X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/d26b60713dda612610cea4344acb625a68bfeb4a..2a99bcdadec88d7483903e2c8420ad3e78fcb805:/charset.plp

diff --git a/charset.plp b/charset.plp
index aa81d0b..d25f23c 100644
--- a/charset.plp
+++ b/charset.plp
@@ -12,7 +12,7 @@ Html({
 		reference common overview table
 	'],
 	stylesheet => [qw'light'],
-	data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'],
+	data => [qw'charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl'],
 });
 
 my @tablist = split m{/+}, $Request || 'default';
@@ -48,6 +48,7 @@ print join " •\n", (
 	}
 	[
 		iso      => 'ISO',
+		win      => 'Windows',
 		dos      => 'DOS',
 		mac      => 'Apple',
 		ebcdic   => 'EBCDIC',
@@ -62,7 +63,6 @@ print join " •\n", (
 		cyrillic => 0,
 		hebrew   => 0,
 	],
-	[ uc => 'Unicode' ],
 );
 :>.
 </p>
@@ -78,8 +78,8 @@ sub tabinput {
 	my $input = shift or return;
 
 	state $ALIAS = {
-		default    => [qw( unicode+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
-		uc         => [qw( U+cols=32 unicode+0-4095 unicode+4096-6319  unicode+6320-8191 )],
+		default    => [qw( u+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
+		unicode    => [qw( uu+cols=32+realsize u+0-4095 u+4096-6319 u+6320-8191 )],
 		us         => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
 		ebcdic     => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
 		iso        => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
@@ -105,7 +105,7 @@ sub tabinput {
 		hebrew     => [qw( iso-8859-8 cp1255 MacHebrew cp862  U59-5F )],
 		thai       => [qw( iso-8859-11 cp874 MacThai )],
 		vietnamese => [qw( viscii cp1258 MacVietnamese )],
-		symbol     => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )],
+		symbols    => [qw( symbol dingbats MacDingbats )],
 
 		# iso-code shorthand
 		1 => 'westeur',
@@ -125,7 +125,7 @@ sub tabinput {
 		return;
 	}
 
-	state $visible = {};  # all present tables
+	state $visible = {'' => 1};  # all present tables
 	my %row = (offset => 0, cols => 16);
 	my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
 
@@ -133,6 +133,11 @@ sub tabinput {
 		state $INHERIT = {
 			'cp437'       => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later
 			'gsm0338'     => ['ascii' => '0-127'],
+			'dingbats'    => ['' => '32-127+160'],
+			'MacDingbats' => ['dingbats' => '128-159'],
+			'AdobeZdingbat'=> ['MacDingbats' => '128-159'], # should be identical but maps to private use
+			'symbol'      => ['' => '32-127+160'],
+			'AdobeSymbol' => ['symbol' => '32-127+160', '' => '32-127+160'], # minor differences, irrelevant except for different '€'
 
 			'iso-8859-2'  => ['iso-8859-1' => '160'],
 			'iso-8859-3'  => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9
@@ -247,7 +252,7 @@ sub tabinput {
 		}
 	}
 
-	if ($input =~ /^U([0-9a-fA-F]+)(?:-([0-9a-fA-F]+))?/) {
+	if ($input =~ m{ \A u ([0-9a-f]+) (?:-([0-9a-f]+))? \z }ix) {
 		my $start = hex($1) << ($2 ? 4 : 8);
 		my $end = $2 ? (hex($2) << 4) + $row{cols} - 1 : $start + 255;
 		$row{table} = join '', map { chr } $start .. $end;
@@ -256,13 +261,13 @@ sub tabinput {
 		$row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
 		$row{offset} = $start % 256;
 	}
-	elsif ($input eq 'U') {
+	elsif (lc $input eq 'uu') {
 		$row{set} = 'Unicode planes';
 		$row{cell} = do 'charset-ucplanes.inc.pl'
 			or Alert('Table data could not be read', $@ || $!);
 		$row{endpoint} = 1023 * $row{cell}->{colsize};
 	}
-	elsif ($input eq 'unicode') {
+	elsif (lc $input eq 'u') {
 		$row{cell} = do 'charset-unicode.inc.pl'
 			or Alert('Table data could not be read', $@ || $!);
 
