Character encodings

<: if ($tablist[0] eq 'default') { say "Overview of Unicode allocation and common latin code pages."; say "Compare alternate charsets:"; } else { say "Charset comparison:"; } sub optionlink { my ($title, $href, $selected) = @_; return sprintf( $selected ? '%s' : '%s', EscapeHTML($title), $href ); } print join " •\n", ( map { join " ·\n", pairmap { optionlink($b || ucfirst $a, '/charset'.($a && "/$a?compare"), $a eq $Request); } @{$_} } [ iso => 'ISO', win => 'Windows', dos => 'DOS', mac => 'Apple', ebcdic => 'EBCDIC', $tablist[0] eq 'default' ? () : ('' => 'common'), ], [ westeur => 'West', centeur => 'Central', norteur => 'North European', turkish => 0, greek => 0, cyrillic => 0, hebrew => 0, ], ); :>.

\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} }++; } } else { $row{offset} = $+{offset}; } $row{endpoint} = $+{endpoint} || 0; } else { Alert("Unknown option $param for charset $input"); } } if (lc $input eq 'wingdings') { require Encode::Wingdings; } if ($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 .. $end; utf8::upgrade($row{table}); # prevent latin1 output $row{endpoint} = $end - $start; $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8; $row{offset} = $start % 256; } 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 (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 && $row{endpoint} < 0x20000 ? 'SMP' : 'allocations' ); } elsif ($input =~ m/^utf-*8$/i) { $row{set} = 'UTF-8'; $row{cell} = do 'charset-utf8.inc.pl' or Alert('Table data could not be read', $@ || $!); $row{endpoint} = 255; } elsif ($row{set} = Encode::resolve_alias($input)) { $row{endpoint} ||= 255; if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') { # array of possibly multiple characters per code point $row{table} = [ map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $row{endpoint} ]; } else { # ~16x faster than decoding in loop; # 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 - $row{offset}) { # replace phi glyph substr($row{table}, 0xED - $row{offset}, 1) = 'ϕ'; } if ($row{offset} < 0x20) { # replace control characters by visible variants 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) = '⎮⌡⎞⎟⎠⎤⎥⎦⎫⎬⎭'; } } $visible->{ascii} = # assume common base $visible->{ $row{set} } = 1; } else { Alert("Encoding $input unknown"); return; } push @request, \%row; } tabinput($_) for @tablist; my $NOCHAR = chr 0xFFFD; sub range_cell { my ($info, $offset) = @_; my $table = $info->{cell} or return; my $def = $table->{$offset} or return; my ($len, $class, $name, $title) = @{$def}; my $cols = $info->{cols}; my $colsize = $table->{colsize} || 1; my $attr = ''; $len /= $colsize; $name //= $len <= 2 ? 'res' : 'reserved'; if (my $part = $offset/$colsize % $cols) { # continued row my $rest = $cols - $part; # remaining $rest = $len if $len < $rest; #TODO: optimise if ($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 = $separate && '…'; } else { # minority on next row push @next, $separate && '"', $title || $name; } $table->{$offset + $colsize*$rest} //= \@next; $class .= ' joind'; } $len = $rest; } elsif (my $rows = int($len / $cols)) { # multiple full rows my $rowsize = $colsize * $cols; if ($len -= $rows * $cols) { # partial row remains $table->{$offset + $rowsize * $rows} //= [$len*$colsize, "$class joinu", '', $title]; $class .= ' joind'; } unless ($info->{realsize}) { # coalesce multiple rows while ($rows > 3) { $info->{skip}->{$offset += $rowsize}++; $rows--; } if ($rows > 2) { $info->{skip}->{$offset += $rowsize} = 0; } } $attr .= sprintf ' rowspan=%d', $rows; $len = $cols; } $attr .= sprintf ' colspan=%d', $len unless $len == 1; $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//; $attr .= sprintf ' class="%s"', $class if $class; $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title; return "$name\n"; } for my $row (@request) { my $cols = $row->{cols}; my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1; 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; printf '

', !$row->{cell} && ' charmap'; my $title = $row->{set}; $title .= " " for $row->{parent} || (); printf '', $title; print '' x ($cols + 1); for my $section (qw{thead}) { print "<$section>'; while ($offset < $row->{endpoint}) { if ($row->{skip}->{$offset}) { $offset += $cols * $colsize; next; } print '

%s
", $rowdiv == 1 ? '+' : '↱'; printf '	%0X', $coldigits, $_ $colsize for 0 .. $cols - 1; print "\n"; } print '
'; if (defined $row->{skip}->{$offset}) { print '⋮'; } else { if (my $rowmod = $offset % $rowdiv) { # offset in column units printf '+%X', $rowmod; } else { # divided row offset printf '%X', ($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}, $offset, 1; my ($cell, $name, $class) = $glyph eq $NOCHAR ? () : $glyphs->glyph_html($glyph); if (exists $get{compare}) { state $visible = {}; $class = ( $cp == ord $glyph ? 'l4' : $row->{parent} && $glyph eq Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' : !$class ? undef : $visible->{$glyph} ? 'l2' : 'l1' ); $visible->{$glyph}++; } say sprintf $class ? '	%s' : '	', $name, $class, $cell; } continue { $offset += $colsize; } } say '

'; } :>

<: if (exists $get{compare}) { :>

unicode

inherited

existing

original

unassigned <: } else { :>

control

whitespace

diacritic

letter

punctuation

quote

symbol

math

currency

numeric

greek

latin

cyrillic

aramaic

brahmic

arabic

syllabic

african

japanese

cjk

chinese

alphabetic

unicode 7.0

proposed

deprecated

unassigned

invalid <: } :>