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_help $usage
  12             $in_cs $out_cs $i @chars
  13             $char $char_id $out_char
  14             $char_unified @chars_unified
  15             $ids $ids_argc %ids $idsdb $geta
  16             $idsdata_file $ids_start $font_start
  17             @CDP @HZK @GT
  18             );
  19 use Getopt::Long;
  20 use utf8;
  21 use Fcntl ':flock';
  22 use Chise_utils ':all';
  23 require 5.008;
  24
  25 ### Options ###
  26
  27 #$opt_order{'UniMulti'}='jcgk';
  28 $opt_order{'UniMulti'}='jGcgkHC';
  29 $opt_order{'UniCNS'}='c';
  30 $opt_order{'UniGB'}='g';
  31 $opt_order{'UniJIS'}='j';
  32 $opt_order{'UniKS'}='k';
  33
  34 $opt_allow_unify=1; # 1=true, 0=false.
  35 $opt_protrude=0;# 1=true, 0=false.
  36
  37 # currently does not work.
  38 $opt_use_kage_for_Ext_B=0;# 1=true, 0=false.
  39
  40 ### End ###
  41
  42 my $strictly_forbidden_after = '「【『［（〈“‘‘（〔｛《{\[\(\x{3016}｛｢';
  43 #       \x{3016} | # white 【
  44
  45 my $forbidden_after = "\x{0000}";
  46
  47 # ￥¥＄$〒♯＃#￠¢￡£＠@§
  48 my $slightly_forbidden_after = '￥¥＄$〒♯＃#￠¢￡£＠@§';
  49
  50 # $strictly_forbidden_before
  51 # All these characters are allowed to protrude
  52 # in the right margin
  53 my $strictly_forbidden_before=
  54     '!,.:;?、。！，．：；？。\)#}’”〉》」』】〕\x{3017}）］｝｝」\]';
  55 ###       \x{3017} | # white 】
  56
  57 my $forbidden_before
  58     = 'ー々ぁぃぅぇぉゃゅょっゎァィゥェォャュョッヮヵヶ';
  59
  60 my $slightly_forbidden_before
  61     = '\x{000a}\#\-‐−‰′″℃゛゜ゝゞヽヾ＂％－ﾞﾟ';
  62
  63 #my $asian = '\x{1100}-\x{11FF}\x{2E80}-\x{D7AF}\x{F900}-\x{FAFF}\x{FE30}-\x{FE4F}\x{FF00}-\x{FFFFFF}'; # need to fix.
  64 my $asian = '\x{2E80}-\x{312f}\x{3190}-\x{ABFF}\x{F900}-\x{FAFF}\x{FE30}-\x{FE4F}\x{FF00}-\x{FFFFFF}'; # need to fix.
  65
  66 my $space = '\x{0020}\x{0009}\x{000A}\x{000C}\x{000D}';
  67
  68 my $hangul='\x{1100}-\x{11ff}\x{3130}-\x{318f}\x{AC00}-\x{D7AF}';
  69
  70 my %tex_meta=('#'=>'\#',
  71               '$'=>'\\textdollar{}',
  72               '%'=>'\%',
  73               '&'=>'\&',
  74               '{'=>'\\textbraceleft{}',
  75               '}'=>'\\textbraceright{}',
  76               '\\'=>'\\textbackslash{}',
  77               '_'=>'\\textunderscore',
  78              );
  79
  80 my $tex_meta_re=join('|',map {quotemeta($_)} keys %tex_meta);
  81
  82 &GetOptions("in=s"=>\$opt_in_cs,
  83             "out=s"=>\$opt_out_cs,
  84             "help",\$opt_help);
  85
  86 $usage=<<EOF;
  87 Usage: $0 -i <input coding system> -o <cmap encoding>
  88     input coding system:
  89       Utf8mcs, Utf8cns, Utf8gb, Utf8jis, Utf8ks, Utf8big5
  90     cmap encoding:
  91       UniCNS, UniGB, UniJIS, UniKS, UniMulti
  92 EOF
  93
  94 if($opt_in_cs or $opt_out_cs){
  95     $in_cs=$opt_in_cs;
  96     $out_cs=$opt_out_cs;
  97 }elsif(@ARGV==0){
  98     ($in_cs,$out_cs)=($0=~/(Utf8.+)To(\w+)/);
  99 }
 100
 101 # $in_cs:
 102 #   Utf8mcs,Utf8cns,Utf8gb,Utf8jis,Utf8ks,Utf8big5
 103 # $out_cs:
 104 #   UniCNS,UniGB,UniJIS,UniKS,UniMulti
 105
 106 $in_cs=~s/Utf8/ucs\@/;
 107
