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 @nibble = (map { $_.0, $_.8 } 0 .. 7);
69 elsif ($row{set} = resolve_alias($input)) {
70 if ($row{set} eq 'Internal') {
71 $row{table} = ' ' x ($endpoint < 255 ? 640 : 8192);
72 $row{set} = 'Unicode BMP';
73 $row{cell} = do 'charset-unicode.inc.pl';
75 elsif ($row{set} eq 'utf-8-strict') {
78 $row{cell} = do 'charset-utf8.inc.pl';
81 $row{table} = decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
85 print "<p>Encoding $input unknown</p>\n";
92 } map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ }
93 $ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default';
94 my $NOCHAR = chr 0xFFFD;
96 for my $cp437 (grep {$request[$_]->{set} eq 'cp437'} 0 .. $#request) {
97 substr($request[$cp437]->{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
98 substr($request[$cp437]->{table}, 0, 32) = pack 'U*', map {hex} qw(
99 2007 263A 263B 2665 2666 2663 2660 2022 25D8 25CB 25D9 2642 2640 266A 266B 263C
100 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 2191 2193 2192 2190 221F 2194 25B2 25BC
105 my ($table, $offset) = @_;
106 my $def = $table->{$offset} or return;
107 my ($len, $class, $name, $title) = @{$def};
110 $name //= $len <= 2 ? 'res' : 'reserved';
112 if (my $part = $offset % 16) {
114 my $cols = 16 - $part; # remaining
115 $cols = $len if $len < $cols; #TODO: optimise
117 # continued on new row
118 $table->{$offset + $cols} = [$len, "$class joinu", $name, $title];
124 elsif (my $rows = $len >> 4) {
126 if ($len -= $rows << 4) {
127 # partial row remains
128 $table->{$offset + $rows * 16} = [$len, "$class joinu", '', $title];
131 $attr .= sprintf ' rowspan=%d', $rows;
135 $attr .= sprintf ' colspan=%d', $len unless $len == 1;
136 $attr .= sprintf ' class="%s"', $class if $class;
137 $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
138 return "<td$attr>$name";
141 for my $row (@request) {
142 printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
143 printf '<caption>%s</caption>', $row->{set};
145 for my $section (qw{thead}) {
146 print "<$section><tr><th>↱";
147 print '<th>', $_ for @nibble;
151 for my $msb (0 .. (length($row->{table}) || 256) - 1 >> 4) {
152 printf '<tr><th>%X', $msb + ($row->{offset} >> 4);
153 for my $lsb (0 .. $#nibble) {
154 my $val = ( ($msb<<4) + $lsb ) * $nibsize;
156 if (ref $row->{cell} eq 'CODE') {
157 print $row->{cell}->($val);
160 print range_cell($row->{cell}, $val);
164 my $glyph = substr $row->{table}, $val, 1;
165 if ($glyph eq $NOCHAR) {
170 print "\n".$glyphs->glyph_cell($glyph);
174 print "</table></div>\n";
181 <table class="glyphs"><tr>
182 <td class="X Cc">control
183 <td class="X Zs"><span>whitespace</span>
184 <td class="X Mn">diacritic<table class="glyphs"><tr>
185 <td class="X Sk">letter
187 <td class="X Po">punctuation<table class="glyphs"><tr>
188 <td class="X Pf">quote
190 <td class="X So">symbol<table class="glyphs"><tr>
191 <td class="X Sm">math
192 <td class="X Sc">currency
194 <td class="X No">numeric
195 <td class="X Greek">greek<table class="glyphs"><tr>
196 <td class="X Latin">latin
197 <td class="X Cyrillic">cyrillic
199 <td class="X Aramaic">aramaic<table class="glyphs"><tr>
200 <td class="X Brahmic">brahmic
201 <td class="X Arabic">arabic
203 <td class="X Syllabic">syllabic<table class="glyphs"><tr>
204 <td class="X African">african
205 <td class="X Hiragana">japanese
206 <td class="X Han">cjk
207 <td class="X Bopomofo">chinese
209 <td class="X Alpha">alphabetic
212 <table class="glyphs"><tr>
213 <td class="X">unicode 7.0
214 <td class="X Xr">proposed
215 <td class="X Xd">deprecated
216 <td class="">unassigned
217 <td class="X Xi">invalid