inCHISE

   1 #!/usr/bin/perl -w -CSD
   2
   3 # ver.0.2
   4
   5 use strict;
   6 use vars qw($omegadb_path
   7             $opt_protrude $opt_allow_unify
   8             $opt_use_kage_for_Ext_B
   9             %opt_order %order %order_map
  10             $opt_in_cs $opt_out_cs
  11             $opt_preserve_spaces
  12             $opt_help $usage
  13             $in_cs $out_cs $i @chars
  14             $char $char_id $out_char
  15             $char_unified @chars_unified
  16             $ids $ids_argc %ids $idsdb $geta
  17             $idsdata_file $ids_start $font_start
  18             @CDP @HZK @GT
  19             );
  20 use Getopt::Long;
  21 use utf8;
  22 use Fcntl ':flock';
  23 use Chise_utils ':all';
  24 require 5.008;
  25
  26 ### Options ###
  27
  28 #$opt_order{'UniMulti'}='jcgk';
  29 $opt_order{'UniMulti'}='jGcgkHC';
  30 $opt_order{'UniCNS'}='c';
  31 $opt_order{'UniGB'}='g';
  32 $opt_order{'UniJIS'}='j';
  33 $opt_order{'UniKS'}='k';
  34
  35 $opt_allow_unify=1; # 1=true, 0=false.
  36 $opt_protrude=0;# 1=true, 0=false.
  37
  38 # currently does not work.
  39 $opt_use_kage_for_Ext_B=0;# 1=true, 0=false.
  40
  41 ### End ###
  42
  43 my $strictly_forbidden_after = '「【『［（〈‘‛“‟（〔｛《{\[\(〖｛｢';
  44
  45 my $forbidden_after = "\x{0000}";
  46
  47 # ￥¥＄$〒♯＃#￠¢￡£＠@§
  48 my $slightly_forbidden_after = '￥¥＄$〒♯＃#￠¢￡£＠@§';
  49
  50 # $strictly_forbidden_before
  51 # All these characters are allowed to protrude
  52 # in the right margin
  53 my $strictly_forbidden_before=
  54     '!,.:;?、。！，．：；？。\)\]}’”〉》」』】〕〗）］｝」';
  55
  56 my $forbidden_before
  57     = 'ー々ぁぃぅぇぉゃゅょっゎァィゥェォャュョッヮヵヶ';
  58
  59 my $slightly_forbidden_before
  60     = '\x{000a}\#\-‐−‰′″℃゛゜ゝゞヽヾ＂％－ﾞﾟ';
  61
  62 #my $asian = '\x{1100}-\x{11FF}\x{2E80}-\x{D7AF}\x{F900}-\x{FAFF}\x{FE30}-\x{FE4F}\x{FF00}-\x{FFFFFF}'; # need to be fixed.
  63 my $asian = '\x{2E80}-\x{312f}\x{3190}-\x{ABFF}\x{F900}-\x{FAFF}\x{FE30}-\x{FE4F}\x{FF00}-\x{FFFFFF}'; # need to be fixed.
  64
  65 my $space = '\x{0020}\x{0009}\x{000A}\x{000C}\x{000D}';
  66
  67 my $hangul='\x{1100}-\x{11ff}\x{3130}-\x{318f}\x{AC00}-\x{D7AF}';
  68
  69 my %tex_meta=('#'=>'\#',
  70               '$'=>'\\textdollar{}',
  71               '%'=>'\%',
  72               '&'=>'\&',
  73               '{'=>'\\textbraceleft{}',
  74               '}'=>'\\textbraceright{}',
  75               '\\'=>'\\textbackslash{}',
  76               '_'=>'\\textunderscore',
  77              );
  78
  79 my $tex_meta_re=join('|',map {quotemeta($_)} keys %tex_meta);
  80
  81 &GetOptions("in=s"=>\$opt_in_cs,
  82             "out=s"=>\$opt_out_cs,
  83             "help",\$opt_help);
  84
  85 $usage=<<EOF;
  86 Usage: $0 -i <input coding system> -o <cmap encoding>
  87     input coding system:
  88       Utf8mcs, Utf8cns, Utf8gb, Utf8jis, Utf8ks, Utf8big5
  89     cmap encoding:
  90       UniCNS, UniGB, UniJIS, UniKS, UniMulti
  91 EOF
  92
  93 if($opt_in_cs or $opt_out_cs){
  94     $in_cs=$opt_in_cs;
  95     $out_cs=$opt_out_cs;
  96 }elsif(@ARGV==0){
  97     ($in_cs
  98      ,$out_cs
  99      ,$opt_preserve_spaces)
 100         =($0=~/(Utf8mcs|Utf8cns|Utf8gb|Utf8jis|Utf8ks|Utf8big5)
 101           To (UniCNS|UniGB|UniJIS|UniKS|UniMulti)
 102           (Sp)?/ox);
 103 }
 104