 108 if($opt_help
 109    or not defined($in_cs)
 110    or not defined($out_cs)){
 111     print $usage;
 112     exit 1;
 113 }
 114
 115 $omegadb_path=~s!/$!!;
 116
 117 $idsdata_file="$omegadb_path/idsdata.pl";
 118 $ids_start=0x00;
 119 $font_start=0;
 120
 121 if(-e $idsdata_file){
 122     open(IDSDATA,"+<:utf8",$idsdata_file) or die;
 123     flock(IDSDATA,LOCK_EX);
 124     seek(IDSDATA,0,0);
 125     while(<IDSDATA>){
 126         eval $_;
 127     }
 128     seek(IDSDATA,0,0);
 129 #         require $idsdata_file;
 130 }else{
 131     open(IDSDATA,">:utf8",$idsdata_file) or die;
 132     flock(IDSDATA,LOCK_EX);
 133     seek(IDSDATA,0,0);
 134 }
 135
 136 $ids_argc=0;
 137 $ids="";
 138
 139 #$geta=pack("U",0x3013);
 140 $geta=pack("U",0xfffd);
 141
 142 @GT=(#"=gt","=gt-k",
 143      "=gt-pj-1","=gt-pj-2","=gt-pj-3","=gt-pj-4","=gt-pj-5",
 144      "=gt-pj-6","=gt-pj-7","=gt-pj-8","=gt-pj-9","=gt-pj-10",
 145      "=gt-pj-11"
 146      #,"=gt-pj-k1","=gt-pj-k2"
 147      );
 148 @HZK=("=hanziku-1","=hanziku-2","=hanziku-3","=hanziku-4",
 149       "=hanziku-5","=hanziku-6","=hanziku-7","=hanziku-8",
 150       "=hanziku-9","=hanziku-10","=hanziku-11","=hanziku-12");
 151 @CDP=("=big5-cdp");
 152
 153 %order_map=('c'=>'UniCNS',
 154             'g'=>'UniGB',
 155             'j'=>'UniJIS',
 156             'k'=>'UniKS',
 157             'G'=>'GT',
 158             'H'=>'HZK',
 159             'C'=>'CDP',
 160             );
 161
 162 foreach $out_cs ('UniCNS','UniGB','UniJIS','UniKS','UniMulti'){
 163     if(defined($opt_order{$out_cs})){
 164         if($opt_order{$out_cs}=~/^[cgjkGHC]+$/){
 165             @{$order{$out_cs}}=map {$order_map{$_}}
 166             (split(//,$opt_order{$out_cs}));
 167         }else{
 168             print STDERR "Invalid order for $out_cs!\n";
 169             exit 1;
 170         }
 171     }
 172 }
 173
 174 while(<>){
 175     utf8::decode($_);
 176
 177     if($in_cs ne 'ucs@mcs'){
 178         s/(.)/&get_char_in_mcs($1,$in_cs)/ge;
 179     }
 180     s/(amp.+?;)/&de_tex_er($1)/ge;
 181 #    s/(&.+?;)/&de_tex_er($1)/ge;
 182     s/([$asian])\s+/$1/go unless($out_cs eq 'UniKS');
 183     s/\s+([$asian])/$1/go unless($out_cs eq 'UniKS');
 184     s/([$asian])\s*([^$asian$space])/$1 $2/go;
 185     s/([^$asian$idc])\s*([$asian])/$1 $2/go;
 186     s/\-\-\-/pack("U",0x2014)/geo;# EM DASH
 187     s/\-\-/pack("U",0x2013)/geo;# EN DASH
 188 #    s/\`\`/pack("U",0x201f)/geo;# DOUBLE HIGH-REVERSED-9 QUOTATION MARK
 189     s/\`\`/pack("U",0x201c)/geo;# _TeX compatible_
 190 #    s/\`/pack("U",0x201b)/geo;# SINGLE HIGH-REVERSED-9 QUOTATION MARK
 191     s/\`/pack("U",0x2018)/geo;# _TeX compatible_
 192     s/\'\'/pack("U",0x201d)/geo;# RIGHT DOUBLE QUOTATION MARK
 193     s/\'/pack("U",0x2019)/geo;# RIGHT DOUBLE QUOTATION MARK
 194     @chars=split(//);
 195   CHAR:
 196     for($i=0;$i<=$#chars;$i++){
 197         $char=$chars[$i];
 198         $char_id=unpack("U",$char);
 199
 200         if($char_id<=0x20){
 201             # add quarter space between asian and non-asian.
