1 #!/usr/bin/perl -w -CSD
6 use vars qw($omegadb_path
7 $opt_protrude $opt_allow_unify
8 %opt_order %order %order_map
11 $in_cs $out_cs $i @chars
12 $char $char_id $out_char
13 $char_unified @chars_unified
14 $ids $ids_argc %ids $idsdb $geta
15 $idsdata_file $ids_start $font_start
21 use Chise_utils ':all';
24 my $omegadb_path="/usr/local/lib/chise/omega";
28 #$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';
35 $opt_allow_unify=1; # 1=true, 0=false.
36 $opt_protrude=0;# 1=true, 0=false.
40 my $strictly_forbidden_after = '「【『[(〈“‘‘(〔{《{\[\(\x{3016}{「';
41 # \x{3016} | # white 【
43 my $forbidden_after = "\x{0000}";
46 my $slightly_forbidden_after = '¥¥$$〒♯##¢¢££@@§';
48 # $strictly_forbidden_before
49 # All these characters are allowed to protrude
51 my $strictly_forbidden_before=
52 '!,.:;?、。!,.:;?。\)#}’”〉》」』】〕\x{3017})]}}」\]';
53 ### \x{3017} | # white 】
56 = '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 my $slightly_forbidden_before
59 = '\x{000a}\#\-‐−‰′″℃゛゜ゝゞヽヾ"%-゙゚';
61 my $asian = '\x{1100}-\x{11FF}\x{2E80}-\x{D7AF}\x{F900}-\x{FAFF}\x{FE30}-\x{FE4F}\x{FF00}-\x{FFFFFF}';
63 my $space = '\x{0020}\x{0009}\x{000A}\x{000C}\x{000D}';
65 &GetOptions("in=s"=>\$opt_in_cs,
66 "out=s"=>\$opt_out_cs,
70 Usage: $0 -i <input coding system> -o <cmap encoding>
72 Utf8mcs, Utf8cns, Utf8gb, Utf8jis, Utf8ks
74 UniCNS, UniGB, UniJIS, UniKS, UniMulti
77 if($opt_in_cs or $opt_out_cs){
81 ($in_cs,$out_cs)=($0=~/(Utf8.+)To(\w+)/);
85 # Utf8mcs,Utf8cns,Utf8gb,Utf8jis,Utf8ks,
87 # UniCNS,UniGB,UniJIS,UniKS,UniMulti
89 $in_cs=~s/Utf8/ucs\@/;
92 or not defined($in_cs)
93 or not defined($out_cs)){
98 $omegadb_path=~s!/$!!;
100 $idsdata_file="$omegadb_path/idsdata.pl";
104 if(-e $idsdata_file){
105 open(IDSDATA,"+<:utf8",$idsdata_file) or die;
106 flock(IDSDATA,LOCK_EX);
112 # require $idsdata_file;
114 open(IDSDATA,">:utf8",$idsdata_file) or die;
115 flock(IDSDATA,LOCK_EX);
122 $geta=pack("U",0x3013);
123 #$geta=pack("U",0xfffd);
126 "=gt-pj-1","=gt-pj-2","=gt-pj-3","=gt-pj-4","=gt-pj-5",
127 "=gt-pj-6","=gt-pj-7","=gt-pj-8","=gt-pj-9","=gt-pj-10",
129 #,"=gt-pj-k1","=gt-pj-k2"
131 @HZK=("=hanziku-1","=hanziku-2","=hanziku-3","=hanziku-4",
132 "=hanziku-5","=hanziku-6","=hanziku-7","=hanziku-8",
133 "=hanziku-9","=hanziku-10","=hanziku-11","=hanziku-12");
136 %order_map=('c'=>'UniCNS',
145 foreach $out_cs ('UniCNS','UniGB','UniJIS','UniKS','UniMulti'){
146 if(defined($opt_order{$out_cs})){
147 if($opt_order{$out_cs}=~/^[cgjkGHC]+$/){
148 @{$order{$out_cs}}=map {$order_map{$_}}
149 (split(//,$opt_order{$out_cs}));
151 print STDERR "Invalid order for $out_cs!\n";
159 if($in_cs ne 'ucs@mcs'){
160 s/(.)/&get_char_in_mcs($1,$in_cs)/ge;
162 s/(amp.+?;)/&de_tex_er($1)/ge;
163 # s/(&.+?;)/&de_tex_er($1)/ge;
166 for($i=0;$i<=$#chars;$i++){
168 $char_id=unpack("U",$char);
173 }elsif($char_id>0x20 and $char_id<=0x02af){
179 print &latin_parse();
181 }elsif($char_id>=0x2ff0 and $char_id<=0x2fff){
182 # Ideographic Description Characters
186 if(($out_char=&get_output_char($char,$out_cs))){
187 print $out_char,&add_break($i);
188 }elsif($char_id >= 0x20000 && $char_id <=0x2a6df){
189 # CJK Unified Ideographs Extension B
190 if(not defined($ids{$char}) and $ids{$char}[1]>=0){
191 $ids{$char}[0]=$font_start;
192 $ids{$char}[1]=$ids_start;
199 print "{\\fontencoding{OT1}\\fontfamily{" .