 105 # $in_cs:
 106 #   Utf8mcs,Utf8cns,Utf8gb,Utf8jis,Utf8ks,Utf8big5
 107 # $out_cs:
 108 #   UniCNS,UniGB,UniJIS,UniKS,UniMulti
 109
 110 $in_cs=~s/Utf8/ucs\@/;
 111
 112 if($opt_help
 113    or not defined($in_cs)
 114    or not defined($out_cs)){
 115     print $usage;
 116     exit 1;
 117 }
 118
 119 $omegadb_path=~s!/$!!;
 120
 121 $idsdata_file="$omegadb_path/idsdata.pl";
 122 $ids_start=0x00;
 123 $font_start=0;
 124
 125 if(-e $idsdata_file){
 126     open(IDSDATA,"+<:utf8",$idsdata_file) or die;
 127     flock(IDSDATA,LOCK_EX);
 128     seek(IDSDATA,0,0);
 129     while(<IDSDATA>){
 130         eval $_;
 131     }
 132     seek(IDSDATA,0,0);
 133 #         require $idsdata_file;
 134 }else{
 135     open(IDSDATA,">:utf8",$idsdata_file) or die;
 136     flock(IDSDATA,LOCK_EX);
 137     seek(IDSDATA,0,0);
 138 }
 139
 140 $ids_argc=0;
 141 $ids="";
 142
 143 #$geta=pack("U",0x3013);
 144 $geta=pack("U",0xfffd);
 145
 146 @GT=(#"=gt","=gt-k",
 147      "=gt-pj-1","=gt-pj-2","=gt-pj-3","=gt-pj-4","=gt-pj-5",
 148      "=gt-pj-6","=gt-pj-7","=gt-pj-8","=gt-pj-9","=gt-pj-10",
 149      "=gt-pj-11"
 150      #,"=gt-pj-k1","=gt-pj-k2"
 151      );
 152 @HZK=("=hanziku-1","=hanziku-2","=hanziku-3","=hanziku-4",
 153       "=hanziku-5","=hanziku-6","=hanziku-7","=hanziku-8",
 154       "=hanziku-9","=hanziku-10","=hanziku-11","=hanziku-12");
 155 @CDP=("=big5-cdp");
 156
 157 %order_map=('c'=>'UniCNS',
 158             'g'=>'UniGB',
 159             'j'=>'UniJIS',
 160             'k'=>'UniKS',
 161             'G'=>'GT',
 162             'H'=>'HZK',
 163             'C'=>'CDP',
 164             );
 165
 166 foreach $out_cs ('UniCNS','UniGB','UniJIS','UniKS','UniMulti'){
 167     if(defined($opt_order{$out_cs})){
 168         if($opt_order{$out_cs}=~/^[cgjkGHC]+$/){
 169             @{$order{$out_cs}}=map {$order_map{$_}}
 170             (split(//,$opt_order{$out_cs}));
 171         }else{
 172             print STDERR "Invalid order for $out_cs!\n";
 173             exit 1;
 174         }
 175     }
 176 }
 177
 178 while(<>){
 179     utf8::decode($_);
 180
 181     print '\relax{}';
 182
 183     if($in_cs ne 'ucs@mcs'){
 184         s/(.)/&get_char_in_mcs($1,$in_cs)/ge;
 185     }
 186     s/(amp.+?;)/&de_tex_er($1)/ge;
 187 #    s/(&.+?;)/&de_tex_er($1)/ge;
 188     s/([$asian])\s+/$1/go unless($opt_preserve_spaces);
 189     s/\s+([$asian])/$1/go unless($opt_preserve_spaces);
 190     s/([$asian])\s*([^$asian$space])/$1 $2/go;
 191     s/([^$asian$idc])\s*([$asian])/$1 $2/go;
 192     s/\-\-\-/pack("U",0x2014)/geo;# EM DASH
 193     s/\-\-/pack("U",0x2013)/geo;# EN DASH
 194 #    s/\`\`/pack("U",0x201f)/geo;# DOUBLE HIGH-REVERSED-9 QUOTATION MARK
 195     s/\`\`/pack("U",0x201c)/geo;# _TeX compatible_
 196 #    s/\`/pack("U",0x201b)/geo;# SINGLE HIGH-REVERSED-9 QUOTATION MARK
 197     s/\`/pack("U",0x2018)/geo;# _TeX compatible_
 198     s/\'\'/pack("U",0x201d)/geo;# RIGHT DOUBLE QUOTATION MARK
 199     s/\'/pack("U",0x2019)/geo;# RIGHT DOUBLE QUOTATION MARK
 200     @chars=split(//);
 201   CHAR:
 202     for($i=0;$i<=$#chars;$i++){
 203         $char=$chars[$i];
 204         $char_id=unpack("U",$char);
 205
 206         if($char_id<=0x20){
 207             # add quarter space between asian and non-asian.