 202             if($i>0 and $i<$#chars){
 203                 if(($chars[$i-1]=~/[$asian]/
 204                     and $chars[$i+1]=~/[^$asian]/
 205                       and $chars[$i+1]=~/[$strictly_forbidden_before$slightly_forbidden_before]/)
 206                    or($chars[$i-1]=~/[^$asian]/
 207                       and $chars[$i+1]=~/[$asian]/
 208                       and $chars[$i-1]=~/[$strictly_forbidden_after$slightly_forbidden_after]/)
 209                    ){
 210                     print '';
 211                 }elsif(($chars[$i-1]=~/[$asian]/
 212                         and $chars[$i+1]=~/[^$asian]/)){
 213                     print '\unskip\kern.25ex';
 214                 }elsif($chars[$i-1]=~/[^$asian]/
 215                        and $chars[$i+1]=~/[$asian]/){
 216                     print '\kern.25ex';
 217                 }else{
 218                     print ' ';
 219                 }
 220             }else{
 221                 print ' ';
 222             }
 223             next CHAR;
 224         }elsif(($char_id>0x20 and $char_id<=0x021f)
 225                # Basic Latin
 226                # Latin-1 Supplement
 227                # Latin Extended-A
 228                # Latin Extended-B (not all)
 229                or($char_id>=0x0250 and $char_id<=0x02af)
 230                # IPA Extensions
 231                or($char_id>=0x0300 and $char_id<=0x033f)
 232                or($char_id>=0x0360 and $char_id<=0x036f)
 233                # Combining Diacritical Marks
 234                or($char_id>=0x1e00 and $char_id<=0x1eff)
 235                # Latin Extended Additional
 236                or($char_id>=0x0370 and $char_id<=0x03ff)
 237                # Greek and Coptic
 238                or($char_id>=0x0400 and $char_id<=0x04ff)
 239                # Cyrillic
 240                or($char_id>=0x0530 and $char_id<=0x058f)
 241                # Armenian
 242                or($char_id>=0x2010 and $char_id<=0x2046)
 243                # General Punctuation (partial)
 244                ){
 245             print &latin_parse();
 246             next CHAR;
 247         }elsif($char_id>=0x2ff0 and $char_id<=0x2fff){
 248             # Ideographic Description Characters
 249             print &ids_parse();
 250             next CHAR;
 251         }else{
 252             if(($out_char=&get_output_char($char,$out_cs))){
 253                 print $out_char,&add_break($i);
 254             }else{
 255                 if($opt_allow_unify){
 256                     @chars_unified=&get_chars_unified($char);
 257                     if(@chars_unified>0){
 258                         foreach $char_unified (@chars_unified){
 259                             if(($out_char
 260                                 =&get_output_char($char_unified,$out_cs))){
 261                                 print $out_char,&add_break($i);
 262                                 next CHAR;
 263                             }
 264                         }
 265                     }
 266                 }
 267                 if($opt_use_kage_for_Ext_B){
 268                     if($char_id >= 0x20000 && $char_id <=0x2a6df){
 269                         # CJK Unified Ideographs Extension B
 270                         if(not defined($ids{$char}) and $ids{$char}[1]>=0){
 271                             $ids{$char}[0]=$font_start;
 272                             $ids{$char}[1]=$ids_start;
 273                             $ids_start++;
 274                             if($ids_start>255){
 275                                 $ids_start=0;
