1 #!/usr/bin/perl -w -CSD
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
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
22 use Chise_utils ':all';
25 my $omegadb_path="/usr/local/lib/chise/omega";
29 #$opt_order{'UniMulti'}='jcgk';
30 $opt_order{'UniMulti'}='jGcgkHC';
31 $opt_order{'UniCNS'}='c';
32 $opt_order{'UniGB'}='g';
33 $opt_order{'UniJIS'}='j';
34 $opt_order{'UniKS'}='k';
36 $opt_allow_unify=1; # 1=true, 0=false.
37 $opt_protrude=0;# 1=true, 0=false.
39 # currently does not work.
40 $opt_use_kage_for_Ext_B=0;# 1=true, 0=false.
44 my $strictly_forbidden_after = '「【『[(〈“‘‘(〔{《{\[\(\x{3016}{「';
45 # \x{3016} | # white 【
47 my $forbidden_after = "\x{0000}";
50 my $slightly_forbidden_after = '¥¥$$〒♯##¢¢££@@§';
52 # $strictly_forbidden_before
53 # All these characters are allowed to protrude
55 my $strictly_forbidden_before=
56 '!,.:;?、。!,.:;?。\)#}’”〉》」』】〕\x{3017})]}}」\]';
57 ### \x{3017} | # white 】
60 = 'ー々ぁぃぅぇぉゃゅょっゎァィゥェォャュョッヮヵヶ';
62 my $slightly_forbidden_before
63 = '\x{000a}\#\-‐−‰′″℃゛゜ゝゞヽヾ"%-゙゚';
65 my $asian = '\x{1100}-\x{11FF}\x{2E80}-\x{D7AF}\x{F900}-\x{FAFF}\x{FE30}-\x{FE4F}\x{FF00}-\x{FFFFFF}';
67 my $space = '\x{0020}\x{0009}\x{000A}\x{000C}\x{000D}';
69 my %tex_meta=('#'=>'\#',
70 '$'=>'\\textdollar{}',
73 '{'=>'\\textbraceleft{}',
74 '}'=>'\\textbraceright{}',
75 '\\'=>'\\textbackslash{}',
76 '_'=>'\\textunderscore',
79 my $tex_meta_re=join('|',map {quotemeta($_)} keys %tex_meta);
81 &GetOptions("in=s"=>\$opt_in_cs,
82 "out=s"=>\$opt_out_cs,
86 Usage: $0 -i <input coding system> -o <cmap encoding>
88 Utf8mcs, Utf8cns, Utf8gb, Utf8jis, Utf8ks
90 UniCNS, UniGB, UniJIS, UniKS, UniMulti
93 if($opt_in_cs or $opt_out_cs){
97 ($in_cs,$out_cs)=($0=~/(Utf8.+)To(\w+)/);
101 # Utf8mcs,Utf8cns,Utf8gb,Utf8jis,Utf8ks,
103 # UniCNS,UniGB,UniJIS,UniKS,UniMulti
105 $in_cs=~s/Utf8/ucs\@/;
108 or not defined($in_cs)
109 or not defined($out_cs)){
114 $omegadb_path=~s!/$!!;
116 $idsdata_file="$omegadb_path/idsdata.pl";
120 if(-e $idsdata_file){
121 open(IDSDATA,"+<:utf8",$idsdata_file) or die;
122 flock(IDSDATA,LOCK_EX);
128 # require $idsdata_file;
130 open(IDSDATA,">:utf8",$idsdata_file) or die;
131 flock(IDSDATA,LOCK_EX);
138 #$geta=pack("U",0x3013);
139 $geta=pack("U",0xfffd);
142 "=gt-pj-1","=gt-pj-2","=gt-pj-3","=gt-pj-4","=gt-pj-5",
143 "=gt-pj-6","=gt-pj-7","=gt-pj-8","=gt-pj-9","=gt-pj-10",
145 #,"=gt-pj-k1","=gt-pj-k2"
147 @HZK=("=hanziku-1","=hanziku-2","=hanziku-3","=hanziku-4",
148 "=hanziku-5","=hanziku-6","=hanziku-7","=hanziku-8",
149 "=hanziku-9","=hanziku-10","=hanziku-11","=hanziku-12");
152 %order_map=('c'=>'UniCNS',
161 foreach $out_cs ('UniCNS','UniGB','UniJIS','UniKS','UniMulti'){
162 if(defined($opt_order{$out_cs})){
163 if($opt_order{$out_cs}=~/^[cgjkGHC]+$/){
164 @{$order{$out_cs}}=map {$order_map{$_}}
165 (split(//,$opt_order{$out_cs}));
167 print STDERR "Invalid order for $out_cs!\n";
175 if($in_cs ne 'ucs@mcs'){
176 s/(.)/&get_char_in_mcs($1,$in_cs)/ge;
178 s/(amp.+?;)/&de_tex_er($1)/ge;