 208             if($i>0 and $i<$#chars){
 209                 if(($chars[$i-1]=~/[$asian]/
 210                     and $chars[$i+1]=~/[^$asian]/
 211                       and $chars[$i+1]=~/[$strictly_forbidden_before$slightly_forbidden_before]/)
 212                    or($chars[$i-1]=~/[^$asian]/
 213                       and $chars[$i+1]=~/[$asian]/
 214                       and $chars[$i-1]=~/[$strictly_forbidden_after$slightly_forbidden_after]/)
 215                    ){
 216                     print '';
 217                 }elsif(($chars[$i-1]=~/[$asian]/
 218                         and $chars[$i+1]=~/[^$asian]/)){
 219                     print '\unskip\kern.25ex';
 220                 }elsif($chars[$i-1]=~/[^$asian]/
 221                        and $chars[$i+1]=~/[$asian]/){
 222                     print '\kern.25ex';
 223                 }else{
 224                     print ' ';
 225                 }
 226             }else{
 227                 print ' ';
 228             }
 229             next CHAR;
 230         }elsif(($char_id>0x20 and $char_id<=0x021f)
 231                # Basic Latin
 232                # Latin-1 Supplement
 233                # Latin Extended-A
 234                # Latin Extended-B (not all)
 235                or($char_id>=0x0250 and $char_id<=0x02af)
 236                # IPA Extensions
 237                or($char_id>=0x0300 and $char_id<=0x033f)
 238                or($char_id>=0x0360 and $char_id<=0x036f)
 239                # Combining Diacritical Marks
 240                or($char_id>=0x1e00 and $char_id<=0x1eff)
 241                # Latin Extended Additional
 242                or($char_id>=0x0370 and $char_id<=0x03ff)
 243                # Greek and Coptic
 244                or($char_id>=0x0400 and $char_id<=0x04ff)
 245                # Cyrillic
 246                or($char_id>=0x0530 and $char_id<=0x058f)
 247                # Armenian
 248                or($char_id>=0x2010 and $char_id<=0x2046)
 249                # General Punctuation (partial)
 250                ){
 251             print &latin_parse();
 252             next CHAR;
 253         }elsif($char_id>=0x2ff0 and $char_id<=0x2fff){
 254             # Ideographic Description Characters
 255             print &ids_parse();
 256             next CHAR;
 257         }else{
 258             if(($out_char=&get_output_char($char,$out_cs))){
 259                 print $out_char,&add_break($i);
 260             }else{
 261                 if($opt_allow_unify){
 262                     @chars_unified=&get_chars_unified($char);
 263                     if(@chars_unified>0){
 264                         foreach $char_unified (@chars_unified){
 265                             if(($out_char
 266                                 =&get_output_char($char_unified,$out_cs))){
 267                                 print $out_char,&add_break($i);
 268                                 next CHAR;
 269                             }
 270                         }
 271                     }
 272                 }
 273                 if($opt_use_kage_for_Ext_B){
 274                     if($char_id >= 0x20000 && $char_id <=0x2a6df){
 275                         # CJK Unified Ideographs Extension B
 276                         if(not defined($ids{$char}) and $ids{$char}[1]>=0){
 277                             $ids{$char}[0]=$font_start;
 278                             $ids{$char}[1]=$ids_start;
 279                             $ids_start++;
 280                             if($ids_start>255){
 281                                 $ids_start=0;
