Html({
title => 'charset cheat sheet',
- version => 'v1.0',
+ version => '1.0',
description => [
"Reference sheet with all glyphs in common character encoding tables,",
"and an overview of Unicode ranges and UTF-8 bytes.",
data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'],
});
+my @tablist = split m{/+}, $Request || 'default';
+
:>
<h1>Character encoding</h1>
<:
+use POSIX qw( ceil );
use Shiar_Sheet::FormatChar;
my $glyphs = Shiar_Sheet::FormatChar->new;
+my @request;
+
+sub tabinput {
+ # generate character table(s)
+ my $input = shift or return;
+
+ state $ALIAS = {
+ default => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
+ us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
+ ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
+ iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
+ dos => [qw( cp437 cp865 cp861 cp860 cp863 cp850 cp857 cp852 cp775
+ cp737 cp869 cp866 cp855 cp862 cp864 )],
+ aix => [qw( cp1006 )],
+ win => [qw( cp1252 cp1250 cp1254 cp1257 cp1258 cp1253 cp1251 cp1255 cp1256 cp874 )],
+ mac => [qw( MacRoman MacRomanian MacRumanian MacCroatian MacCentralEurRoman MacTurkish MacIcelandic MacSami
+ MacGreek MacCyrillic MacHebrew MacArabic MacFarsi MacThai )],
+
+ # languages
+ westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )],
+ centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB
+ turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )],
+ baltic => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
+ nordic => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
+ cyrillic => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855 )], # MacUkrainian is broken
+ arabic => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi )],
+ greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 )],
+ hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 )],
+ thai => [qw( iso-8859-11 cp874 MacThai )],
+ vietnamese => [qw( viscii cp1258 MacVietnamese )],
+ symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )],
+
+ # iso-code shorthand
+ 1 => 'westeur',
+ 2 => 'centeur',
+ 3 => 'turkish',
+ 4 => 'baltic',
+ 5 => 'cyrillic',
+ 6 => 'arabic',
+ 7 => 'greek',
+ 8 => 'hebrew',
+ 9 => 'turkish',
+ 10 => 'nordic',
+ 11 => 'thai',
+ };
+ if (my $follow = $ALIAS->{$input}) {
+ tabinput($_) for ref $follow ? @{$follow} : $follow;
+ return;
+ }
+
+ state $visible = {}; # all present tables
+ my %row = (offset => 0, cols => 16);
+ my $endpoint = 255;
+ my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
+
+ if (not defined $params) {
+ use List::Util qw( first pairfirst pairs );
+
+ state $INHERIT = {
+ 'cp437' => ['cp850' => 0], # ascii range overridden later
+ 'gsm0338' => ['ascii' => '0-127'],
+
+ 'iso-8859-2' => ['iso-8859-1' => '160'],
+ 'iso-8859-3' => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9
+ 'iso-8859-4' => ['iso-8859-2' => '160'],
+ 'iso-8859-5' => ['iso-8859-1' => '160'],
+ 'iso-8859-6' => ['cp1256' => '128', 'iso-8859-1' => '160'],
+ 'iso-8859-7' => ['iso-8859-1' => '160'],
+ 'iso-8859-8' => ['iso-8859-1' => '160'],
+ 'iso-8859-9' => ['iso-8859-1' => '208-223+240'],
+ 'iso-8859-10' => ['iso-8859-4' => '160'],
+ 'iso-8859-11' => ['iso-8859-1' => '160'],
+ 'iso-8859-13' => ['iso-8859-4' => '160'],
+ 'iso-8859-14' => ['iso-8859-1' => '160'],
+ 'iso-8859-15' => ['iso-8859-1' => '160-191'],
+ 'iso-8859-16' => ['iso-8859-2' => '160'],
+ 'hp-roman8' => ['iso-8859-1' => '160'],
+
+ 'cp1252' => ['iso-8859-1' => '128-159'],
+ 'cp1250' => ['iso-8859-2' => '128-191', 'cp1252' => '128'],
+ 'cp1254' => ['iso-8859-9' => '128-159', 'cp1252' => '128-159+208'],
+ 'cp874' => ['iso-8859-11' => '128-159', 'cp1252' => '128'], # windows-874 actually cp1162
+ 'cp1257' => ['iso-8859-13' => '128-159+255', 'cp1252' => '128'],
+ 'cp1251' => ['cp1252' => '128'],
+ 'cp1253' => ['cp1252' => '128'],
+ 'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'],
+ 'cp1256' => ['cp1252' => '128'],
+ 'cp1258' => ['cp1252' => '128-159+192'],
+
+ 'cp850' => ['cp437' => '144'],
+ 'cp860' => ['cp437' => '128-175'],
+ 'cp861' => ['cp865' => '128-175'],
+ 'cp863' => ['cp437' => '128-175'],
+ 'cp865' => ['cp437' => '144-175'],
+ 'cp852' => ['cp850' => '128', 'cp437' => '128'],