179 # s/(&.+?;)/&de_tex_er($1)/ge;
182 for($i=0;$i<=$#chars;$i++){
184 $char_id=unpack("U",$char);
189 }elsif($char=~m/($tex_meta_re)/o){
192 }elsif($char_id>0x20 and $char_id<=0x02af){
198 print &latin_parse();
200 }elsif($char_id>=0x2ff0 and $char_id<=0x2fff){
201 # Ideographic Description Characters
205 if(($out_char=&get_output_char($char,$out_cs))){
206 print $out_char,&add_break($i);
208 if($opt_allow_unify){
209 @chars_unified=&get_chars_unified($char);
210 if(@chars_unified>0){
211 foreach $char_unified (@chars_unified){
213 =&get_output_char($char_unified,$out_cs))){
214 print $out_char,&add_break($i);
220 if($opt_use_kage_for_Ext_B){
221 if($char_id >= 0x20000 && $char_id <=0x2a6df){
222 # CJK Unified Ideographs Extension B
223 if(not defined($ids{$char}) and $ids{$char}[1]>=0){
224 $ids{$char}[0]=$font_start;
225 $ids{$char}[1]=$ids_start;
232 print "{\\fontencoding{OT1}\\fontfamily{" .
233 sprintf("chise%03d",$ids{$char}[0]) .
234 "}\\selectfont\\char$ids{$char}[1]}",&add_break($i);
238 if($ids=&get_ids($char)){
239 print &get_macro_for_ids($ids),&add_break($i);
241 print '\rule{1ex}{1ex}',&add_break($i);
248 print IDSDATA 'use utf8;',"\n";
249 foreach $ids (keys %ids){
250 print IDSDATA '$ids{\'',$ids,'\'}='
251 ,'[',join ",",@{$ids{$ids}},"];\n";
253 print IDSDATA '$font_start=',$font_start,";\n";
254 print IDSDATA '$ids_start=',$ids_start,";\n";
256 flock(IDSDATA,LOCK_UN);
263 my($output_char,$atr,$value);
264 $er=~/^(amp)(.*)(;)$/
265 and $prefix=$1,$er=$2,$suffix=$3;
266 $prefix or $prefix="",$suffix or $suffix="";
267 if($er=~/^U[\+|\-]([a-fA-F\d]+)/){
268 $output_char=pack("U",hex($1));
269 }elsif($er=~/^(?:I\-)?($er_prefix_re)\-?([0-9a-fA-F]+)$/){
270 ($atr,$value)=($1,$2);
271 unless($er_alias{$atr}=~/daikanwa|gt/){
274 ($output_char)=&get_chars_matching($er_alias{$atr},$value);
279 return $prefix.$er.$suffix;
287 if(($chars[$i+1]=~m/[$strictly_forbidden_before]/o)
288 and($chars[$i+2]=~m/[$strictly_forbidden_before]/o)){
289 return "\\CJKunbreakablekernone ";
290 }elsif($opt_protrude){
291 if(($chars[$i+1]=~m/[$strictly_forbidden_before]/o)
292 and($chars[$i+2]=~m/[^$strictly_forbidden_before]/o)){
293 return "\\CJKunbreakablekernone \\CJKprotrude ";
298 and($chars[$i+1]=~m/[$strictly_forbidden_before]/o)){
299 return "\\CJKunbreakablekernone ";
301 if($chars[$i]=~m/[$strictly_forbidden_after]/o){
302 return "\\CJKunbreakablekernone ";
305 and($chars[$i+1]=~m/[$forbidden_before]/o)){
306 return "\\CJKunbreakablekerntwo ";
309 if($chars[$i]=~m/[$forbidden_after]/o){
310 return "\\CJKunbreakablekerntwo ";
313 and($chars[$i+1]=~m/[$slightly_forbidden_before]/o)){
314 return "\\CJKunbreakablekernthree ";
316 if($chars[$i]=~m/[$slightly_forbidden_after]/o){
317 return "\\CJKunbreakablekernthree ";
319 if($chars[$i]=~m/[$asian]/o){
320 return "\\CJKbreakablekern ";
322 if(($i<$#chars)and($chars[$i+1]=~m/[$asian]/o)){
323 return "\\CJKbreakablekern ";
329 # return: string for output with TeX macro.