 282                                 $font_start++;
 283                             }
 284                         }
 285                         print "{\\fontencoding{OT1}\\fontfamily{",
 286                             sprintf("chise%03d",$ids{$char}[0]) ,
 287                                 "}\\selectfont\\char$ids{$char}[1]}",&add_break($i);
 288                         next CHAR;
 289                     }
 290                 }
 291                 if($ids=&get_ids($char)){
 292                     print &get_macro_for_ids($ids),&add_break($i);
 293                 }else{
 294                     print '\rule{1ex}{1ex}',&add_break($i);
 295                 }
 296             }
 297         }
 298     }
 299 }
 300
 301 print IDSDATA 'use utf8;',"\n";
 302 foreach $ids (keys %ids){
 303     print IDSDATA '$ids{\'',$ids,'\'}='
 304     ,'[',join ",",@{$ids{$ids}},"];\n";
 305 }
 306 print IDSDATA '$font_start=',$font_start,";\n";
 307 print IDSDATA '$ids_start=',$ids_start,";\n";
 308 print IDSDATA "1;";
 309 flock(IDSDATA,LOCK_UN);
 310
 311 exit 0;
 312
 313 sub de_tex_er{
 314     my($er)=@_;
 315     my($prefix,$suffix);
 316     my($output_char,$atr,$value);
 317     $er=~/^(amp)(.*)(;)$/
 318         and $prefix=$1,$er=$2,$suffix=$3;
 319     $prefix or $prefix="",$suffix or $suffix="";
 320     if($er=~/^U[\+|\-]([a-fA-F\d]+)/){
 321         $output_char=pack("U",hex($1));
 322     }elsif($er=~/^(?:I\-)?($er_prefix_re)\-?([0-9a-fA-F]+)$/){
 323         ($atr,$value)=($1,$2);
 324         unless($er_alias{$atr}=~/daikanwa|gt/){
 325             $value=hex($value);
 326         }
 327         ($output_char)=&get_chars_matching($er_alias{$atr},$value);
 328     }
 329     if($output_char){
 330         return $output_char;
 331     }else{
 332         return $prefix.$er.$suffix;
 333     }
 334 }
 335
 336 sub add_break{
 337     my($i)=@_;
 338
 339     if($i<($#chars-1)){
 340         if(($chars[$i+1]=~m/[$strictly_forbidden_before]/o)
 341            and($chars[$i+2]=~m/[$strictly_forbidden_before]/o)){
 342             return "\\CJKunbreakablekernone{}";
 343         }elsif($opt_protrude){
 344             if(($chars[$i+1]=~m/[$strictly_forbidden_before]/o)
 345                and($chars[$i+2]=~m/[^$strictly_forbidden_before]/o)){
 346                 return "\\CJKunbreakablekernone \\CJKprotrude{}";
 347             }
 348         }
 349     }
 350     if(($i<$#chars)
 351        and($chars[$i+1]eq" ")){
 352        # preserve space.
 353         return "";
 354     }
 355     if(($i<$#chars)
 356        and($chars[$i+1]=~m/[$strictly_forbidden_before]/o)){
 357         return "\\CJKunbreakablekernone{}";
 358     }
 359     if($chars[$i]=~m/[$strictly_forbidden_after]/o){
 360         return "\\CJKunbreakablekernone{}";
 361     }
 362     if(($i<$#chars)
 363        and($chars[$i+1]=~m/[$forbidden_before]/o)){
 364         return "\\CJKunbreakablekerntwo{}";
 365
 366     }
 367     if($chars[$i]=~m/[$forbidden_after]/o){
 368         return "\\CJKunbreakablekerntwo{}";
 369     }
 370     if(($i<$#chars)
 371        and($chars[$i+1]=~m/[$slightly_forbidden_before]/o)){
 372         return "\\CJKunbreakablekernthree{}";
 373     }
 374     if($chars[$i]=~m/[$slightly_forbidden_after]/o){
 375         return "\\CJKunbreakablekernthree{}";
 376     }
 377     if($chars[$i]=~m/[$asian]/o){
 378         return "\\CJKbreakablekern{}";
 379     }
 380     if(($i<$#chars)and($chars[$i+1]=~m/[$asian]/o)){
 381         return "\\CJKbreakablekern{}";
 382     }
 383 }
 384
 385 sub latin_parse{
 386     # arguments: none
 387     # return: string for output with TeX macro.
