charset: keep original endpoint during character overrides

[sheet.git] / charset.plp

diff --git a/charset.plp b/charset.plp
index 5955c390a6da4327d490972a7bc3ea585659c873..5358a37b47e716e6627b72ea80c706cfe8a92058 100644 (file)
--- 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';
@@ -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 wingdings wingdings2 wingdings3 webdings )],
 
                # 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 +133,15 @@ 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 '€'
+                       'wingdings'   => ['' => '32'],
+                       'wingdings2'  => ['' => '32'],
+                       'wingdings3'  => ['' => '32'],
+                       'webdings'    => ['' => '32'],
 
                        'iso-8859-2'  => ['iso-8859-1' => '160'],
                        'iso-8859-3'  => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9
@@ -247,6 +256,10 @@ sub tabinput {
                }
        }
 
+       if ($input =~ m{ \A (?:wing|web)dings \d* \z }ix) {
+               eval "require Encode::\u$input";
+       }
+
        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;
@@ -295,24 +308,35 @@ sub tabinput {
                        # substr strings is twice as fast as splitting to an array
                        $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint});
                }
-               $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 $row{endpoint} >= 0xED) {
                                # replace phi glyph
-                               substr($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 $row{endpoint} >= 0x60) {
+                               # replace radical extender by closest unicode equivalent
+                               substr($row{table}, 0x60 - $row{offset},  1) = '│';
+                       }
+                       if ($row{offset} <= 0xBD and $row{endpoint} >= 0xFF) {
+                               substr($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) = '⎮⌡⎞⎟⎠⎤⎥⎦⎫⎬⎭';
+                       }
+               }
+
+               $row{endpoint} -= $row{offset};
 
                $visible->{ascii} =  # assume common base
                $visible->{ $row{set} } = 1;
@@ -404,7 +428,7 @@ 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}) {