331 my $out_str=$chars[$i];
334 $char_id=unpack("U",$chars[$i]);
335 if($char_id>0x20 and $char_id<=0x02af){
336 $out_str.=$chars[$i];
343 return '{\fontencoding{UT1}\fontfamily{omlgc}\selectfont '.$out_str.'}';
348 # return: character for output,
350 # or GETA character if ids is invalid.
351 my($ids,$ids_argc)=&ids_rest("",0,$chars[$i]);
357 print STDERR "IDS parse error: $ids\n";
361 ($ids,$ids_argc)=&ids_rest($ids,$ids_argc,$chars[$i]);
363 if(($char=&get_char_for_ids($ids))
364 and($out_char=&get_output_char($char,$out_cs))){
367 return &get_macro_for_ids($ids);
374 # arguments: <ids>, <rest number of arguments for ids>, <character>
375 # return: ids and rest number of arguments for ids.
376 my($ids,$ids_argc,$char)=@_;
378 $argc=&ids_argc($char);
380 $ids_argc+=$ids_argc==0?$argc:$argc-1;
385 return ($ids,$ids_argc);
388 sub get_macro_for_ids{
390 # return: TeX macro for ids
391 # or GETA character if ids is invalid for KAGE.
393 # $ids=&normalize_ids($ids,"UniJIS");
394 return $geta if(($ids!~/[$idc]/)
395 or($ids=~/[\x{10000}-]/));
397 if(not defined($ids{$ids})){
398 $ids{$ids}[0]=$font_start;
399 $ids{$ids}[1]=$ids_start;
406 return "{\\fontencoding{OT1}\\fontfamily{"
407 .sprintf("chise%03d",$ids{$ids}[0])
408 ."}\\selectfont\\char$ids{$ids}[1]}";
412 # argument: <ids>, <output coding system>
413 # return: ids or GETA character if ids is invalid for KAGE.
415 $out_cs=~s/Uni(.+)/'ucs@'.lc($1)/e;
418 my($char,$output_char_id);
419 while($ids=~m/(.)/g){
423 }elsif($output_char_id=&get_char_attribute($char,"=$out_cs")
424 or $output_char_id=&get_char_attribute($char,"=ucs")
425 or $output_char_id=&get_char_attribute($char,"=>$out_cs")
426 or $output_char_id=&get_char_attribute($char,"=>ucs")
427 or $output_char_id=&get_char_attribute($char,"=>ucs*")
429 $output_ids.=pack("U",$output_char_id);
439 # return: character in output coding system or TeX macro or undef.
440 my($char,$out_cs)=@_;
441 my($out_char_id,$suffix);
444 foreach $out_cs (@{$order{$out_cs}}){
445 if($out_cs eq 'UniJIS'
446 and &get_char_attribute($char,"vnd-adobe-cid-unijis-utf16-h")){
447 if($out_char_id=&get_char_attribute($char,'=ucs@jis')
448 or $out_char_id=&get_char_attribute($char,'=ucs')
449 or $out_char_id=&get_char_attribute($char,'=>ucs@jis')
450 or $out_char_id=&get_char_attribute($char,'=>ucs')
451 or $out_char_id=&get_char_attribute($char,'=>ucs*')
453 return '{\selectjisfont\char'.$out_char_id.'}';
455 }elsif($out_cs eq 'UniGB'
456 and &get_char_attribute($char,"vnd-adobe-cid-unigb-ucs2-h")){
457 if($out_char_id=&get_char_attribute($char,'=ucs@gb')
458 or $out_char_id=&get_char_attribute($char,'=ucs')
459 or $out_char_id=&get_char_attribute($char,'=>ucs@gb')
460 or $out_char_id=&get_char_attribute($char,'=>ucs')
461 or $out_char_id=&get_char_attribute($char,'=>ucs*')
463 return '{\selectgbsfont\char'.$out_char_id.'}';
465 }elsif($out_cs eq 'UniCNS'
466 and &get_char_attribute($char,"vnd-adobe-cid-unicns-ucs2-h")){
467 if($out_char_id=&get_char_attribute($char,'=ucs@cns')
