X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/1526acc93e4a99f79545a62d5700d9020fe59b5b..42d8bbb170d20da856db097f3cec280e37578c31:/charset.plp

diff --git a/charset.plp b/charset.plp
index 309139e..c3477f1 100644
--- a/charset.plp
+++ b/charset.plp
@@ -31,7 +31,7 @@ sub tabinput {
 	my $input = shift or return;
 
 	state $ALIAS = {
-		default    => [qw( unicode+0-192 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
+		default    => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
 		us         => [qw( cp437 cp863 AdobeStandardEncoding gsm0338 )],
 		ebcdic     => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )],
 		iso        => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
@@ -99,14 +99,15 @@ sub tabinput {
 			'hp-roman8'   => ['iso-8859-1' => '160'],
 
 			'cp1252'      => ['iso-8859-1' => '128-159'],
-			'cp1250'      => ['iso-8859-2' => '128-191'],
-			'cp1254'      => ['iso-8859-9' => '128-159'],
-			'cp874'       => ['iso-8859-11' => '128-159'], # windows-874 actually cp1162
-			'cp1257'      => ['iso-8859-13' => '128'],
+			'cp1250'      => ['iso-8859-2' => '128-191', 'cp1252' => '128'],
+			'cp1254'      => ['iso-8859-9' => '128-159', 'cp1252' => '128'],
+			'cp874'       => ['iso-8859-11' => '128-159', 'cp1252' => '128'], # windows-874 actually cp1162
+			'cp1257'      => ['iso-8859-13' => '128', 'cp1252' => '128'],
 			'cp1251'      => ['cp1252' => '128'],
 			'cp1253'      => ['cp1252' => '128'],
 			'cp1255'      => ['iso-8859-8' => '128-223', 'cp1252' => '128'],
 			'cp1256'      => ['cp1252' => '128'],
+			'cp1258'      => ['cp1252' => '128'],
 
 			'cp850'       => ['cp437' => '144'],
 			'cp860'       => ['cp437' => '128-175'],
@@ -137,18 +138,21 @@ sub tabinput {
 
 		my @parents = @{ $INHERIT->{$input} || [] };
 
-		if (my ($parent, $part) = pairfirst { $visible->{$a} } @parents) {
+		if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
 			$row{parent} = $parent;
 			$params = $part;
+			$params = 128 unless $visible->{$parent}
+				or ($input eq 'MacCroatian' and defined $visible->{MacRomanian});
 		}
-		elsif ($visible->{ascii}) {
-			$row{parent} = $parents[0] || 'ascii';
+		elsif (defined $visible->{ascii}) {
+			$row{parent} = $parents[0];
 			$params = $parents[1] || 128;
 			$params = 128 if $params >= 128;  # ascii offset at most
 		}
 		elsif (@parents) {
 			$row{parent} = $parents[0];
 		}
+		$visible->{$_} //= 0 for $row{parent};
 	}
 
 	if (defined $params and $params =~ m/^ (\d+) (-\d+)? /x) {
@@ -161,28 +165,28 @@ sub tabinput {
 		my $end = $2 ? hex($2) << 4 : $start + 240;
 		$row{table} = join '', map { chr } $start .. $end+15;
 		utf8::upgrade($row{table});  # prevent latin1 output
-		$row{len} = length $row{table};
+		$row{endpoint} = $end + 14 - $start;
 		$row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
 	}
 	elsif ($input eq 'U') {
-		$row{len} = 1024;
 		$row{set} = 'Unicode planes';
 		$row{cell} = do 'charset-ucplanes.inc.pl'
 			or Alert('Table data could not be read', $@ || $!);
 		$row{cols} *= 2;
+		$row{endpoint} = 1023 * $row{cell}->{colsize};
 	}
 	elsif ($row{set} = Encode::resolve_alias($input)) {
 		if ($row{set} eq 'Internal') {
-			$row{len} = $endpoint < 255 ? 640 : 8192;
 			$row{set} = 'Unicode BMP';
 			$row{cell} = do 'charset-unicode.inc.pl'
 				or Alert('Table data could not be read', $@ || $!);
+			$row{endpoint} = ($endpoint || 8191) * $row{cell}->{colsize};
 		}
 		elsif ($row{set} eq 'utf-8-strict') {
-			$row{len} = 256;
 			$row{set} = 'UTF-8';
 			$row{cell} = do 'charset-utf8.inc.pl'
 				or Alert('Table data could not be read', $@ || $!);
+			$row{endpoint} = 255;
 		}
 		else {
 			if ($row{set} eq 'MacHebrew') {
@@ -196,7 +200,7 @@ sub tabinput {
 				# substr strings is twice as fast as splitting to an array
 				$row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
 			}
-			$row{len} = $endpoint - $row{offset};
+			$row{endpoint} = $endpoint - $row{offset};
 
 			if ($row{set} eq 'cp437' and !$row{offset}) {
 				substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
@@ -281,7 +285,6 @@ for my $row (@request) {
 	my $rowdiv = 16 ** $coldigits;  # row divide for column digits
 	$rowdiv = 1 if $rowdiv != $cols * $colsize;  # divide only if all columns are matched
 	my $offset = 0;
-	my $endpoint = $offset + $row->{len} * $colsize;
 
 	printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
 	my $title = $row->{set};
@@ -295,7 +298,7 @@ for my $row (@request) {
 		print "\n";
 	}
 	print '<tbody>';
-	while ($offset < $endpoint - 1) {
+	while ($offset < $row->{endpoint}) {
 		print '<tr><th>';
 		{
 			if (my $rowmod = $offset % $rowdiv) {
@@ -320,6 +323,20 @@ for my $row (@request) {
 				next;
 			}
 
+			if (exists $get{compare}) {
+				state $visible = {};
+				my $cp = $offset + $row->{offset};
+				printf '<td class="%s" title="%3$s">%2$s',
+					$cp == ord $glyph ? 'l4' :
+					$row->{parent} && $glyph eq
+						Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
+					$visible->{$glyph} ? 'l2' :
+					'l1',
+					$glyphs->glyph_html($glyph);
+				$visible->{$glyph}++;
+				next;
+			}
+
 			print "\n".$glyphs->glyph_cell($glyph);
 		}
 		continue {
@@ -334,7 +351,13 @@ for my $row (@request) {
 <hr>
 
 <div class="legend">
-	<table class="glyphs"><tr>
+	<table class="glyphs"><tr><: if (exists $get{compare}) { :>
+	<td class="X l4">unicode
+	<td class="X l3">inherited
+	<td class="X l2">existing
+	<td class="X l1">original
+	<td class="">unassigned
+<: } else { :>
 	<td class="X Cc">control
 	<td class="X Zs"><span>whitespace</span>
 	<td class="X Mn">diacritic<table class="glyphs"><tr>
@@ -371,6 +394,6 @@ for my $row (@request) {
 	<td class="X Xd">deprecated
 	<td class="">unassigned
 	<td class="X Xi">invalid
-	</table>
+<: } :>	</table>
 </div>