200 sprintf("chise%03d",$ids{$char}[0]) .
201 "}\\selectfont\\char$ids{$char}[1]}",&add_break($i);
204 if($opt_allow_unify){
205 @chars_unified=&get_chars_unified($char);
206 if(@chars_unified>0){
207 foreach $char_unified (@chars_unified){
209 =&get_output_char($char_unified,$out_cs))){
210 print $out_char,&add_break($i);
216 if($ids=&get_ids($char)){
217 print &get_macro_for_ids($ids),&add_break($i);
219 print '\rule{1ex}{1ex}',&add_break($i);
226 print IDSDATA 'use utf8;',"\n";
227 foreach $ids (keys %ids){
228 print IDSDATA '$ids{\'',$ids,'\'}='
229 ,'[',join ",",@{$ids{$ids}},"];\n";
231 print IDSDATA '$font_start=',$font_start,";\n";
232 print IDSDATA '$ids_start=',$ids_start,";\n";
234 flock(IDSDATA,LOCK_UN);
241 my($output_char,$atr,$value);
242 $er=~/^(amp)(.*)(;)$/
243 and $prefix=$1,$er=$2,$suffix=$3;
244 $prefix or $prefix="",$suffix or $suffix="";
245 if($er=~/^U[\+|\-]([a-fA-F\d]+)/){
246 $output_char=pack("U",hex($1));
247 }elsif($er=~/^(?:I\-)?($er_prefix_re)\-?([0-9a-fA-F]+)$/){
248 ($atr,$value)=($1,$2);
249 unless($er_alias{$atr}=~/daikanwa|gt/){
252 ($output_char)=&get_chars_matching($er_alias{$atr},$value);
257 return $prefix.$er.$suffix;
265 if(($chars[$i+1]=~m/[$strictly_forbidden_before]/o)
266 and($chars[$i+2]=~m/[$strictly_forbidden_before]/o)){
267 return "\\CJKunbreakablekernone ";
268 }elsif($opt_protrude){
269 if(($chars[$i+1]=~m/[$strictly_forbidden_before]/o)
270 and($chars[$i+2]=~m/[^$strictly_forbidden_before]/o)){
271 return "\\CJKunbreakablekernone \\CJKprotrude ";
276 and($chars[$i+1]=~m/[$strictly_forbidden_before]/o)){
277 return "\\CJKunbreakablekernone ";
279 if($chars[$i]=~m/[$strictly_forbidden_after]/o){
280 return "\\CJKunbreakablekernone ";
283 and($chars[$i+1]=~m/[$forbidden_before]/o)){
284 return "\\CJKunbreakablekerntwo ";
287 if($chars[$i]=~m/[$forbidden_after]/o){
288 return "\\CJKunbreakablekerntwo ";
291 and($chars[$i+1]=~m/[$slightly_forbidden_before]/o)){
292 return "\\CJKunbreakablekernthree ";
294 if($chars[$i]=~m/[$slightly_forbidden_after]/o){
295 return "\\CJKunbreakablekernthree ";
297 if($chars[$i]=~m/[$asian]/o){
298 return "\\CJKbreakablekern ";
300 if(($i<$#chars)and($chars[$i+1]=~m/[$asian]/o)){
301 return "\\CJKbreakablekern ";
307 # return: string for output with TeX macro.