468 or $out_char_id=&get_char_attribute($char,'=ucs')
469 or $out_char_id=&get_char_attribute($char,'=>ucs@cns')
470 or $out_char_id=&get_char_attribute($char,'=>ucs')
471 or $out_char_id=&get_char_attribute($char,'=>ucs*')
473 return '{\selectcnsfont\char'.$out_char_id.'}';
475 }elsif($out_cs eq 'UniKS'
476 and &get_char_attribute($char,"vnd-adobe-cid-uniks-ucs2-h")){
477 if($out_char_id=&get_char_attribute($char,'=ucs@ks')
478 or $out_char_id=&get_char_attribute($char,'=ucs')
479 or $out_char_id=&get_char_attribute($char,'=>ucs@ks')
480 or $out_char_id=&get_char_attribute($char,'=>ucs')
481 or $out_char_id=&get_char_attribute($char,'=>ucs*')
483 return '{\selectksxfont\char'.$out_char_id.'}';
485 }elsif($out_cs eq 'GT'){
486 return $gt if($gt=&get_macro_for_GT($char));
487 }elsif($out_cs eq 'HZK'){
488 return $hzk if($hzk=&get_macro_for_HZK($char));
489 }elsif($out_cs eq 'CDP'){
490 return $cdp if($cdp=&get_macro_for_CDP($char));
497 # argument: <character>
501 # $ids=&get_char_attribute($char,"ids-aggregated")
502 # or $ids=&get_char_attribute($char,"ids");
503 $ids=&get_char_attribute($char,"ids-decomposed")
504 or $ids=&get_char_attribute($char,"ids");
505 # or $ids=&get_char_attribute($char,"ideographic-structure");
510 sub get_char_for_ids{
511 # argument: <ideographic description sequence>
512 # return: char or undef.
516 if(($output_char)=&get_chars_matching("ids",$ids)){
524 # argument: <char>, <input coding system>
525 # return: char in ucs@mcs.
529 return $char if($in_cs eq 'ucs@mcs');
531 if(($output_char)=&get_chars_matching("=$in_cs",unpack("U",$char))){
538 sub get_chars_unified{
540 my($chars,$ucs,$char_ucs);
543 if($chars=&get_char_attribute($char,'->ucs-unified')){
544 $chars=~s/^\((.*)\)$/$1/;
545 return (split(/\s*\?/,$chars));
546 }elsif($ucs=&get_char_attribute($char,'=>ucs*')
547 or $ucs=&get_char_attribute($char,'=>ucs')){
548 $char_ucs=pack("U",$ucs);
549 if($chars=&get_char_attribute($char_ucs,'->ucs-unified')){
550 $chars=~s/^\((.*)\)$/$1/;
551 @chars=grep {not /^$char$/}
552 (split(/\s*\?/,$chars));
553 push(@chars,$char_ucs);
559 sub get_macro_for_GT{
561 # return: TeX macro for GT fonts or undef.
565 if($gt=&get_char_attribute($char,$_)){
566 m/gt\-pj\-(\d+)/ and $GT=$1;
571 return "{\\fontencoding{OT1}\\fontfamily{"
572 .sprintf("gt%02d",$GT)
573 ."}\\selectfont\\char".($gt|0x8080)."}";
579 sub get_macro_for_HZK{
581 # return: TeX macro for Hanziku fonts or undef.
585 if($hzk=&get_char_attribute($char,$_)){
586 m/hanziku\-(\d+)/ and $HZK=$1;
591 return "{\\fontencoding{OT1}\\fontfamily{".sprintf("hzk%02d",$HZK)."}\\selectfont\\char".$hzk."}";
597 sub get_macro_for_CDP{
599 # return: TeX macro for CDP fonts or undef.
603 if($cdp=&get_char_attribute($char,$_)){
608 $ucs=&get_char_attribute(&get_chars_matching("=big5-pua",$cdp),"=ucs");
610 return "{\\fontencoding{OT1}\\fontfamily{cdp}\\selectfont\\char"
614 print STDERR "This should not happen.\n";
615 print STDERR "ucs code point of CDP: $cdp not found.\n";