X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=inCHISE;h=3fb24fa64bf1b4a8a154bf783ef1c00977a062a5;hb=850d741ce7ed32d75cec3753d0cfff8b47acf96b;hp=9f7a731f7ca075696055b492bb52bfbd7bdfc49f;hpb=f9120418868c224a29b458672f041e6d209d8762;p=chise%2Fomega.git

diff --git a/inCHISE b/inCHISE
index 9f7a731..3fb24fa 100755
--- a/inCHISE
+++ b/inCHISE
@@ -60,7 +60,7 @@ my $forbidden_before
 my $slightly_forbidden_before
     = '\x{000a}\#\-‐−‰′″℃゛゜ゝゞヽヾ＂％－ﾞﾟ';
 
-my $asian = '\x{1100}-\x{11FF}\x{2E80}-\x{D7AF}\x{F900}-\x{FAFF}\x{FE30}-\x{FE4F}\x{FF00}-\x{FFFFFF}';
+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.
 
 my $space = '\x{0020}\x{0009}\x{000A}\x{000C}\x{000D}';
 
@@ -170,11 +170,21 @@ foreach $out_cs ('UniCNS','UniGB','UniJIS','UniKS','UniMulti'){
 
 while(<>){
     utf8::decode($_);
+
     if($in_cs ne 'ucs@mcs'){
 	s/(.)/&get_char_in_mcs($1,$in_cs)/ge;
     }
     s/(amp.+?;)/&de_tex_er($1)/ge;
 #    s/(&.+?;)/&de_tex_er($1)/ge;
+    s/([$asian])\s+([$asian])/$1$2/g unless($out_cs eq 'UniKS');
+    s/([$asian])\s*([^$asian])/$1 $2/g;
+    s/([^$asian$idc])\s*([$asian])/$1 $2/g;
+    s/\-\-\-/pack("U",0x2014)/geo;# EM DASH
+    s/\-\-/pack("U",0x2013)/geo;# EN DASH
+    s/\`\`/pack("U",0x201f)/geo;# DOUBLE HIGH-REVERSED-9 QUOTATION MARK
+    s/\`/pack("U",0x201b)/geo;# SINGLE HIGH-REVERSED-9 QUOTATION MARK
+    s/\'\'/pack("U",0x201d)/geo;# RIGHT DOUBLE QUOTATION MARK
+    s/\'/pack("U",0x2019)/geo;# RIGHT DOUBLE QUOTATION MARK
     @chars=split(//);
   CHAR:
     for($i=0;$i<=$#chars;$i++){
@@ -182,7 +192,25 @@ while(<>){
 	$char_id=unpack("U",$char);
 
 	if($char_id<=0x20){
-	    print $chars[$i];
+	    # add quarter space between asian and non-asian.
+	    if($i>0 and $i<$#chars){
+		if(($chars[$i-1]=~/[$asian]/
+		    and $chars[$i+1]=~/[^$asian]/
+		      and $chars[$i+1]=~/[$strictly_forbidden_before$slightly_forbidden_before]/)
+		   or($chars[$i-1]=~/[^$asian]/
+		      and $chars[$i+1]=~/[$asian]/
+		      and $chars[$i-1]=~/[$strictly_forbidden_after$slightly_forbidden_after]/)
+		   ){
+		    print '';
+		}elsif(($chars[$i-1]=~/[$asian]/
+			and $chars[$i+1]=~/[^$asian]/)
+		       or($chars[$i-1]=~/[^$asian]/
+			  and $chars[$i+1]=~/[$asian]/)){
+		    print '{\selectjisfont\hspace{.25ex}}';
+		}else{
+		    print ' ';
+		}
+	    }
 	    next CHAR;
 	}elsif($char=~m/($tex_meta_re)/o){
 	    print $tex_meta{$1};
@@ -205,6 +233,8 @@ while(<>){
                # Cyrillic
 	       or($char_id>=0x0530 and $char_id<=0x058f)
                # Armenian
+	       or($char_id>=0x2010 and $char_id<=0x2046)
+	       # General Punctuation (partial)
 	       ){
 	    print &latin_parse();
 	    next CHAR;
@@ -240,8 +270,8 @@ while(<>){
 				$font_start++;
 			    }
 			}
-			print "{\\fontencoding{OT1}\\fontfamily{" .
-			    sprintf("chise%03d",$ids{$char}[0]) .
+			print "{\\fontencoding{OT1}\\fontfamily{",
+			    sprintf("chise%03d",$ids{$char}[0]) ,
 				"}\\selectfont\\char$ids{$char}[1]}",&add_break($i);
 			next CHAR;
 		    }
@@ -297,41 +327,46 @@ sub add_break{
     if($i<($#chars-1)){
 	if(($chars[$i+1]=~m/[$strictly_forbidden_before]/o)
 	   and($chars[$i+2]=~m/[$strictly_forbidden_before]/o)){
-	    return "\\CJKunbreakablekernone ";
+	    return "\\CJKunbreakablekernone{}";
 	}elsif($opt_protrude){
 	    if(($chars[$i+1]=~m/[$strictly_forbidden_before]/o)
 	       and($chars[$i+2]=~m/[^$strictly_forbidden_before]/o)){
-		return "\\CJKunbreakablekernone \\CJKprotrude ";
+		return "\\CJKunbreakablekernone \\CJKprotrude{}";
 	    }
 	}
     }
     if(($i<$#chars)
+       and($chars[$i+1]eq" ")){
+       # preserve space.
+ 	return "";
+    }
+    if(($i<$#chars)
        and($chars[$i+1]=~m/[$strictly_forbidden_before]/o)){
-	return "\\CJKunbreakablekernone ";
+	return "\\CJKunbreakablekernone{}";
     }
     if($chars[$i]=~m/[$strictly_forbidden_after]/o){
-	return "\\CJKunbreakablekernone ";
+	return "\\CJKunbreakablekernone{}";
     }
     if(($i<$#chars)
        and($chars[$i+1]=~m/[$forbidden_before]/o)){
-	return "\\CJKunbreakablekerntwo ";
+	return "\\CJKunbreakablekerntwo{}";
 