 276                                 $font_start++;
 277                             }
 278                         }
 279                         print "{\\fontencoding{OT1}\\fontfamily{",
 280                             sprintf("chise%03d",$ids{$char}[0]) ,
 281                                 "}\\selectfont\\char$ids{$char}[1]}",&add_break($i);
 282                         next CHAR;
 283                     }
 284                 }
 285                 if($ids=&get_ids($char)){
 286                     print &get_macro_for_ids($ids),&add_break($i);
 287                 }else{
 288                     print '\rule{1ex}{1ex}',&add_break($i);
 289                 }
 290             }
 291         }
 292     }
 293 }
 294
 295 print IDSDATA 'use utf8;',"\n";
 296 foreach $ids (keys %ids){
 297     print IDSDATA '$ids{\'',$ids,'\'}='
 298     ,'[',join ",",@{$ids{$ids}},"];\n";
 299 }
 300 print IDSDATA '$font_start=',$font_start,";\n";
 301 print IDSDATA '$ids_start=',$ids_start,";\n";
 302 print IDSDATA "1;";
 303 flock(IDSDATA,LOCK_UN);
 304
 305 exit 0;
 306
 307 sub de_tex_er{
 308     my($er)=@_;
 309     my($prefix,$suffix);
 310     my($output_char,$atr,$value);
 311     $er=~/^(amp)(.*)(;)$/
 312         and $prefix=$1,$er=$2,$suffix=$3;
 313     $prefix or $prefix="",$suffix or $suffix="";
 314     if($er=~/^U[\+|\-]([a-fA-F\d]+)/){
 315         $output_char=pack("U",hex($1));
 316     }elsif($er=~/^(?:I\-)?($er_prefix_re)\-?([0-9a-fA-F]+)$/){
 317         ($atr,$value)=($1,$2);
 318         unless($er_alias{$atr}=~/daikanwa|gt/){
 319             $value=hex($value);
 320         }
 321         ($output_char)=&get_chars_matching($er_alias{$atr},$value);
 322     }
 323     if($output_char){
 324         return $output_char;
 325     }else{
 326         return $prefix.$er.$suffix;
 327     }
 328 }
 329
 330 sub add_break{
 331     my($i)=@_;
 332
 333     if($i<($#chars-1)){
 334         if(($chars[$i+1]=~m/[$strictly_forbidden_before]/o)
 335            and($chars[$i+2]=~m/[$strictly_forbidden_before]/o)){
 336             return "\\CJKunbreakablekernone{}";
 337         }elsif($opt_protrude){
 338             if(($chars[$i+1]=~m/[$strictly_forbidden_before]/o)
 339                and($chars[$i+2]=~m/[^$strictly_forbidden_before]/o)){
 340                 return "\\CJKunbreakablekernone \\CJKprotrude{}";
 341             }
 342         }
 343     }
 344     if(($i<$#chars)
 345        and($chars[$i+1]eq" ")){
 346        # preserve space.
 347         return "";
 348     }
 349     if(($i<$#chars)
 350        and($chars[$i+1]=~m/[$strictly_forbidden_before]/o)){
 351         return "\\CJKunbreakablekernone{}";
 352     }
 353     if($chars[$i]=~m/[$strictly_forbidden_after]/o){
 354         return "\\CJKunbreakablekernone{}";
 355     }
 356     if(($i<$#chars)
 357        and($chars[$i+1]=~m/[$forbidden_before]/o)){
 358         return "\\CJKunbreakablekerntwo{}";
 359
 360     }
 361     if($chars[$i]=~m/[$forbidden_after]/o){
 362         return "\\CJKunbreakablekerntwo{}";
 363     }
 364     if(($i<$#chars)
 365        and($chars[$i+1]=~m/[$slightly_forbidden_before]/o)){
 366         return "\\CJKunbreakablekernthree{}";
 367     }
 368     if($chars[$i]=~m/[$slightly_forbidden_after]/o){
 369         return "\\CJKunbreakablekernthree{}";
 370     }
 371     if($chars[$i]=~m/[$asian]/o){
 372         return "\\CJKbreakablekern{}";
 373     }
 374     if(($i<$#chars)and($chars[$i+1]=~m/[$asian]/o)){
 375         return "\\CJKbreakablekern{}";
 376     }
 377 }
 378
 379 sub latin_parse{
 380     # arguments: none
 381     # return: string for output with TeX macro.