309 my $out_str=$chars[$i];
312 $char_id=unpack("U",$chars[$i]);
313 if($char_id>0x20 and $char_id<=0x02af){
314 $out_str.=$chars[$i];
321 return '{\normalfont {'.$out_str.'}}';
326 # return: character for output,
328 # or GETA character if ids is invalid.
329 my($ids,$ids_argc)=&ids_rest("",0,$chars[$i]);
335 print STDERR "IDS parse error: $ids\n";
339 ($ids,$ids_argc)=&ids_rest($ids,$ids_argc,$chars[$i]);
341 if(($char=&get_char_for_ids($ids))
342 and($out_char=&get_output_char($char,$out_cs))){
345 return &get_macro_for_ids($ids);
352 # arguments: <ids>, <rest number of arguments for ids>, <character>
353 # return: ids and rest number of arguments for ids.
354 my($ids,$ids_argc,$char)=@_;
356 $argc=&ids_argc($char);
358 $ids_argc+=$ids_argc==0?$argc:$argc-1;
363 return ($ids,$ids_argc);
366 sub get_macro_for_ids{
368 # return: TeX macro for ids
369 # or GETA character if ids is invalid for KAGE.
371 $ids=&normalize_ids($ids,"UniJIS");
372 return $geta if(($ids!~/[$idc]/)
373 or($ids=~/[\x{10000}-]/));
375 if(not defined($ids{$ids}) and $ids{$ids}[1]>=0){
376 $ids{$ids}[0]=$font_start;
377 $ids{$ids}[1]=$ids_start;
384 return "{\\fontencoding{OT1}\\fontfamily{"
385 .sprintf("chise%03d",$ids{$ids}[0])
386 ."}\\selectfont\\char$ids{$ids}[1]}";
390 # argument: <ids>, <output coding system>
391 # return: ids or GETA character if ids is invalid for KAGE.
393 $out_cs=~s/Uni(.+)/'ucs@'.lc($1)/e;
396 my($char,$output_char_id);
397 while($ids=~m/(.)/g){
401 }elsif($output_char_id=&get_char_attribute($char,"=$out_cs")){
402 $output_ids.=pack("U",$output_char_id);
403 }elsif($output_char_id=&get_char_attribute($char,"=ucs")){
404 $output_ids.=pack("U",$output_char_id);
405 }elsif($output_char_id=&get_char_attribute($char,"=>$out_cs")){
406 $output_ids.=pack("U",$output_char_id);
407 }elsif($output_char_id=&get_char_attribute($char,"=>ucs")){
408 $output_ids.=pack("U",$output_char_id);
418 # return: character in output coding system or TeX macro or undef.
