-<:
-use utf8;
-use strict;
-use warnings;
-use open IO => ':utf8';
-
-our $VERSION = '1.0';
-
-$header{content_type} = 'text/html; charset=utf-8';
+<(common.inc.plp)><:
+
+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'],
+});
-:><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
- "http://www.w3.org/TR/html4/loose.dtd">
-<html>
-
-<head>
-<title>charset cheat sheet</title>
-<meta http-equiv="content-type" content="utf-8">
-<link rel="stylesheet" type="text/css" media="all" href="/base.css">
-</head>
-
-<body>
+:>
<h1>Character encoding</h1>
<:
-my $diinfo = do 'digraphs.inc.pl';
-my %di = map { $diinfo->{$_}->[0] => $_ } grep { ref $diinfo->{$_} }
- keys %$diinfo;
+use POSIX qw( ceil );
+use Shiar_Sheet::FormatChar;
+my $glyphs = Shiar_Sheet::FormatChar->new;
+my $cols = 16; # columns
-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(utf-8 iso-8859-1 cp437)],
- 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)],
+# 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 = grep { defined } map {
- $_ ? (resolve_alias($_) or print("Encoding $_ unknown") && ()) : ();
+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 Alert('Table data could not be read', $@ || $!);
+ $cols *= 2;
+ }
+ elsif ($row{set} = Encode::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 Alert('Table data could not be read', $@ || $!);
+ }
+ elsif ($row{set} eq 'utf-8-strict') {
+ $row{table} = undef;
+ $row{set} = 'UTF-8';
+ $row{cell} = do 'charset-utf8.inc.pl'
+ or Alert('Table data could not be read', $@ || $!);
+ }
+ else {
+ $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
+ }
+ }
+ else {
+ Alert("Encoding <q>$input</q> unknown");
+ }
+ $row{set} ? \%row : ();
+ }
+ else {
+ ();
+ }
} map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ }
- $ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default';
-my @tables = map { $_ eq 'utf-8-strict' ? undef : decode($_, pack 'C*', 0..255) } @request;
+ $Request =~ /\w/ ? split(m{[/+\s]}, $Request) : 'default';
my $NOCHAR = chr 0xFFFD;
-for my $cp437 (grep {$request[$_] eq 'cp437'} 0 .. $#request) {
- substr($tables[$cp437], 237, 1) = pack 'U*', 0x3D5; # phi sign
- substr($tables[$cp437], 0, 32) = pack 'U*', map {hex} qw(
+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 quote {
- local $_ = shift;
- s/"/"/g;
- s/</</g;
- s/>/>/g;
- return $_;
-}
-
-sub printcell_utf8 {
- my ($value) = @_;
- if ($value <= 0x7F) {
- print '<td rowspan="8" colspan="16" class="X di-a"',
- ' title="U+0000 – U+007F">Single byte ASCII'
- if $value == 0;
- }
- elsif ($value <= 0xBF) {
- print '<td rowspan="4" colspan="16" class="X di-d"',
- '>Multi-byte continuation'
- if $value == 0x80;
- }
- elsif ($value <= 0xC1) {
- print '<td colspan="2" class="X di-b" style="border-right:none; border-bottom:none"',
- ' title="U+0000 – U+007F">(Overl.)'
- if $value == 0xC0;
- }
- elsif ($value <= 0xDF) {
- print '<td rowspan="2" colspan="14" class="X di-prop" style="border-left:none"',
- ' title="U+0080 – U+03FF">2-byte sequence start'
- if $value == 0xC2;
- print '<td rowspan="1" colspan="16" class="X di-prop" style="border-top:none"',
- ' title="U+0400 – U+07FF">'
- if $value == 0xD0;
- }
- elsif ($value <= 0xEF) {
- print '<td colspan="16" class="X di-prop"',
- ' title="U+0800 – U+FFFF">3-byte sequence start'
- if $value == 0xE0;
- }
- elsif ($value <= 0xF4) {
- print '<td colspan="5" class="X di-prop" style="border-right:none"',
- ' title="U+1·0000 – U+10·FFFF">4-byte sequence'
- if $value == 0xF0;
- }
- elsif ($value <= 0xF7) {
- print '<td colspan="3" class="X di-b" style="border-left:none"',
- ' title="U+11·0000 – U+1FF·FFFF">(Overflow)'
- if $value == 0xF5;
- }
- elsif ($value <= 0xFB) {
- print '<td colspan="4" class="X di-b"',
- ' title="U+200·0000 – U+3FFF·FFFF">5-byte'
- if $value == 0xF8;
- }
- elsif ($value <= 0xFD) {
- print '<td colspan="2" class="X di-b"',
- ' title="U+4000·0000 – 7FFFF·FFFF">6-byte'
- if $value == 0xFC;
- }
- elsif ($value <= 0xFF) {
- print '<td colspan="2" class="di-invalid">Invalid'
- if $value == 0xFE;
+sub range_cell {
+ my ($table, $offset) = @_;
+ my $def = $table->{$offset} or return;
