22 (choseong pieup-tikeut)
#       4 DD (choseong ssangtikeut)         23 (choseong sios-tikeut)
#       5 R  (choseong rieul)               24 (choseong sios-pieup)
#       6 M  (choseong mieum)               25 (choseong pieup-sios)
#       7 B  (choseong pieup)               26 (choseong pansios)
#       8 BB (choseong ssangpieup)          27 (choseong yesieung)
#       9 S  (choseong sios)                28 (choseong pieup-cieuc)
#      10 SS (choseong ssangsios)           29 (choseong sios-cieuc)
#      11 ieung (choseong ieung)            30 (choseong yeorinhieuh)
#      12 J  (choseong cieuc)
#      13 JJ (choseong ssangcieuc)
#      14 C  (choseong chieuch)
#      15 K  (choseong khieukh)
#      16 T  (choseong thieuth)
#      17 P  (choseong phieuph)
#      18 H  (choseong hieuh)
#

# Middle Letter index values:
#
#    Modern Letters:                     Archaic Letters (no Romanization):
#
#       0 Filler (blank)                    22 YO-YA   (jungseong yo-ya)
#       1 A   (jungseong a)                 23 YO-YAE  (jungseong yo-yae)
#       2 AE  (jungseong ae)                24 YO-I    (jungseong yo-i)
#       3 YA  (jungseong ya)                25 YU-YEO  (jungseong yu-yeo)
#       4 YAE (jungseong yae)               26 YU-YE   (jungseong yu-ye)
#       5 EO  (jungseong eo)                27 YU-I    (jungseong yu-i)
#       6 E   (jungseong e)                 28 araea   (jungseong araea)
#       7 YEO (jungseong yeo)               29 araea-i (jungseong araea-i)
#       8 YE  (jungseong ye)
#       9 O   (jungseong o)
#      10 WA  (jungseong wa)
#      11 WAE (jungseong wae)
#      12 OE  (jungseong oe)
#      13 YO  (jungseong yo)
#      14 U   (jungseong u)
#      15 WEO (jungseong weo)
#      16 WE  (jungseong we)
#      17 WI  (jungseong wi)
#      18 YU  (jungseong yu)
#      19 EU  (jungseong eu)
#      20 YI  (jungseong yi)
#      21 I   (jungseong i)
#

# Terminal (Final) Letter index values:
#
#    Modern Letters:                     Archaic Letters (no Romanization):
#
#       0 Filler (blank)                    28 (jongseong rieul-hieuh)
#       1 G  (jongseong kiyeok)             29 (jongseong mieum-kiyeok)
#       2 GG (jongseong ssangkiyeok)        30 (jongseong yeorinhieuh)
#       3 GS (jongseong kiyeok-sios)        31 (jongseong yesieung)
#       4 N  (jongseong nieun)
#       5 NJ (jongseong nieun-cieuc)
#       6 NH (jongseong niuen-hieuh)
#       7 D  (jongseong tikeut)
#       8 L  (jongseong rieul)
#       9 LG (jongseong rieul-kiyeok)
#      10 LM (jongseong rieul-mieum)
#      11 LB (jongseong rieul-pieup)
#      12 LS (jongseong rieul-sios)
#      13 LT (jongseong rieul-thieuth)
#      14 LP (jongseong rieul-phieuph)
#      15 LH (jongseong rieul-hieuh)
#      16 M  (jongseong mieum)
#      17 B  (jongseong pieup)
#      18 BS (jongseong pieup-sios)
#      19 S  (jongseong sios)
#      20 SS (jongseong ssangsios)
#      21 NG (jongseong ieung)
#      22 J  (jongseong cieuc)
#      23 C  (jongseong chieuch)
#      24 K  (jongseong khieukh)
#      25 T  (jongseong thieuth)
#      26 P  (jongseong phieuph)
#      27 H  (jongseong hieuh)
#


# The base font index for leading consonants
  @lconBase= (
          1,  11,  21,  31,  41,  51,  # G, GG, N, D, DD, R
         61,  71,  81,  91, 101, 111,  # M, B, BB, S, SS, ieung
        121, 131, 141, 151, 161, 171,  # J, JJ, C, K, T, P
        181,                           # H -- end of modern set
        191, 201, 211, 221, 231, 241,  # 
        251, 261, 271, 281, 291, 301
  );

  # The base index for vowels

  @vowBase = (
        0,311,314,317,320,323,   # (Fill), A, AE, YA, YAE, EO
        326,329,332,335,339,343, # E, YEO, YE, O, WA, WAE
        347,351,355,358,361,364, # OI, YO, U, WEO, WE, WI
        367,370,374,378,         # YU, EU, UI, I     -- end of modern set
        381, 384, 387,           # YO-YA, YO-YAE, YO-YI
        390, 393, 396,           # YU-YEO, YU-YE, YU-I
        399, 402                 # araea, araea-i
  );

  # The base font index for trailing consonants

  @tconBase = (
        # modern trailing consonants (filler + 27)
        0,
        405, 409, 413, 417, 421,  #  G, GG, GS,  N, NJ
        425, 429, 433, 437, 441,  # NH,  D,  L, LG, LM
        445, 449, 453, 457, 461,  # LB, LS, LT, LP, LH
        465, 469, 473, 477, 481,  #  M,  B, BS,  S, SS
        485, 489, 493, 497, 501,  # NG,  J,  C,  K,  T
        505, 509,                 #  P,  H     -- end of modern set
        513, 517, 521, 525
   );