419 my($char,$out_cs)=@_;
420 my($out_char_id,$suffix);
423 foreach $out_cs (@{$order{$out_cs}}){
424 if($out_cs eq 'UniJIS'
425 and &get_char_attribute($char,"vnd-adobe-cid-unijis-utf16-h")){
426 if($out_char_id=&get_char_attribute($char,'=ucs@jis')
427 or $out_char_id=&get_char_attribute($char,'=ucs')
428 or $out_char_id=&get_char_attribute($char,'=>ucs@jis')
429 or $out_char_id=&get_char_attribute($char,'=>ucs')
431 return '{\selectjisfont\char'.$out_char_id.'}';
433 }elsif($out_cs eq 'UniGB'
434 and &get_char_attribute($char,"vnd-adobe-cid-unigb-ucs2-h")){
435 if($out_char_id=&get_char_attribute($char,'=ucs@gb')
436 or $out_char_id=&get_char_attribute($char,'=ucs')
437 or $out_char_id=&get_char_attribute($char,'=>ucs@gb')
438 or $out_char_id=&get_char_attribute($char,'=>ucs')
440 return '{\selectgbsfont\char'.$out_char_id.'}';
442 }elsif($out_cs eq 'UniCNS'
443 and &get_char_attribute($char,"vnd-adobe-cid-unicns-ucs2-h")){
444 if($out_char_id=&get_char_attribute($char,'=ucs@cns')
445 or $out_char_id=&get_char_attribute($char,'=ucs')
446 or $out_char_id=&get_char_attribute($char,'=>ucs@cns')
447 or $out_char_id=&get_char_attribute($char,'=>ucs')
449 return '{\selectcnsfont\char'.$out_char_id.'}';
451 }elsif($out_cs eq 'UniKS'
452 and &get_char_attribute($char,"vnd-adobe-cid-uniks-ucs2-h")){
453 if($out_char_id=&get_char_attribute($char,'=ucs@ks')
454 or $out_char_id=&get_char_attribute($char,'=ucs')
455 or $out_char_id=&get_char_attribute($char,'=>ucs@ks')
456 or $out_char_id=&get_char_attribute($char,'=>ucs*')
457 or $out_char_id=&get_char_attribute($char,'=>ucs')
459 return '{\selectksxfont\char'.$out_char_id.'}';
461 }elsif($out_cs eq 'GT'){
462 return $gt if($gt=&get_macro_for_GT($char));
463 }elsif($out_cs eq 'HZK'){
464 return $hzk if($hzk=&get_macro_for_HZK($char));
465 }elsif($out_cs eq 'CDP'){
466 return $cdp if($cdp=&get_macro_for_CDP($char));
473 # argument: <character>
477 $ids=&get_char_attribute($char,"ids-aggregated")
478 or $ids=&get_char_attribute($char,"ids");
479 # or $ids=&get_char_attribute($char,"ideographic-structure");
484 sub get_char_for_ids{
485 # argument: <ideographic description sequence>
486 # return: char or undef.
490 if(($output_char)=&get_chars_matching("ids",$ids)){
498 # argument: <char>, <input coding system>
499 # return: char in ucs@mcs.
503 return $char if($in_cs eq 'ucs@mcs');
505 if(($output_char)=&get_chars_matching("=$in_cs",unpack("U",$char))){
512 sub get_chars_unified{
514 my($chars,$ucs,$char_ucs);
517 if($chars=&get_char_attribute($char,'->ucs-unified')){
518 $chars=~s/^\((.*)\)$/$1/;
519 return (split(/\s*\?/,$chars));
520 }elsif($ucs=&get_char_attribute($char,'=>ucs*')
521 or $ucs=&get_char_attribute($char,'=>ucs')){
522 $char_ucs=pack("U",$ucs);
523 if($chars=&get_char_attribute($char_ucs,'->ucs-unified')){
524 $chars=~s/^\((.*)\)$/$1/;
525 @chars=grep {not /^$char$/}
526 (split(/\s*\?/,$chars));
527 push(@chars,$char_ucs);
533 sub get_macro_for_GT{
535 # return: TeX macro for GT fonts or undef.
539 if($gt=&get_char_attribute($char,$_)){
540 m/gt\-pj\-(\d+)/ and $GT=$1;
545 return "{\\fontencoding{OT1}\\fontfamily{"
546 .sprintf("gt%02d",$GT)
547 ."}\\selectfont\\char".($gt|0x8080)."}";
553 sub get_macro_for_HZK{
555 # return: TeX macro for Hanziku fonts or undef.
559 if($hzk=&get_char_attribute($char,$_)){
560 m/hanziku\-(\d+)/ and $HZK=$1;
565 return "{\\fontencoding{OT1}\\fontfamily{".sprintf("hzk%02d",$HZK)."}\\selectfont\\char".$hzk."}";
571 sub get_macro_for_CDP{
573 # return: TeX macro for CDP fonts or undef.
577 if($cdp=&get_char_attribute($char,$_)){
582 $ucs=&get_char_attribute(&get_chars_matching("=big5-pua",$cdp),"=ucs");
584 return "{\\fontencoding{OT1}\\fontfamily{cdp}\\selectfont\\char"
588 print STDERR "This should not happen.\n";
589 print STDERR "ucs code point of CDP: $cdp not found.\n";