 382     my($char_id);
 383     my $out_str="";
 384     while($i<=$#chars){
 385         $char_id=unpack("U",$chars[$i]);
 386         if($chars[$i]=~m/($tex_meta_re)/o){
 387             $out_str.=$tex_meta{$chars[$i]};
 388         }elsif(($char_id>0x20 and $char_id<=0x021f)
 389            or($char_id>=0x0250 and $char_id<=0x02af)# IPA Extensions
 390            or($char_id>=0x0300 and $char_id<=0x033f)# Combining Diacritical Marks
 391            or($char_id>=0x0360 and $char_id<=0x036f)
 392            or($char_id>=0x0370 and $char_id<=0x03ff)# Greek and Coptic
 393            or($char_id>=0x0400 and $char_id<=0x04ff)# Cyrillic
 394            or($char_id>=0x0530 and $char_id<=0x058f)# Armenian
 395            ){
 396             $out_str.=$chars[$i];
 397         }elsif($char_id>=0x1e00 and $char_id<=0x1eff){
 398             # Latin Extended Additional
 399             # 0x1e00 -> 0x0600, etc.
 400             $out_str.=pack("U",$char_id-0x1800);
 401         }elsif($char_id>=0x2010 and $char_id<=0x2046){
 402             # General Punctuation (partial)
 403             $out_str.=pack("U",$char_id-0x1000);
 404         }else{
 405             $i--;
 406             last;
 407         }
 408         $i++;
 409     }
 410     return '{\fontencoding{OT1}\fontfamily{omlgc}\selectfont '.$out_str.'}';###UT1?
 411 }
 412
 413 sub ids_parse{
 414     # arguments: none
 415     # return: character for output,
 416     #          TeX macro for ids,
 417     #          or GETA character if ids is invalid.
 418     my($ids,$ids_argc)=&ids_rest("",0,$chars[$i]);
 419
 420     while($ids_argc>0){
 421         # We are in IDS.
 422         $i++;
 423         if($i>$#chars){
 424             print STDERR "IDS parse error: $ids\n";
 425             return $geta;
 426         }
 427
 428         ($ids,$ids_argc)=&ids_rest($ids,$ids_argc,$chars[$i]);
 429         if($ids_argc==0){
 430             if(($char=&get_char_for_ids($ids))
 431                and($out_char=&get_output_char($char,$out_cs))){
 432                 return $out_char;
 433             }else{
 434                 return &get_macro_for_ids($ids);
 435             }
 436         }
 437     }
 438 }
 439
 440 sub ids_rest{
 441     # arguments: <ids>, <rest number of arguments for ids>, <character>
 442     # return: ids and rest number of arguments for ids.
 443     my($ids,$ids_argc,$char)=@_;
 444     my($argc);
 445     $argc=&ids_argc($char);
 446     if($argc){
 447         $ids_argc+=$ids_argc==0?$argc:$argc-1;
 448     }else{
 449         $ids_argc--;
 450     }
 451     $ids.=$char;
 452     return ($ids,$ids_argc);
 453 }
 454
 455 sub get_macro_for_ids{
 456     # argument: <ids>
 457     # return: TeX macro for ids
 458     #          or GETA character if ids is invalid for KAGE.
 459     my($ids)=@_;
 460     # $ids=&normalize_ids($ids,"UniJIS");
 461     return $geta if(($ids!~/[$idc]/)
 462                     or($ids=~/[\x{10000}-]/));
 463                     #irregular for KAGE.
 464     if(not defined($ids{$ids})){
 465         $ids{$ids}[0]=$font_start;
 466         $ids{$ids}[1]=$ids_start;
 467         $ids_start++;
 468     }
 469     if($ids_start>255){
 470         $ids_start=0;
 471         $font_start++;
 472     }
 473     return "{\\fontencoding{OT1}\\fontfamily{"
 474         .sprintf("chise%03d",$ids{$ids}[0])
 475         ."}\\selectfont\\char$ids{$ids}[1]}";
 476 }
 477
 478 sub normalize_ids{
 479     # argument: <ids>, <output coding system>
 480     # return: ids or GETA character if ids is invalid for KAGE.
