4 title => 'charset cheat sheet',
7 "Reference sheet with all glyphs in common character encoding tables,",
8 "and an overview of Unicode ranges and UTF-8 bytes.",
11 charset codepage unicode ascii utf8 latin glyph character encoding
12 reference common overview table
14 stylesheet => [qw'light'],
15 data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'],
19 <h1>Character encoding</h1>
22 use Shiar_Sheet::FormatChar;
23 my $glyphs = Shiar_Sheet::FormatChar->new;
24 my @nibble = (0..9, 'A'..'F');
27 use Encode qw(decode resolve_alias);
28 # generate character table(s)
29 # (~16x faster than decoding in loop;
30 # substr strings is twice as fast as splitting to an array)
32 # default => [qw(unicode utf-8 iso-8859-1 cp437 -cp1252- --iso-8859-15- -koi8-f)],
33 default => [qw(unicode- utf-8 iso-8859-1 -cp1252- --iso-8859-15- cp437 -cp850)],
34 0 => [qw(cp437 cp863)],
35 1 => [qw(iso-8859-1 cp1252 MacRoman cp850)],
36 2 => [qw(iso-8859-2 cp1250 cp852 MacCentralEurRoman MacCroatian MacRumanian)],
37 5 => [qw(koi8-f iso-8859-5 cp1251 MacCyrillic cp855 cp866)],
38 7 => [qw(iso-8859-7 cp1253 MacGreek cp737 cp869)],
39 8 => [qw(iso-8859-8 cp1255 MacHebrew cp862)],
43 my %row = (offset => 0);
45 if ($input =~ s/^--//) {
46 $row{offset} = $endpoint > 160 ? 160 : 48;
48 elsif ($input =~ s/^-//) {
49 $row{offset} = $endpoint > 128 ? 128 : 32;
51 if ($input =~ s/-$//) {
52 $endpoint = $row{offset} ? $row{offset} < 160 ? 159 : 191 : 127;
55 if ($input =~ /^U([0-9a-f]+)(?:-([0-9a-f]+))?/) {
56 my $start = hex($1) << ($2 ? 4 : 8);
57 my $end = $2 ? hex($2) << 4 : $start + 240;
58 $row{table} = join '', map { chr } $start .. $end+15;
59 utf8::upgrade($row{table}); # prevent latin1 output
60 $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
62 elsif ($input eq 'U') {
63 $row{table} = ' ' x 1024;
64 $row{set} = 'Unicode planes';
65 $row{cell} = do 'charset-ucplanes.inc.pl'
66 or printf "<p class=error>Table data could not be read: <em>%s</em>.</p>\n", $@ || $!;
67 @nibble = (map { $_.0, $_.8 } 0 .. 7);
70 elsif ($row{set} = resolve_alias($input)) {
71 if ($row{set} eq 'Internal') {
72 $row{table} = ' ' x ($endpoint < 255 ? 640 : 8192);
73 $row{set} = 'Unicode BMP';
74 $row{cell} = do 'charset-unicode.inc.pl'
75 or printf "<p class=error>Table data could not be read: <em>%s</em>.</p>\n", $@ || $!;
77 elsif ($row{set} eq 'utf-8-strict') {
80 $row{cell} = do 'charset-utf8.inc.pl'
81 or printf "<p class=error>Table data could not be read: <em>%s</em>.</p>\n", $@ || $!;
84 $row{table} = decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
88 print "<p>Encoding $input unknown</p>\n";
95 } map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ }
96 $ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default';
97 my $NOCHAR = chr 0xFFFD;
99 for my $cp437 (grep {$request[$_]->{set} eq 'cp437'} 0 .. $#request) {
100 substr($request[$cp437]->{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
101 substr($request[$cp437]->{table}, 0, 32) = pack 'U*', map {hex} qw(
102 2007 263A 263B 2665 2666 2663 2660 2022 25D8 25CB 25D9 2642 2640 266A 266B 263C
103 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 2191 2193 2192 2190 221F 2194 25B2 25BC
108 my ($table, $offset) = @_;
109 my $def = $table->{$offset} or return;
110 my ($len, $class, $name, $title) = @{$def};
114 $name //= $len <= 2 ? 'res' : 'reserved';
116 if (my $part = $offset/$nibsize % 16) {
118 my $cols = 16 - $part; # remaining
119 $cols = $len if $len < $cols; #TODO: optimise
121 # continued on new row
122 $table->{$offset + $nibsize*$cols} = [$len*$nibsize, "$class joinu", $name, $title];
128 elsif (my $rows = $len >> 4) {
130 if ($len -= $rows << 4) {
131 # partial row remains
132 $table->{$offset + $nibsize*$rows * 16} = [$len*$nibsize, "$class joinu", '', $title];
135 $attr .= sprintf ' rowspan=%d', $rows;
139 $attr .= sprintf ' colspan=%d', $len unless $len == 1;
140 $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
141 $attr .= sprintf ' class="%s"', $class if $class;
142 $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
143 return "<td$attr>$name";
146 for my $row (@request) {
147 printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
148 printf '<caption>%s</caption>', $row->{set};
150 for my $section (qw{thead}) {
151 print "<$section><tr><th>↱";
152 print '<th>', $_ for @nibble;
156 for my $msb (0 .. (length($row->{table}) || 256) - 1 >> 4) {
157 printf '<tr><th>%X', ($msb + ($row->{offset} >> 4)) * $nibsize;
158 for my $lsb (0 .. $#nibble) {
159 my $val = ( ($msb<<4) + $lsb ) * $nibsize;
161 if (ref $row->{cell} eq 'CODE') {
162 print $row->{cell}->($val);
165 print range_cell($row->{cell}, $val);
169 my $glyph = substr $row->{table}, $val, 1;
170 if ($glyph eq $NOCHAR) {
175 print "\n".$glyphs->glyph_cell($glyph);
179 print "</table></div>\n";
186 <table class="glyphs"><tr>
187 <td class="X Cc">control
188 <td class="X Zs"><span>whitespace</span>
189 <td class="X Mn">diacritic<table class="glyphs"><tr>
190 <td class="X Sk">letter
192 <td class="X Po">punctuation<table class="glyphs"><tr>
193 <td class="X Pf">quote
195 <td class="X So">symbol<table class="glyphs"><tr>
196 <td class="X Sm">math
197 <td class="X Sc">currency
199 <td class="X No">numeric
200 <td class="X Greek">greek<table class="glyphs"><tr>
201 <td class="X Latin">latin
202 <td class="X Cyrillic">cyrillic
204 <td class="X Aramaic">aramaic<table class="glyphs"><tr>
205 <td class="X Brahmic">brahmic
206 <td class="X Arabic">arabic
208 <td class="X Syllabic">syllabic<table class="glyphs"><tr>
209 <td class="X African">african
210 <td class="X Hiragana">japanese
211 <td class="X Han">cjk
212 <td class="X Bopomofo">chinese
214 <td class="X Alpha">alphabetic
217 <table class="glyphs"><tr>
218 <td class="X">unicode 7.0
219 <td class="X Xr">proposed
220 <td class="X Xd">deprecated
221 <td class="">unassigned
222 <td class="X Xi">invalid