    # The mapping from vowels to leading consonant type
    # in absence of trailing consonant

    @lconMap1 = (
        0,0,0,0,0,0,     # (Fill), A, AE, YA, YAE, EO
        0,0,0,1,3,3,     # E, YEO, YE, O, WA, WAE
        3,1,2,4,4,4,     # OI, YO, U, WEO, WE, WI
        2,1,3,0,         # YU, EU, UI, I    -- end of modern set
        3,3,3,4,4,4,     # YO-YA, YO-YAE, YO-I, YU-YEO, YU-YE, YU-I
        1,3              # araea, araea-i
    );

    # The mapping from vowels to leading consonant type
    # in presence of trailing consonant

    @lconMap2 = (
        5,5,5,5,5,5,     #  (Fill), A, AE, YA, YAE, EO
        5,5,5,6,8,8,     #  E, YEO, YE, O, WA, WAE
        8,6,7,9,9,9,     #  OI, YO, U, WEO, WE, WI
        7,6,8,5,         #  YU, EU, UI, I    -- end of modern set
        8,8,8,9,9,9,     # YO-YA, YO-YAE, YO-I, YU-YEO, YU-YE, YU-I
        6,8              # araea, araea-i
    );

    #  vowel type ; 1 = o and its alikes, 0 = others

    @vowType = (
        0,0,0,0,0,0,
        0,0,0,1,1,1,
        1,1,0,0,0,0,
        0,1,1,0,         # end of modern set
        1,1,1,0,0,0,1,1
    );

    #  The mapping from trailing consonants to vowel type

    @tconType = (
        0, 1, 1, 1, 2, 1,
        1, 1, 1, 1, 1, 1,
        1, 1, 1, 1, 1, 1,
        1, 1, 1, 1, 1, 1,
        1, 1, 1, 1,        # end of moder set
        1, 1, 1, 1
    );

    #  The mapping from vowels to trailing consonant type

    @tconMap = (
        0, 0, 2, 0, 2, 1,  # (Fill), A, AE, YA, YAE, EO
        2, 1, 2, 3, 0, 0,  # E, YEO, YE, O, WA, WAE
        0, 3, 3, 1, 1, 1,  # OI, YO, U, WEO, WE, WI
        3, 3, 0, 1,        # YU, EU, UI, I   -- end of modern set
        0, 0, 0, 1, 1, 1,  # YO-YA, YO-YAE, YO-I, YU-YEO, YU-YE, YU-I
        3, 0               # araea, araea-i
    );



# read in BITMAP patterns for Jamos from JOHAB-encoded BDF font file 
# thru STDIN

$BITMAP=0;
while (<>) {
  if (/^ENCODING\s+(\d+)/) { $i = $1; $jamo[$i]=""; }
  elsif (/^BITMAP/) { $BITMAP=1; }
  elsif (/^ENDCHAR/) { $BITMAP=0; 
  }
  elsif ($BITMAP) { 
    y/a-f/A-F/; 
    s/\n$//; 
    $jamo[$i] = $jamo[$i] . $_;
  }
}

for ( $j=0 ; $j <  11172 ;  $j++ ) {

   $init = int( $j / 21 / 28) ;
   $medial = int($j / 28 ) % 21+1 ; 
   $final = $j % 28;
  
   printf ("%04X:%64s\n", $j+0xAC00, &compose_hangul($init,$medial,$final)) or die ("Cannot print to stdout.\n");

}

sub compose_hangul
{
   local($l,$m,$f) = @_; 

   @l_bit = unpack("a2" x 32, $jamo[&get_ind($l,$m,$f,1)]);
   @m_bit = unpack("a2" x 32, $jamo[&get_ind($l,$m,$f,2)]);
   @f_bit = unpack("a2" x 32, $jamo[&get_ind($l,$m,$f,3)]);
 

   for ( $i = 0; $i < 32; $i++) {
      $bit[$i]=sprintf("%02X", 
         hex($l_bit[$i]) | hex($m_bit[$i]) | hex($f_bit[$i]) ) or die ("Cannot print to stdout.\n");
   }

   return pack("a2" x 32, @bit );

}

sub get_ind
{
  local($l,$m,$f,$wh) = @_;

# ($l = 0 && $l < 19 && $m =0 && $m < 21  && $f =0 && $f < 28) or 
#   die ("$0: get_ind() : invalid Jamo index\n");

  if ( $wh == 1 ) {  # leading consonant index; no final consonant if $f==0
    $ind = $lconBase[$l] + 
           ($f > 0 ?  $lconMap2[$m] : $lconMap1[$m] ) ;
  } 
  elsif ( $wh == 2 ) { # medial vowel index

     $ind = $vowBase[$m];
     if ( $vowType[$m] == 1 ) {
     # For vowels 'o' and alikes,
     # Giyeok and Kieuk get special treatment
         $ind += ( ($l==0 || $l == 15) ? 0 : 1)
                 + ($f > 0 ? 2 : 0 );
     }
     else {
         $ind+= $tconType[$f];
     }
  }
  else {
      if ($f == 0) {
         $ind = 0;
      }
      else {
         $ind = $tconBase[$f] + $tconMap[$m];
      }
  }
  return $ind;
}