@@ -298,19 +303,29 @@ sub tabinput {
 		$row{endpoint} -= $row{offset};
 
 		if ($row{set} eq 'cp437') {
-			for my $phipos (237 - $row{offset}) {
-				next if $phipos < 0 or $phipos > $row{endpoint};
+			if ($row{offset} <= 0xED and $row{endpoint} >= 0xED - $row{offset}) {
 				# replace phi glyph
-				substr($row{table}, $phipos, 1) = pack 'U*', 0x3D5;
+				substr($row{table}, 0xED - $row{offset}, 1) = 'ϕ';
 			}
-			if ($row{offset} == 0) {
+			if ($row{offset} < 0x20) {
 				# replace control characters by visible variants
-				substr($row{table}, 0, 32) = pack 'U*', map {hex} qw(
-					2007 263A 263B 2665 2666 2663 2660 2022
-					25D8 25CB 25D9 2642 2640 266A 266B 263C
-					25BA 25C4 2195 203C 00B6 00A7 25AC 21A8
-					2191 2193 2192 2190 221F 2194 25B2 25BC
-				);
+				my $sub = substr ' ☺☻♥♦♣♠•◘○◙♂♀♪♫☼►◄↕‼¶§▬↨↑↓→←∟↔▲▼', $row{offset};
+				substr($row{table}, 0, length $sub) = $sub;
+			}
+		}
+		elsif ($row{set} eq 'symbol') {
+			if ($row{offset} <= 0x60 and $row{endpoint} >= 0x60 - $row{offset}) {
+				# replace radical extender by closest unicode equivalent
+				substr($row{table}, 0x60 - $row{offset},  1) = '│';
+			}
+			if ($row{offset} <= 0xBD and $row{endpoint} >= 0xFF - $row{offset}) {
+				substr($row{table}, 0xBD - $row{offset},  2) = '⏐⎯'; # arrow extenders
+				substr($row{table}, 0xD2 - $row{offset},  3) = '®©™'; # serif variants
+				substr($row{table}, 0xE0 - $row{offset},  1) = '◊'; # replace lookalike, should match AdobeSymbol
+				substr($row{table}, 0xE2 - $row{offset},  3) = '®©™'; # sans-serif variants
+				substr($row{table}, 0xE6 - $row{offset}, 10) = '⎛⎜⎝⎡⎢⎣⎧⎨⎩⎪';
+				substr($row{table}, 0xF0 - $row{offset},  1) = '€';
+				substr($row{table}, 0xF4 - $row{offset}, 11) = '⎮⌡⎞⎟⎠⎤⎥⎦⎫⎬⎭';
 			}
 		}
 
@@ -346,15 +361,16 @@ sub range_cell {
 		if ($len -= $rest) {
 			# continued on new row
 			my @next = ($len * $colsize, "$class joinu");
+			my $separate = $cols - $len > $rest;  # columns not on next row
 			if ($len > $rest) {
 				# minority remains
 				push @next, $name, $title;
 				$title ||= $name;
-				$name = '';
+				$name = $separate && '…';
 			}
 			else {
 				# minority on next row
-				push @next, '"', $title || $name;
+				push @next, $separate && '"', $title || $name;
 			}
 			$table->{$offset + $colsize*$rest} //= \@next;
 			$class .= ' joind';
@@ -389,7 +405,7 @@ sub range_cell {
 	$attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
 	$attr .= sprintf ' class="%s"', $class if $class;
 	$attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
-	return "<td$attr>$name";
+	return "<td$attr>$name\n";
 }
 
 for my $row (@request) {
@@ -403,7 +419,7 @@ for my $row (@request) {
 	printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
 	my $title = $row->{set};
 	$title .= " <aside>(over $_)</aside>"
-		for grep { $_ ne 'iso-8859-1' } $row->{parent} // ();
+		for $row->{parent} || ();
 	printf '<caption>%s</caption>', $title;
 	print '<col>' x ($cols + 1);
 	for my $section (qw{thead}) {
@@ -432,39 +448,39 @@ for my $row (@request) {
 				printf '%X', ($offset + $row->{offset}) / $rowdiv;
 			}
 		}
+		say '';
+
 		for (1 .. $cols) {
 			if ($row->{cell}) {
 				print range_cell($row, $offset);
 				next;
 			}
 
+			my $cp = $offset + $row->{offset};
 			my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
 				substr $row->{table}, $offset, 1;
-			if ($glyph eq $NOCHAR) {
-				print '<td>';
-				next;
-			}
+			my ($cell, $name, $class) = $glyph eq $NOCHAR ? () :
+				$glyphs->glyph_html($glyph);
 
 			if (exists $get{compare}) {
 				state $visible = {};
-				my $cp = $offset + $row->{offset};
-				printf '<td class="%s" title="%3$s">%2$s',
+				$class = (
 					$cp == ord $glyph ? 'l4' :
 					$row->{parent} && $glyph eq
 						Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
+					!$class ? undef :
 					$visible->{$glyph} ? 'l2' :
-					'l1',
-					$glyphs->glyph_html($glyph);
+					'l1'
+				);
 				$visible->{$glyph}++;
-				next;
 			}
 
-			print "\n".$glyphs->glyph_cell($glyph);
+			say sprintf $class ? '<td title="%s" class="X %s">%s' : '<td title="%s">',
+				$name, $class, $cell;
 		}
 		continue {
 			$offset += $colsize;
 		}
-		print "\n";
 	}
 	say '</table></div>';
 }