 481     my($ids,$out_cs)=@_;
 482     $out_cs=~s/Uni(.+)/'ucs@'.lc($1)/e;
 483
 484     my $output_ids="";
 485     my($char,$output_char_id);
 486     while($ids=~m/(.)/g){
 487         $char=$1;
 488         if($char=~/[$idc]/){
 489             $output_ids.=$char;
 490         }elsif($output_char_id=&get_char_attribute($char,"=$out_cs")
 491            or $output_char_id=&get_char_attribute($char,"=ucs")
 492            or $output_char_id=&get_char_attribute($char,"=>$out_cs")
 493            or $output_char_id=&get_char_attribute($char,"=>ucs")
 494            or $output_char_id=&get_char_attribute($char,"=>ucs*")
 495               ){
 496             $output_ids.=pack("U",$output_char_id);
 497         }else{
 498             return $geta;
 499         }
 500     }
 501     return $output_ids;
 502 }
 503
 504 sub get_output_char{
 505     # argument: <char>
 506     # return: character in output coding system or TeX macro or undef.
 507     my($char,$out_cs)=@_;
 508     my($out_char_id,$suffix);
 509     my($gt,$hzk,$cdp);
 510
 511     foreach $out_cs (@{$order{$out_cs}}){
 512         if($out_cs eq 'UniJIS'
 513            and &get_char_attribute($char,"vnd-adobe-cid-unijis-utf16-h")){
 514             if($out_char_id=&get_char_attribute($char,'=ucs@jis')
 515                or $out_char_id=&get_char_attribute($char,'=ucs')
 516                or $out_char_id=&get_char_attribute($char,'=>ucs@jis')
 517                or $out_char_id=&get_char_attribute($char,'=>ucs')
 518                or $out_char_id=&get_char_attribute($char,'=>ucs*')
 519                ){
 520                 return "{\\selectjisfont\\char$out_char_id}";
 521             }
 522         }elsif($out_cs eq 'UniGB'
 523                and &get_char_attribute($char,"vnd-adobe-cid-unigb-ucs2-h")){
 524             if($out_char_id=&get_char_attribute($char,'=ucs@gb')
 525                or $out_char_id=&get_char_attribute($char,'=ucs')
 526                or $out_char_id=&get_char_attribute($char,'=>ucs@gb')
 527                or $out_char_id=&get_char_attribute($char,'=>ucs')
 528                or $out_char_id=&get_char_attribute($char,'=>ucs*')
 529                ){
 530                 return "{\\selectgbsfont\\char$out_char_id}";
 531             }
 532         }elsif($out_cs eq 'UniCNS'
 533                and &get_char_attribute($char,"vnd-adobe-cid-unicns-ucs2-h")){
 534             if($out_char_id=&get_char_attribute($char,'=ucs@cns')
 535                or $out_char_id=&get_char_attribute($char,'=ucs')
 536                or $out_char_id=&get_char_attribute($char,'=>ucs@cns')
 537                or $out_char_id=&get_char_attribute($char,'=>ucs')
 538                or $out_char_id=&get_char_attribute($char,'=>ucs*')
 539                ){
 540                 return "{\\selectcnsfont\\char$out_char_id}";
 541             }
 542         }elsif($out_cs eq 'UniKS'
 543                and &get_char_attribute($char,"vnd-adobe-cid-uniks-ucs2-h")){
 544             if($out_char_id=&get_char_attribute($char,'=ucs@ks')
 545                or $out_char_id=&get_char_attribute($char,'=ucs')
 546                or $out_char_id=&get_char_attribute($char,'=>ucs@ks')
 547                or $out_char_id=&get_char_attribute($char,'=>ucs')
 548                or $out_char_id=&get_char_attribute($char,'=>ucs*')
 549                ){
 550                 return "{\\selectksxfont\\char$out_char_id}";
 551             }
 552         }elsif($out_cs eq 'GT'){
 553             return $gt if($gt=&get_macro_for_GT($char));
 554         }elsif($out_cs eq 'HZK'){
 555             return $hzk if($hzk=&get_macro_for_HZK($char));
 556         }elsif($out_cs eq 'CDP'){
 557             return $cdp if($cdp=&get_macro_for_CDP($char));
 558         }
 559     }
 560     return undef;
 561 }
 562
 563 sub get_ids{
 564     # argument: <character>
 565     # return: ids
 566     my($char)=@_;
 567     my $ids="";
 568     $ids=&get_char_attribute($char,"ids-aggregated")
 569         or $ids=&get_char_attribute($char,"ids");
 570 #    $ids=&get_char_attribute($char,"ids-decomposed")
 571 #       or $ids=&get_char_attribute($char,"ids");
 572 #         or $ids=&get_char_attribute($char,"ideographic-structure");
 573 #    $ids=~s/[? ()]//g;
 574     return $ids;
 575 }
 576
 577 sub get_char_for_ids{
 578     # argument: <ideographic description sequence>
 579     # return: char or undef.