     }
     if($chars[$i]=~m/[$forbidden_after]/o){
-	return "\\CJKunbreakablekerntwo ";
+	return "\\CJKunbreakablekerntwo{}";
     }
     if(($i<$#chars)
        and($chars[$i+1]=~m/[$slightly_forbidden_before]/o)){
-	return "\\CJKunbreakablekernthree ";
+	return "\\CJKunbreakablekernthree{}";
     }
     if($chars[$i]=~m/[$slightly_forbidden_after]/o){
-	return "\\CJKunbreakablekernthree ";
+	return "\\CJKunbreakablekernthree{}";
     }
     if($chars[$i]=~m/[$asian]/o){
-	return "\\CJKbreakablekern ";
+	return "\\CJKbreakablekern{}";
     }
     if(($i<$#chars)and($chars[$i+1]=~m/[$asian]/o)){
-	return "\\CJKbreakablekern ";
+	return "\\CJKbreakablekern{}";
     }
 }
 
@@ -339,8 +374,7 @@ sub latin_parse{
     # arguments: none
     # return: string for output with TeX macro.
     my($char_id);
-    my $out_str=$chars[$i];
-    $i++;
+    my $out_str="";
     while($i<=$#chars){
 	$char_id=unpack("U",$chars[$i]);
 	if(($char_id>0x20 and $char_id<=0x021f)
@@ -356,13 +390,16 @@ sub latin_parse{
 	    # Latin Extended Additional
             # 0x1e00 -> 0x0600, etc.
 	    $out_str.=pack("U",$char_id-0x1800);
+	}elsif($char_id>=0x2010 and $char_id<=0x2046){
+	    # General Punctuation (partial)
+	    $out_str.=pack("U",$char_id-0x1000);
 	}else{
 	    $i--;
 	    last;
 	}
 	$i++;
     }
-    return '{\fontencoding{UT1}\fontfamily{omlgc}\selectfont '.$out_str.'}';
+    return '{\fontencoding{OT1}\fontfamily{omlgc}\selectfont '.$out_str.'}';###UT1?
 }
 
 sub ids_parse{
@@ -472,7 +509,7 @@ sub get_output_char{
 	       or $out_char_id=&get_char_attribute($char,'=>ucs')
 	       or $out_char_id=&get_char_attribute($char,'=>ucs*')
 	       ){
-		return '{\selectjisfont\char'.$out_char_id.'}';
+		return "{\\selectjisfont\\char$out_char_id}";
 	    }
 	}elsif($out_cs eq 'UniGB'
 	       and &get_char_attribute($char,"vnd-adobe-cid-unigb-ucs2-h")){
@@ -482,7 +519,7 @@ sub get_output_char{
 	       or $out_char_id=&get_char_attribute($char,'=>ucs')
 	       or $out_char_id=&get_char_attribute($char,'=>ucs*')
 	       ){
-		return '{\selectgbsfont\char'.$out_char_id.'}';
+		return "{\\selectgbsfont\\char$out_char_id}";
 	    }
 	}elsif($out_cs eq 'UniCNS'
 	       and &get_char_attribute($char,"vnd-adobe-cid-unicns-ucs2-h")){
@@ -492,7 +529,7 @@ sub get_output_char{
 	       or $out_char_id=&get_char_attribute($char,'=>ucs')
 	       or $out_char_id=&get_char_attribute($char,'=>ucs*')
 	       ){
-		return '{\selectcnsfont\char'.$out_char_id.'}';
+		return "{\\selectcnsfont\\char$out_char_id}";
 	    }
 	}elsif($out_cs eq 'UniKS'
 	       and &get_char_attribute($char,"vnd-adobe-cid-uniks-ucs2-h")){
@@ -502,7 +539,7 @@ sub get_output_char{
 	       or $out_char_id=&get_char_attribute($char,'=>ucs')
 	       or $out_char_id=&get_char_attribute($char,'=>ucs*')
 	       ){
-		return '{\selectksxfont\char'.$out_char_id.'}';
+		return "{\\selectksxfont\\char$out_char_id}";
 	    }
 	}elsif($out_cs eq 'GT'){
 	    return $gt if($gt=&get_macro_for_GT($char));
@@ -610,7 +647,9 @@ sub get_macro_for_HZK{
 	}
     }
     if($hzk){
-	return "{\\fontencoding{OT1}\\fontfamily{".sprintf("hzk%02d",$HZK)."}\\selectfont\\char".$hzk."}";
+	return "{\\fontencoding{OT1}\\fontfamily{"
+	    .sprintf("hzk%02d",$HZK)
+	    ."}\\selectfont\\char".$hzk."}";
     }else{
 	return undef;
     }
@@ -629,9 +668,7 @@ sub get_macro_for_CDP{
     if($cdp){
 	$ucs=&get_char_attribute(&get_chars_matching("=big5-pua",$cdp),"=ucs");
 	if($ucs){
-	    return "{\\fontencoding{OT1}\\fontfamily{cdp}\\selectfont\\char"
-		.$ucs.
-		    "}";
+	    return "{\\fontencoding{OT1}\\fontfamily{cdp}\\selectfont\\char$ucs}";
 	}else{
 	    print STDERR "This should not happen.\n";
 	    print STDERR "ucs code point of CDP: $cdp not found.\n";