 388     my($char_id);
 389     my $out_str="";
 390     while($i<=$#chars){
 391         $char_id=unpack("U",$chars[$i]);
 392         if($chars[$i]=~m/($tex_meta_re)/o){
 393             $out_str.=$tex_meta{$chars[$i]};
 394         }elsif(($char_id>0x20 and $char_id<=0x021f)
 395            or($char_id>=0x0250 and $char_id<=0x02af)# IPA Extensions
 396            or($char_id>=0x0300 and $char_id<=0x033f)# Combining Diacritical Marks
 397            or($char_id>=0x0360 and $char_id<=0x036f)
 398            or($char_id>=0x0370 and $char_id<=0x03ff)# Greek and Coptic
 399            or($char_id>=0x0400 and $char_id<=0x04ff)# Cyrillic
 400            or($char_id>=0x0530 and $char_id<=0x058f)# Armenian
 401            ){
 402             $out_str.=$chars[$i];
 403         }elsif($char_id>=0x1e00 and $char_id<=0x1eff){
 404             # Latin Extended Additional
 405             # 0x1e00 -> 0x0600, etc.
 406             $out_str.=pack("U",$char_id-0x1800);
 407         }elsif($char_id>=0x2010 and $char_id<=0x2046){
 408             # General Punctuation (partial)
 409             $out_str.=pack("U",$char_id-0x1000);
 410         }else{
 411             $i--;
 412             last;
 413         }
 414         $i++;
 415     }
 416     return '{\fontencoding{OT1}\fontfamily{omlgc}\selectfont '.$out_str.'}';###UT1?
 417 }
 418
 419 sub ids_parse{
 420     # arguments: none
 421     # return: character for output,
 422     #          TeX macro for ids,
 423     #          or GETA character if ids is invalid.
 424     my($ids,$ids_argc)=&ids_rest("",0,$chars[$i]);
 425
 426     while($ids_argc>0){
 427         # We are in IDS.
 428         $i++;
 429         if($i>$#chars){
 430             print STDERR "IDS parse error: $ids\n";
 431             return $geta;
 432         }
 433
 434         ($ids,$ids_argc)=&ids_rest($ids,$ids_argc,$chars[$i]);
 435         if($ids_argc==0){
 436             if(($char=&get_char_for_ids($ids))
 437                and($out_char=&get_output_char($char,$out_cs))){
 438                 return $out_char;
 439             }else{
 440                 return &get_macro_for_ids($ids);
 441             }
 442         }
 443     }
 444 }
 445
 446 sub ids_rest{
 447     # arguments: <ids>, <rest number of arguments for ids>, <character>
 448     # return: ids and rest number of arguments for ids.
 449     my($ids,$ids_argc,$char)=@_;
 450     my($argc);
 451     $argc=&ids_argc($char);
 452     if($argc){
 453         $ids_argc+=$ids_argc==0?$argc:$argc-1;
 454     }else{
 455         $ids_argc--;
 456     }
 457     $ids.=$char;
 458     return ($ids,$ids_argc);
 459 }
 460
 461 sub get_macro_for_ids{
 462     # argument: <ids>
 463     # return: TeX macro for ids
 464     #          or GETA character if ids is invalid for KAGE.
 465     my($ids)=@_;
 466     # $ids=&normalize_ids($ids,"UniJIS");
 467     return $geta if(($ids!~/[$idc]/)
 468                     or($ids=~/[\x{10000}-]/));
 469                     #irregular for KAGE.
 470     if(not defined($ids{$ids})){
 471         $ids{$ids}[0]=$font_start;
 472         $ids{$ids}[1]=$ids_start;
 473         $ids_start++;
 474     }
 475     if($ids_start>255){
 476         $ids_start=0;
 477         $font_start++;
 478     }
 479     return "{\\fontencoding{OT1}\\fontfamily{"
 480         .sprintf("chise%03d",$ids{$ids}[0])
 481         ."}\\selectfont\\char$ids{$ids}[1]}";
 482 }
 483
 484 sub normalize_ids{
 485     # argument: <ids>, <output coding system>
 486     # return: ids or GETA character if ids is invalid for KAGE.