+ 'cp857' => ['cp850' => '128-175+208-239', 'cp437' => '128'],
+ 'cp775' => ['cp850' => '128'], # partial cp437
+ 'cp866' => ['cp437' => '128-175+224'],
+ 'cp855' => ['cp437' => '128'],
+ 'cp1006' => ['iso-8859-6' => '160', 'cp437' => '128'],
+ 'cp737' => ['cp437' => '128-175+224'],
+ 'cp869' => ['cp437' => '128'],
+ 'cp862' => ['cp437' => '128-159'],
+ 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants
+
+ 'koi8-u' => ['koi8-r' => 128],
+ 'koi8-f' => ['koi8-u' => 128],
+
+ 'MacRomanian' => ['MacRoman' => '160-191+208-223'],
+ 'MacRumanian' => ['MacRomanian' => '160-191+208-223', 'MacRoman' => '160-191+208-223'],
+ 'MacCroatian' => ['MacRoman' => '160'],
+ 'MacCentralEurRoman' => ['MacRoman' => '128'],
+ 'MacIcelandic'=> ['MacRoman' => '160-175+208-239'], #TODO: gaps at C/E
+ 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned
+ 'MacSami' => ['MacIcelandic' => '144', 'MacRoman' => '144'],
+ 'MacGreek' => ['MacRoman' => '128'],
+ 'MacCyrillic' => ['MacRoman' => '128'],
+ 'MacHebrew' => ['iso-8859-8' => '128', 'MacRoman' => '128-143+160'], # partial ascii
+ 'MacArabic' => ['iso-8859-6' => '128', 'cp864' => '128', 'MacRoman' => '128'], #TODO: multiple parents
+ 'MacFarsi' => ['MacArabic' => '176-191', 'MacRoman' => '128'],
+
+ 'cp37' => ['posix-bc' => '0'],
+ 'posix-bc' => ['cp1047' => '64'],
+ 'cp500' => ['cp37' => '64-95+176-191'],
+ 'cp1047' => ['cp37' => '16-95+160-191'], #TODO: gap at 3/4
+ 'cp1026' => ['cp37' => '64'],
+ 'cp875' => ['cp37' => '48'],
-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;
+ };
+
+ my @parents = @{ $INHERIT->{$input} || [] };
+
+ if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
+ $row{parent} = $parent;
+ $params = $part;
+ $params = 128 unless $visible->{$parent}
+ or ($input eq 'MacCroatian' and defined $visible->{MacRomanian});
}
- elsif ($input =~ s/^-//) {
- $row{offset} = $endpoint > 128 ? 128 : 32;
+ elsif (defined $visible->{ascii}) {
+ $row{parent} = $parents[0];
+ $params = $parents[1] || 128;
+ $params = 128 if $params >= 128; # ascii offset at most
}
- if ($input =~ s/-$//) {
- $endpoint = $row{offset} ? $row{offset} < 160 ? 159 : 191 : 127;
+ elsif (@parents) {
+ $row{parent} = $parents[0];
+ }
+ $visible->{$_} //= 0 for $row{parent};
+ }
+
+ if (defined $params) {
+ $params =~ m{
+ \A (?<offset> \d+)
+ (?: (?: [-] (?<stop> \d+) )? (?: [+] (?<restart> \d+) ) )?
+ (?: [-] (?<endpoint> \d+) )? \z
+ }x or Alert([
+ "Unknown range parameters for $input",
+ "<q>$params</q> is not in format start(-stop)(+restart(-end))",
+ ]);
+
+ $row{offset} = $+{offset};
+ $endpoint = $+{endpoint} if $+{endpoint};
+ if (my $restart = $+{restart}) {
+ my $skip = int(($+{stop} || $row{offset}) / $row{cols});
+ for ($skip + 1 .. ($restart / $row{cols}) - 1) {
+ $row{skip}->{ $_ * $row{cols} - $row{offset} }++;
+ }
}
+ }
- if ($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', $start >> 8;
+ if ($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{endpoint} = $end + 14 - $start;
+ $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
+ }
+ elsif ($input eq 'U') {
+ $row{set} = 'Unicode planes';
+ $row{cell} = do 'charset-ucplanes.inc.pl'
+ or Alert('Table data could not be read', $@ || $!);
+ $row{cols} *= 2;
+ $row{endpoint} = 1023 * $row{cell}->{colsize};
+ }
+ elsif ($row{set} = Encode::resolve_alias($input)) {
+ if ($row{set} eq 'Internal') {
+ $row{set} = 'Unicode BMP';
+ $row{cell} = do 'charset-unicode.inc.pl'
+ or Alert('Table data could not be read', $@ || $!);
+ $row{endpoint} = ($endpoint || 8191) * $row{cell}->{colsize};
}
- elsif ($input eq 'U') {
- $row{table} = ' ' x 512;
- $row{set} = 'Unicode planes';
- $row{cell} = do 'charset-ucplanes.inc.pl';
+ elsif ($row{set} eq 'utf-8-strict') {
+ $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} = resolve_alias($input)) {
- if ($row{set} eq 'Internal') {
- $row{table} = ' ' x ($endpoint < 255 ? 640 : 8192);
- $row{set} = 'Unicode BMP';
- $row{cell} = do 'charset-unicode.inc.pl';
- }
- elsif ($row{set} eq 'utf-8-strict') {
- $row{table} = undef;
- $row{set} = 'UTF-8';
- $row{cell} = do 'charset-utf8.inc.pl';
+ else {
+ if ($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} .. $endpoint
+ ];
}
else {
- $row{table} = decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
