<(common.inc.plp)><:
use 5.014;
Html({
title => 'charset cheat sheet',
version => '1.0',
description => [
"Reference sheet with all glyphs in common character encoding tables,",
"and an overview of Unicode ranges and UTF-8 bytes.",
],
keywords => [qw'
charset codepage unicode ascii utf8 latin glyph character encoding
reference common overview table
'],
stylesheet => [qw'light'],
data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'],
});
:>
Character encoding
<:
use Shiar_Sheet::FormatChar;
my $glyphs = Shiar_Sheet::FormatChar->new;
my @nibble = (0..9, 'A'..'F');
my $nibsize = 1;
use Encode qw(decode resolve_alias);
# generate character table(s)
# (~16x faster than decoding in loop;
# substr strings is twice as fast as splitting to an array)
my %ALIAS = (
# default => [qw(unicode utf-8 iso-8859-1 cp437 -cp1252- --iso-8859-15- -koi8-f)],
default => [qw(unicode- utf-8 iso-8859-1 -cp1252- --iso-8859-15- cp437 -cp850)],
0 => [qw(cp437 -cp863)],
1 => [qw(iso-8859-1 -cp1252 -MacRoman -cp850)],
2 => [qw(iso-8859-2 -cp1250 -cp852 -MacCentralEurRoman -MacCroatian -MacRumanian)],
5 => [qw(koi8-f -iso-8859-5 -cp1251 -MacCyrillic -cp855 -cp866)],
7 => [qw(iso-8859-7 -cp1253 -MacGreek -cp737 -cp869)],
8 => [qw(iso-8859-8 -cp1255 -MacHebrew -cp862)],
);
my @request = map {
if (my $input = $_) {
my %row = (offset => 0);
my $endpoint = 255;
if ($input =~ s/^--//) {
$row{offset} = $endpoint > 160 ? 160 : 48;
}
elsif ($input =~ s/^-//) {
$row{offset} = $endpoint > 128 ? 128 : 32;
}
if ($input =~ s/-$//) {
$endpoint = $row{offset} ? $row{offset} < 160 ? 159 : 191 : 127;
}
if ($row{offset}) {
$row{setnote} = 'over cp437' if $input eq 'cp850';
$row{setnote} = 'over iso-8859-1' if $input =~ /^iso-8859-|^cp125/;
}
if ($input =~ /^U([0-9a-f]+)(?:-([0-9a-f]+))?/) {
my $start = hex($1) << ($2 ? 4 : 8);
my $end = $2 ? hex($2) << 4 : $start + 240;
$row{table} = join '', map { chr } $start .. $end+15;
utf8::upgrade($row{table}); # prevent latin1 output
$row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
}
elsif ($input eq 'U') {
$row{table} = ' ' x 1024;
$row{set} = 'Unicode planes';
$row{cell} = do 'charset-ucplanes.inc.pl'
or printf "Table data could not be read: %s.
\n", $@ || $!;
@nibble = (map { $_.0, $_.8 } 0 .. 7);
$nibsize = 8;
}
elsif ($row{set} = resolve_alias($input)) {
if ($row{set} eq 'Internal') {
$row{table} = ' ' x ($endpoint < 255 ? 640 : 8192);
$row{set} = 'Unicode BMP';
$row{cell} = do 'charset-unicode.inc.pl'
or printf "Table data could not be read: %s.
\n", $@ || $!;
}
elsif ($row{set} eq 'utf-8-strict') {
$row{table} = undef;
$row{set} = 'UTF-8';
$row{cell} = do 'charset-utf8.inc.pl'
or printf "Table data could not be read: %s.
\n", $@ || $!;
}
else {
$row{table} = decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
}
}
else {
say "Encoding $input
unknown
";
}
$row{set} ? \%row : ();
}
else {
();
}
} map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ }
$ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default';
my $NOCHAR = chr 0xFFFD;
for my $cp437 (grep {$request[$_]->{set} eq 'cp437'} 0 .. $#request) {
substr($request[$cp437]->{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
substr($request[$cp437]->{table}, 0, 32) = pack 'U*', map {hex} qw(
2007 263A 263B 2665 2666 2663 2660 2022 25D8 25CB 25D9 2642 2640 266A 266B 263C
25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 2191 2193 2192 2190 221F 2194 25B2 25BC
);
}
sub range_cell {
my ($table, $offset) = @_;
my $def = $table->{$offset} or return;
my ($len, $class, $name, $title) = @{$def};
my $attr = '';
$len /= $nibsize;
$name //= $len <= 2 ? 'res' : 'reserved';
if (my $part = $offset/$nibsize % 16) {
# continued row
my $cols = 16 - $part; # remaining
$cols = $len if $len < $cols; #TODO: optimise
if ($len -= $cols) {
# continued on new row
$table->{$offset + $nibsize*$cols} = [$len*$nibsize, "$class joinu", $name, $title];
$name = '';
$class .= ' joind';
}
$len = $cols;
}
elsif (my $rows = $len >> 4) {
# multiple full rows
if ($len -= $rows << 4) {
# partial row remains
$table->{$offset + $nibsize*$rows * 16} = [$len*$nibsize, "$class joinu", '', $title];
$class .= ' joind';
}
$attr .= sprintf ' rowspan=%d', $rows;
$len = 16;
}
$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";
}
for my $row (@request) {
printf '', !$row->{cell} && ' charmap';
my $title = $row->{set};
$title .= " " for $row->{setnote} // ();
printf '%s', $title;
print '' x 17;
for my $section (qw{thead}) {
print "<$section>| ↱";
print ' | ', $_ for @nibble;
print "\n";
}
print ' |
|---|
';
for my $msb (0 .. (length($row->{table}) || 256) - 1 >> 4) {
printf '| %X', ($msb + ($row->{offset} >> 4)) * $nibsize;
for my $lsb (0 .. $#nibble) {
my $val = ( ($msb<<4) + $lsb ) * $nibsize;
if ($row->{cell}) {
if (ref $row->{cell} eq 'CODE') {
print $row->{cell}->($val);
next;
}
print range_cell($row->{cell}, $val);
next;
}
my $glyph = substr $row->{table}, $val, 1;
if ($glyph eq $NOCHAR) {
print ' | ';
next;
}
print "\n".$glyphs->glyph_cell($glyph);
}
print "\n";
}
print " |
|---|
| control
| whitespace
| diacritic
| punctuation
| symbol
| numeric
| greek
| aramaic
| syllabic
| african
| japanese
| cjk
| chinese
|
| alphabetic
|
| unicode 7.0
| proposed
| deprecated
| unassigned
| invalid
|