X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/fc86468387a30434587a1ec138256813bf5cf7b1..29b407ddf91b653e6e8380f73e221628a392376f:/charset.plp diff --git a/charset.plp b/charset.plp index 5955c39..6d60e27 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'; @@ -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', @@ -125,14 +125,24 @@ 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; if (not defined $params) { state $INHERIT = { + 'ascii' => ['' => '0-127'], 'cp437' => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later 'gsm0338' => ['ascii' => '0-127'], + 'dingbats' => ['' => '32-127+160'], + 'MacDingbats' => ['dingbats' => '128-159'], + 'AdobeZdingbat'=> ['MacDingbats' => '128-159'], # should be identical but maps to private use + 'symbol' => ['' => '32-127+160'], + 'AdobeSymbol' => ['symbol' => '32-127+160', '' => '32-127+160'], # minor differences, irrelevant except for different '€' + 'wingdings' => ['' => '32'], + 'wingdings2' => ['' => '32'], + 'wingdings3' => ['' => '32'], + 'webdings' => ['' => '32'], 'iso-8859-2' => ['iso-8859-1' => '160'], 'iso-8859-3' => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9 @@ -229,24 +239,28 @@ sub tabinput { elsif ($param =~ m{ \A cols = (\d+) \z }x) { $row{cols} = $1; } - elsif ($param =~ m{ \A (? \d+) (?: [-] (? \d+) )? \z }x) { + elsif ($param =~ m{ \A (? \d+) (?: [-] (? \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 $param for charset $input"); } } + if ($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; @@ -255,24 +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' ); } @@ -283,6 +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 @@ -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; @@ -339,7 +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 @@ -399,12 +423,12 @@ 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} * $colsize || 0; printf '
', !$row->{cell} && ' charmap'; my $title = $row->{set}; $title .= " " - for grep { $_ ne 'iso-8859-1' } $row->{parent} // (); + for $row->{parent} || (); printf '', $title; print '' x ($cols + 1); for my $section (qw{thead}) { @@ -413,7 +437,7 @@ for my $row (@request) { print "\n"; } print ''; - while ($offset < $row->{endpoint}) { + while ($offset < $row->{endpoint} * $colsize) { if ($row->{skip}->{$offset}) { $offset += $cols * $colsize; next;
%s