+ my ($len, $class, $name, $title) = @{$def};
+
+ 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;
}
- else {
- print "\n".'<td class="X">?';
+ 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";
}
-print "<ul>\n";
+for my $row (@request) {
+ 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; # divider of row headers
-my @nibble = (0..9, 'A'..'F');
-for my $tablenum (0 .. $#tables) {
- print '<li><table class="glyphs">';
- printf '<caption>%s</caption>', $request[$tablenum] eq 'utf-8-strict' ? 'UTF-8' : $request[$tablenum];
- print '<col>';
+ printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
+ my $title = $row->{set};
+ $title .= " <aside>($_)</aside>" for $row->{setnote} // ();
+ printf '<caption>%s</caption>', $title;
+ print '<col>' x ($cols + 1);
for my $section (qw{thead}) {
print "<$section><tr><th>↱";
- print '<th>', $_ for @nibble;
+ printf '<th>%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
print "\n";
}
print '<tbody>';
- for my $msb (0 .. $#nibble) {
- print '<tr><th>', $nibble[$msb];
- for my $lsb (0 .. $#nibble) {
- if ($request[$tablenum] eq 'utf-8-strict') {
- printcell_utf8(($msb<<4) + $lsb);
+ for my $msb (0 .. ((length($row->{table}) || 256) - 1) / $cols) {
+ print '<tr><th>';
+ {
+ my $rowlabel = ($msb + int($row->{offset} / $cols)) * $cols * $colsize;
+ if (my $rowmod = $rowlabel % $rowdiv) {
+ # offset in column units
+ printf '<small>+%X</small>', $rowmod;
+ }
+ else {
+ # divided row offset
+ printf '%X', $rowlabel / $rowdiv;
+ }
+ }
+ for my $lsb (0 .. $cols - 1) {
+ my $val = ( ($msb * $cols) + $lsb ) * $colsize;
+ if ($row->{cell}) {
+ print range_cell($row->{cell}, $val);
next;
}
- my $glyph = substr $tables[$tablenum], ($msb<<4) + $lsb, 1;
+ my $glyph = substr $row->{table}, $val, 1;
if ($glyph eq $NOCHAR) {
print '<td>';
next;
}
- my $info = [ord $glyph];
- if (defined (my $mnem = $di{ord $glyph})) {
- $info = $diinfo->{$mnem};
- }
- my ($codepoint, $name, $prop, $script, $string) = @$info;
-
- $glyph = quote($string || $glyph);
- my $desc = sprintf 'U+%04X%s', $codepoint, $name && " ($name)";
- my @class = ('X', grep {$_} $prop, $script);
-
- $glyph = "<span>$glyph</span>" if $prop eq 'Zs';
- printf "\n".'<td class="%s" title="%s">%s',
- join(' ', @class), quote($desc), $glyph;
+ print "\n".$glyphs->glyph_cell($glyph);
}
print "\n";
}
- print "</table>\n";
+ say '</table></div>';
}
-print "</ul>\n";
-
:>
<hr>
-<p class="footer">
- <a href="http://sheet.shiar.nl/" rel="home">sheet.shiar.nl</a>/charset
- <a href="git://git.shiar.nl/sheet" rel="vcs-git" title="Git repository"><:= "v$VERSION" :></a>
- created by <a href="http://shiar.nl/" rel="author">Shiar</a> •
- <a title="Licensed under the GNU Affero General Public License, version 3" rel="copyright"
- href="http://www.fsf.org/licensing/licenses/agpl-3.0.html">AGPLv3</a> •
- last update <:
- use Time::Format qw(time_format);
- print time_format('yyyy-mm-dd', (stat $ENV{SCRIPT_FILENAME})[9]);
- :>
-</p>
-
-</html>
+<div class="legend">
+ <table class="glyphs"><tr>
+ <td class="X Cc">control
+ <td class="X Zs"><span>whitespace</span>
+ <td class="X Mn">diacritic<table class="glyphs"><tr>
+ <td class="X Sk">letter
+ </table>
+ <td class="X Po">punctuation<table class="glyphs"><tr>
+ <td class="X Pf">quote
+ </table>
+ <td class="X So">symbol<table class="glyphs"><tr>
+ <td class="X Sm">math
+ <td class="X Sc">currency
+ </table>
+ <td class="X No">numeric
+ <td class="X Greek">greek<table class="glyphs"><tr>
+ <td class="X Latin">latin
+ <td class="X Cyrillic">cyrillic
+ </table>
+ <td class="X Aramaic">aramaic<table class="glyphs"><tr>
+ <td class="X Brahmic">brahmic
+ <td class="X Arabic">arabic
+ </table>
+ <td class="X Syllabic">syllabic<table class="glyphs"><tr>
+ <td class="X African">african
+ <td class="X Hiragana">japanese
+ <td class="X Han">cjk
+ <td class="X Bopomofo">chinese
+ </table>
+ <td class="X Alpha">alphabetic
+ </table>
+
+ <table class="glyphs"><tr>
+ <td class="X">unicode 7.0
+ <td class="X Xr">proposed
+ <td class="X Xd">deprecated
+ <td class="">unassigned
+ <td class="X Xi">invalid
+ </table>
+</div>
+
git.shiar.nl / sheet.git / blobdiff