 487     my($ids,$out_cs)=@_;
 488     $out_cs=~s/Uni(.+)/'ucs@'.lc($1)/e;
 489
 490     my $output_ids="";
 491     my($char,$output_char_id);
 492     while($ids=~m/(.)/g){
 493         $char=$1;
 494         if($char=~/[$idc]/){
 495             $output_ids.=$char;
 496         }elsif($output_char_id=&get_char_attribute($char,"=$out_cs")
 497            or $output_char_id=&get_char_attribute($char,"=ucs")
 498            or $output_char_id=&get_char_attribute($char,"=>$out_cs")
 499            or $output_char_id=&get_char_attribute($char,"=>ucs")
 500            or $output_char_id=&get_char_attribute($char,"=>ucs*")
 501               ){
 502             $output_ids.=pack("U",$output_char_id);
 503         }else{
 504             return $geta;
 505         }
 506     }
 507     return $output_ids;
 508 }
 509
 510 sub get_output_char{
 511     # argument: <char>
 512     # return: character in output coding system or TeX macro or undef.
 513     my($char,$out_cs)=@_;
 514     my($out_char_id,$suffix);
 515     my($gt,$hzk,$cdp);
 516
 517     foreach $out_cs (@{$order{$out_cs}}){
 518         if($out_cs eq 'UniJIS'
 519            and &get_char_attribute($char,"vnd-adobe-cid-unijis-utf16-h")){
 520             if($out_char_id=&get_char_attribute($char,'=ucs@jis')
 521                or $out_char_id=&get_char_attribute($char,'=ucs')
 522                or $out_char_id=&get_char_attribute($char,'=>ucs@jis')
 523                or $out_char_id=&get_char_attribute($char,'=>ucs')
 524                or $out_char_id=&get_char_attribute($char,'=>ucs*')
 525                ){
 526                 return "{\\selectjisfont\\char$out_char_id}";
 527             }
 528         }elsif($out_cs eq 'UniGB'
 529                and &get_char_attribute($char,"vnd-adobe-cid-unigb-ucs2-h")){
 530             if($out_char_id=&get_char_attribute($char,'=ucs@gb')
 531                or $out_char_id=&get_char_attribute($char,'=ucs')
 532                or $out_char_id=&get_char_attribute($char,'=>ucs@gb')
 533                or $out_char_id=&get_char_attribute($char,'=>ucs')
 534                or $out_char_id=&get_char_attribute($char,'=>ucs*')
 535                ){
 536                 return "{\\selectgbsfont\\char$out_char_id}";
 537             }
 538         }elsif($out_cs eq 'UniCNS'
 539                and &get_char_attribute($char,"vnd-adobe-cid-unicns-ucs2-h")){
 540             if($out_char_id=&get_char_attribute($char,'=ucs@cns')
 541                or $out_char_id=&get_char_attribute($char,'=ucs')
 542                or $out_char_id=&get_char_attribute($char,'=>ucs@cns')
 543                or $out_char_id=&get_char_attribute($char,'=>ucs')
 544                or $out_char_id=&get_char_attribute($char,'=>ucs*')
 545                ){
 546                 return "{\\selectcnsfont\\char$out_char_id}";
 547             }
 548         }elsif($out_cs eq 'UniKS'
 549                and &get_char_attribute($char,"vnd-adobe-cid-uniks-ucs2-h")){
 550             if($out_char_id=&get_char_attribute($char,'=ucs@ks')
 551                or $out_char_id=&get_char_attribute($char,'=ucs')
 552                or $out_char_id=&get_char_attribute($char,'=>ucs@ks')
 553                or $out_char_id=&get_char_attribute($char,'=>ucs')
 554                or $out_char_id=&get_char_attribute($char,'=>ucs*')
 555                ){
 556                 return "{\\selectksxfont\\char$out_char_id}";
 557             }
 558         }elsif($out_cs eq 'GT'){
 559             return $gt if($gt=&get_macro_for_GT($char));
 560         }elsif($out_cs eq 'HZK'){
 561             return $hzk if($hzk=&get_macro_for_HZK($char));
 562         }elsif($out_cs eq 'CDP'){
 563             return $cdp if($cdp=&get_macro_for_CDP($char));
 564         }
 565     }
 566     return undef;
 567 }
 568
 569 sub get_ids{
 570     # argument: <character>
 571     # return: ids
 572     my($char)=@_;
 573     my $ids="";
 574     $ids=&get_char_attribute($char,"ids-aggregated")
 575         or $ids=&get_char_attribute($char,"ids");
 576 #    $ids=&get_char_attribute($char,"ids-decomposed")
 577 #       or $ids=&get_char_attribute($char,"ids");
 578 #         or $ids=&get_char_attribute($char,"ideographic-structure");
 579 #    $ids=~s/[? ()]//g;
 580     return $ids;
 581 }
 582
 583 sub get_char_for_ids{
 584     # argument: <ideographic description sequence>
 585     # return: char or undef.