+ # ~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} .. $endpoint);
}
+ $row{endpoint} = $endpoint - $row{offset};
+
+ if ($row{set} eq 'cp437' and !$row{offset}) {
+ substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
+ 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
+ );
+ }
+
+ $visible->{ascii} = # assume common base
+ $visible->{ $row{set} } = 1;
}
- else {
- print "<p>Encoding $input unknown</p>\n";
- }
- \%row;
}
else {
- ();
+ Alert("Encoding <q>$input</q> unknown");
+ return;
}
-} map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ }
- $ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default';
+ push @request, \%row;
+}
+tabinput($_) for @tablist;
+
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[$cp437]->{table}, 0, 32) = pack 'U*', map {hex} qw(
- 2007 263A 263B 2665 2666 2663 2660 2022 25D8 25CB 25D9 2642 2640 266A 266B 263C
- 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 2191 2193 2192 2190 221F 2194 25B2 25BC
- );
+sub range_cell {
+ my ($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");
+ if ($len > $rest) {
+ # minority remains
+ push @next, $name, $title;
+ $title ||= $name;
+ $name = '';
+ }
+ else {
+ # minority on next row
+ push @next, '"', $title || $name;
+ }
+ $table->{$offset + $colsize*$rest} //= \@next;
+ $class .= ' joind';
+ }
+ $len = $rest;
+ }
+ elsif (my $rows = int($len / $cols)) {
+ # multiple full rows
+ if ($len -= $rows * $cols) {
+ # partial row remains
+ $table->{$offset + $colsize*$rows * $cols} //= [$len*$colsize, "$class joinu", '', $title];
+ $class .= ' joind';
+ }
+ $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 $title;
+ return "<td$attr>$name";
}
-my @nibble = (0..9, 'A'..'F');
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 $offset = 0;
+
printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
- printf '<caption>%s</caption>', $row->{set};
- print '<col>' x 17;
+ my $title = $row->{set};
+ $title .= " <aside>(over $_)</aside>"
+ for grep { $_ ne 'iso-8859-1' } $row->{parent} // ();
+ printf '<caption>%s</caption>', $title;
+ print '<col>' x ($cols + 1);
for my $section (qw{thead}) {
- print "<$section><tr><th>↱";
- print '<th>', $_ for @nibble;
+ print "<$section><tr><th>", $rowdiv == 1 ? '+' : '↱';
+ printf '<th>%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
print "\n";
}
print '<tbody>';
- for my $msb (0 .. (length($row->{table}) || 256) - 1 >> 4) {
- printf '<tr><th>%X', $msb + ($row->{offset} >> 4);
- for my $lsb (0 .. $#nibble) {
+ while ($offset < $row->{endpoint}) {
+ if ($row->{skip}->{$offset}) {
+ $offset += $cols * $colsize;
+ next;
+ }
+
+ print '<tr><th>';
+ {
+ if (my $rowmod = $offset % $rowdiv) {
+ # offset in column units
+ printf '<small>+%X</small>', $rowmod;
+ }
+ else {
+ # divided row offset
+ printf '%X', ($offset + $row->{offset}) / $rowdiv;
+ }
+ }
+ for (1 .. $cols) {
if ($row->{cell}) {
- print $row->{cell}->(($msb<<4) + $lsb);
+ print range_cell($row, $offset);
next;
}
- my $glyph = substr $row->{table}, ($msb<<4) + $lsb, 1;
+ my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
+ substr $row->{table}, $offset, 1;
if ($glyph eq $NOCHAR) {
print '<td>';
next;
}
+ if (exists $get{compare}) {
+ state $visible = {};
+ my $cp = $offset + $row->{offset};
+ printf '<td class="%s" title="%3$s">%2$s',
+ $cp == ord $glyph ? 'l4' :
+ $row->{parent} && $glyph eq
+ Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
+ $visible->{$glyph} ? 'l2' :
+ 'l1',
+ $glyphs->glyph_html($glyph);
+ $visible->{$glyph}++;
+ next;
+ }
+
print "\n".$glyphs->glyph_cell($glyph);
}
+ continue {
+ $offset += $colsize;
+ }
print "\n";
}
- print "</table></div>\n";
+ say '</table></div>';
}
:>
<hr>
<div class="legend">
- <table class="glyphs"><tr>
+ <table class="glyphs"><tr><: if (exists $get{compare}) { :>
+ <td class="X l4">unicode
+ <td class="X l3">inherited
+ <td class="X l2">existing
+ <td class="X l1">original
+ <td class="">unassigned
+<: } else { :>
<td class="X Cc">control
<td class="X Zs"><span>whitespace</span>
<td class="X Mn">diacritic<table class="glyphs"><tr>
<td class="X Xd">deprecated
<td class="">unassigned
<td class="X Xi">invalid
- </table>
+<: } :> </table>
</div>