 580     my($ids)=@_;
 581     my($output_char);
 582
 583     if(($output_char)=&get_chars_matching("ids",$ids)){
 584         return $output_char;
 585     }else{
 586         return undef;
 587     }
 588 }
 589
 590 sub get_char_in_mcs{
 591     # argument: <char>, <input coding system>
 592     # return:   char in ucs@mcs.
 593     my($char,$in_cs)=@_;
 594     my($output_char);
 595
 596     return $char if($in_cs eq 'ucs@mcs');
 597
 598     if(($output_char)=&get_chars_matching("=$in_cs",unpack("U",$char))){
 599         return $output_char;
 600     }else{
 601         return $char;
 602     }
 603 }
 604
 605 sub get_chars_unified{
 606     my($char)=@_;
 607     my($chars,$ucs,$char_ucs);
 608     my(@chars);
 609
 610     if($chars=&get_char_attribute($char,'->ucs-unified')){
 611         $chars=~s/^\((.*)\)$/$1/;
 612         return (split(/\s*\?/,$chars));
 613     }elsif($ucs=&get_char_attribute($char,'=>ucs*')
 614           or $ucs=&get_char_attribute($char,'=>ucs')){
 615         $char_ucs=pack("U",$ucs);
 616         if($chars=&get_char_attribute($char_ucs,'->ucs-unified')){
 617             $chars=~s/^\((.*)\)$/$1/;
 618             @chars=grep {not /^$char$/}
 619                 (split(/\s*\?/,$chars));
 620             push(@chars,$char_ucs);
 621             return @chars;
 622         }
 623     }
 624 }
 625
 626 sub get_macro_for_GT{
 627     # argument: <char>
 628     # return: TeX macro for GT fonts or undef.
 629     my($char)=@_;
 630     my($gt,$GT);
 631     foreach (@GT){
 632         if($gt=&get_char_attribute($char,$_)){
 633             m/gt\-pj\-(\d+)/ and $GT=$1;
 634             last;
 635         }
 636     }
 637     if($gt){
 638         return "{\\fontencoding{OT1}\\fontfamily{"
 639             .sprintf("gt%02d",$GT)
 640             ."}\\selectfont\\char".($gt|0x8080)."}";
 641     }else{
 642         return undef;
 643     }
 644 }
 645
 646 sub get_macro_for_HZK{
 647     # argument: <char>
 648     # return: TeX macro for Hanziku fonts or undef.
 649     my($char)=@_;
 650     my($hzk,$HZK);
 651     foreach (@HZK){
 652         if($hzk=&get_char_attribute($char,$_)){
 653             m/hanziku\-(\d+)/ and $HZK=$1;
 654             last;
 655         }
 656     }
 657     if($hzk){
 658         return "{\\fontencoding{OT1}\\fontfamily{"
 659             .sprintf("hzk%02d",$HZK)
 660             ."}\\selectfont\\char".$hzk."}";
 661     }else{
 662         return undef;
 663     }
 664 }
 665
 666 sub get_macro_for_CDP{
 667     # argument: <char>
 668     # return: TeX macro for CDP fonts or undef.
 669     my($char)=@_;
 670     my($cdp,$ucs);
 671     foreach (@CDP){
 672         if($cdp=&get_char_attribute($char,$_)){
 673             last;
 674         }
 675     }
 676     if($cdp){
 677         $ucs=&get_char_attribute(&get_chars_matching("=big5-pua",$cdp),"=ucs");
 678         if($ucs){
 679             return "{\\fontencoding{OT1}\\fontfamily{cdp}\\selectfont\\char$ucs}";
 680         }else{
 681             print STDERR "This should not happen.\n";
 682             print STDERR "ucs code point of CDP: $cdp not found.\n";
 683         }
 684     }else{
 685         return undef;
 686     }
 687 }