 586     my($ids)=@_;
 587     my($output_char);
 588
 589     if(($output_char)=&get_chars_matching("ids",$ids)){
 590         return $output_char;
 591     }else{
 592         return undef;
 593     }
 594 }
 595
 596 sub get_char_in_mcs{
 597     # argument: <char>, <input coding system>
 598     # return:   char in ucs@mcs.
 599     my($char,$in_cs)=@_;
 600     my($output_char);
 601
 602     return $char if($in_cs eq 'ucs@mcs');
 603
 604     if(($output_char)=&get_chars_matching("=$in_cs",unpack("U",$char))){
 605         return $output_char;
 606     }else{
 607         return $char;
 608     }
 609 }
 610
 611 sub get_chars_unified{
 612     my($char)=@_;
 613     my($chars,$ucs,$char_ucs);
 614     my(@chars);
 615
 616     if($chars=&get_char_attribute($char,'->ucs-unified')){
 617         $chars=~s/^\((.*)\)$/$1/;
 618         return (split(/\s*\?/,$chars));
 619     }elsif($ucs=&get_char_attribute($char,'=>ucs*')
 620           or $ucs=&get_char_attribute($char,'=>ucs')){
 621         $char_ucs=pack("U",$ucs);
 622         if($chars=&get_char_attribute($char_ucs,'->ucs-unified')){
 623             $chars=~s/^\((.*)\)$/$1/;
 624             @chars=grep {not /^$char$/}
 625                 (split(/\s*\?/,$chars));
 626             push(@chars,$char_ucs);
 627             return @chars;
 628         }
 629     }
 630 }
 631
 632 sub get_macro_for_GT{
 633     # argument: <char>
 634     # return: TeX macro for GT fonts or undef.
 635     my($char)=@_;
 636     my($gt,$GT);
 637     foreach (@GT){
 638         if($gt=&get_char_attribute($char,$_)){
 639             m/gt\-pj\-(\d+)/ and $GT=$1;
 640             last;
 641         }
 642     }
 643     if($gt){
 644         return "{\\fontencoding{OT1}\\fontfamily{"
 645             .sprintf("gt%02d",$GT)
 646             ."}\\selectfont\\char".($gt|0x8080)."}";
 647     }else{
 648         return undef;
 649     }
 650 }
 651
 652 sub get_macro_for_HZK{
 653     # argument: <char>
 654     # return: TeX macro for Hanziku fonts or undef.
 655     my($char)=@_;
 656     my($hzk,$HZK);
 657     foreach (@HZK){
 658         if($hzk=&get_char_attribute($char,$_)){
 659             m/hanziku\-(\d+)/ and $HZK=$1;
 660             last;
 661         }
 662     }
 663     if($hzk){
 664         return "{\\fontencoding{OT1}\\fontfamily{"
 665             .sprintf("hzk%02d",$HZK)
 666             ."}\\selectfont\\char".$hzk."}";
 667     }else{
 668         return undef;
 669     }
 670 }
 671
 672 sub get_macro_for_CDP{
 673     # argument: <char>
 674     # return: TeX macro for CDP fonts or undef.
 675     my($char)=@_;
 676     my($cdp,$ucs);
 677     foreach (@CDP){
 678         if($cdp=&get_char_attribute($char,$_)){
 679             last;
 680         }
 681     }
 682     if($cdp){
 683         $ucs=&get_char_attribute(&get_chars_matching("=big5-pua",$cdp),"=ucs");
 684         if($ucs){
 685             return "{\\fontencoding{OT1}\\fontfamily{cdp}\\selectfont\\char$ucs}";
 686         }else{
 687             print STDERR "This should not happen.\n";
 688             print STDERR "ucs code point of CDP: $cdp not found.\n";
 689         }
 690     }else{
 691         return undef;
 692     }
 693 }