charset: omit unassigned half of ascii table

[sheet.git] / charset.plp

diff --git a/charset.plp b/charset.plp
index 25bc44b6e14ac26f77b98ff3a08be0da33e91fd3..6d60e2733774720f1d3bdabb85a346dafcd912d6 100644 (file)
--- a/charset.plp
+++ b/charset.plp
@@ -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 )],
-               symbols    => [qw( symbol dingbats MacDingbats wingdings )],
+               symbols    => [qw( symbol dingbats MacDingbats wingdings wingdings2 wingdings3 webdings )],
 
                # iso-code shorthand
                1 => 'westeur',
@@ -131,6 +131,7 @@ sub tabinput {
 
        if (not defined $params) {
                state $INHERIT = {
+                       'ascii'       => ['' => '0-127'],
                        'cp437'       => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later
                        'gsm0338'     => ['ascii' => '0-127'],
                        'dingbats'    => ['' => '32-127+160'],
@@ -139,6 +140,9 @@ sub tabinput {
                        '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
@@ -235,26 +239,26 @@ sub tabinput {
                elsif ($param =~ m{ \A cols = (\d+) \z }x) {
                        $row{cols} = $1;
                }
-               elsif ($param =~ m{ \A (?<offset> \d+) (?: [-] (?<endpoint> \d+) )? \z }x) {
+               elsif ($param =~ m{ \A (?<start> \d+) (?: [-] (?<end> \d+) )? \z }x) {
                        if (defined $row{endpoint}) {
                                # extend earlier range
-                               my $skip = int(($row{endpoint} || $row{offset}) / $row{cols});
-                               for ($skip + 1 .. ($+{offset} / $row{cols}) - 1) {
-                                       $row{skip}->{ $_ * $row{cols} - $row{offset} }++;
+                               my $skip = int(($row{endpoint} || $row{startpoint}) / $row{cols});
+                               for ($skip + 1 .. ($+{start} / $row{cols}) - 1) {
+                                       $row{skip}->{ $_ * $row{cols} - $row{startpoint} }++;
                                }
                        }
                        else {
-                               $row{offset} = $+{offset};
+                               $row{startpoint} = $+{start};
                        }
-                       $row{endpoint} = $+{endpoint} || 0;
+                       $row{endpoint} = $+{end} || 0;
                }
                else {
                        Alert("Unknown option <q>$param</q> for charset $input");
                }
        }
 
-       if (lc $input eq 'wingdings') {
-               require Encode::Wingdings;
+       if ($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) {
@@ -265,24 +269,22 @@ sub tabinput {
                $row{endpoint} = $end - $start;
                $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
                $row{offset} = $start % 256;
+               $row{startpoint} = 0;
        }
        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};
+               $row{endpoint} ||= 1023;
        }
        elsif (lc $input eq 'u') {
                $row{cell} = do 'charset-unicode.inc.pl'
                        or Alert('Table data could not be read', $@ || $!);
 
                $row{endpoint} ||= 8191;
-               $row{endpoint}  *= $row{cell}->{colsize};
-               $row{startpoint} = $row{cell}->{colsize} * $row{offset};
-               $row{offset} = 0;
                $row{set} = 'Unicode ' . (
-                       $row{startpoint} <  0x10000 && $row{endpoint} < 0x10000 ? 'BMP' :
-                       $row{startpoint} >= 0x10000 && $row{endpoint} < 0x20000 ? 'SMP' :
+                       $row{startpoint} <  0x1000 && $row{endpoint} < 0x1000 ? 'BMP' :
+                       $row{startpoint} >= 0x1000 && $row{endpoint} < 0x2000 ? 'SMP' :
                        'allocations'
                );
        }
@@ -293,6 +295,7 @@ sub tabinput {
                $row{endpoint} = 255;
        }
        elsif ($row{set} = Encode::resolve_alias($input)) {
+               $row{offset} = delete $row{startpoint};
                $row{endpoint} ||= 255;
                if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') {
                        # array of possibly multiple characters per code point
@@ -305,10 +308,9 @@ 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 ($row{set} eq 'cp437') {
-                       if ($row{offset} <= 0xED and $row{endpoint} >= 0xED - $row{offset}) {
+                       if ($row{offset} <= 0xED and $row{endpoint} >= 0xED) {
                                # replace phi glyph
                                substr($row{table}, 0xED - $row{offset}, 1) = 'ϕ';
                        }
@@ -319,11 +321,11 @@ sub tabinput {
                        }
                }
                elsif ($row{set} eq 'symbol') {
-                       if ($row{offset} <= 0x60 and $row{endpoint} >= 0x60 - $row{offset}) {
+                       if ($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 - $row{offset}) {
+                       if ($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
@@ -334,6 +336,8 @@ sub tabinput {
                        }
                }
 
+               $row{endpoint} -= $row{offset};
+
                $visible->{ascii} =  # assume common base
                $visible->{ $row{set} } = 1;
        }
@@ -359,7 +363,7 @@ sub range_cell {
        $len /= $colsize;
        $name //= $len <= 2 ? 'res' : 'reserved';
 
-       if (my $part = $offset/$colsize % $cols) {
+       if (my $part = ($offset/$colsize - $info->{startpoint}) % $cols) {
                # continued row
                my $rest = $cols - $part;  # remaining
                $rest = $len if $len < $rest; #TODO: optimise
@@ -419,7 +423,7 @@ for my $row (@request) {
        my $coldigits = ceil(log($colsize * $cols) / log(16));  # uniform length of hexadecimal header
        my $rowdiv = 16 ** $coldigits;  # row divide for column digits
        $rowdiv = 1 if $rowdiv != $cols * $colsize;  # divide only if all columns are matched
-       my $offset = $row->{startpoint} || 0;
+       my $offset = $row->{startpoint} * $colsize || 0;
 
        printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
        my $title = $row->{set};
@@ -433,7 +437,7 @@ for my $row (@request) {
                print "\n";
        }
        print '<tbody>';
-       while ($offset < $row->{endpoint}) {
+       while ($offset < $row->{endpoint} * $colsize) {
                if ($row->{skip}->{$offset}) {
                        $offset += $cols * $colsize;
                        next;