5 title => 'charset cheat sheet',
8 "Reference sheet with all glyphs in common character encoding tables,",
9 "and an overview of Unicode ranges and UTF-8 bytes.",
12 charset codepage unicode ascii utf8 latin glyph character encoding
13 reference common overview table
15 stylesheet => [qw'light'],
16 data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'],
20 <h1>Character encoding</h1>
23 use Shiar_Sheet::FormatChar;
24 my $glyphs = Shiar_Sheet::FormatChar->new;
25 my @nibble = (0..9, 'A'..'F');
28 use Encode qw(decode resolve_alias);
29 # generate character table(s)
30 # (~16x faster than decoding in loop;
31 # substr strings is twice as fast as splitting to an array)
33 # default => [qw(unicode utf-8 iso-8859-1 cp437 -cp1252- --iso-8859-15- -koi8-f)],
34 default => [qw(unicode- utf-8 iso-8859-1 -cp1252- --iso-8859-15- cp437 -cp850)],
35 0 => [qw(cp437 -cp863)],
36 1 => [qw(iso-8859-1 -cp1252 -MacRoman -cp850)],
37 2 => [qw(iso-8859-2 -cp1250 -cp852 -MacCentralEurRoman -MacCroatian -MacRumanian)],
38 5 => [qw(koi8-f -iso-8859-5 -cp1251 -MacCyrillic -cp855 -cp866)],
39 7 => [qw(iso-8859-7 -cp1253 -MacGreek -cp737 -cp869)],
40 8 => [qw(iso-8859-8 -cp1255 -MacHebrew -cp862)],
44 my %row = (offset => 0);
46 if ($input =~ s/^--//) {
47 $row{offset} = $endpoint > 160 ? 160 : 48;
49 elsif ($input =~ s/^-//) {
50 $row{offset} = $endpoint > 128 ? 128 : 32;
52 if ($input =~ s/-$//) {
53 $endpoint = $row{offset} ? $row{offset} < 160 ? 159 : 191 : 127;
56 $row{setnote} = 'over cp437' if $input eq 'cp850';
57 $row{setnote} = 'over iso-8859-1' if $input =~ /^iso-8859-|^cp125/;
60 if ($input =~ /^U([0-9a-f]+)(?:-([0-9a-f]+))?/) {
61 my $start = hex($1) << ($2 ? 4 : 8);
62 my $end = $2 ? hex($2) << 4 : $start + 240;
63 $row{table} = join '', map { chr } $start .. $end+15;
64 utf8::upgrade($row{table}); # prevent latin1 output
65 $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
67 elsif ($input eq 'U') {
68 $row{table} = ' ' x 1024;
69 $row{set} = 'Unicode planes';
70 $row{cell} = do 'charset-ucplanes.inc.pl'
71 or printf "<p class=error>Table data could not be read: <em>%s</em>.</p>\n", $@ || $!;
72 @nibble = (map { $_.0, $_.8 } 0 .. 7);
75 elsif ($row{set} = resolve_alias($input)) {
76 if ($row{set} eq 'Internal') {
77 $row{table} = ' ' x ($endpoint < 255 ? 640 : 8192);
78 $row{set} = 'Unicode BMP';
79 $row{cell} = do 'charset-unicode.inc.pl'
80 or printf "<p class=error>Table data could not be read: <em>%s</em>.</p>\n", $@ || $!;
82 elsif ($row{set} eq 'utf-8-strict') {
85 $row{cell} = do 'charset-utf8.inc.pl'
86 or printf "<p class=error>Table data could not be read: <em>%s</em>.</p>\n", $@ || $!;
89 $row{table} = decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
93 say "<p class=error>Encoding <q>$input</q> unknown</p>";
95 $row{set} ? \%row : ();
100 } map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ }
101 $ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default';
102 my $NOCHAR = chr 0xFFFD;
104 for my $cp437 (grep {$request[$_]->{set} eq 'cp437'} 0 .. $#request) {
105 substr($request[$cp437]->{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
106 substr($request[$cp437]->{table}, 0, 32) = pack 'U*', map {hex} qw(
107 2007 263A 263B 2665 2666 2663 2660 2022 25D8 25CB 25D9 2642 2640 266A 266B 263C
108 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 2191 2193 2192 2190 221F 2194 25B2 25BC
113 my ($table, $offset) = @_;
114 my $def = $table->{$offset} or return;
115 my ($len, $class, $name, $title) = @{$def};
119 $name //= $len <= 2 ? 'res' : 'reserved';
121 if (my $part = $offset/$nibsize % 16) {
123 my $cols = 16 - $part; # remaining
124 $cols = $len if $len < $cols; #TODO: optimise
126 # continued on new row
127 $table->{$offset + $nibsize*$cols} = [$len*$nibsize, "$class joinu", $name, $title];
133 elsif (my $rows = $len >> 4) {
135 if ($len -= $rows << 4) {
136 # partial row remains
137 $table->{$offset + $nibsize*$rows * 16} = [$len*$nibsize, "$class joinu", '', $title];
140 $attr .= sprintf ' rowspan=%d', $rows;
144 $attr .= sprintf ' colspan=%d', $len unless $len == 1;
145 $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
146 $attr .= sprintf ' class="%s"', $class if $class;
147 $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
148 return "<td$attr>$name";
151 for my $row (@request) {
152 printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
153 my $title = $row->{set};
154 $title .= " <aside>($_)</aside>" for $row->{setnote} // ();
155 printf '<caption>%s</caption>', $title;
157 for my $section (qw{thead}) {
158 print "<$section><tr><th>↱";
159 print '<th>', $_ for @nibble;
163 for my $msb (0 .. (length($row->{table}) || 256) - 1 >> 4) {
164 printf '<tr><th>%X', ($msb + ($row->{offset} >> 4)) * $nibsize;
165 for my $lsb (0 .. $#nibble) {
166 my $val = ( ($msb<<4) + $lsb ) * $nibsize;
168 if (ref $row->{cell} eq 'CODE') {
169 print $row->{cell}->($val);
172 print range_cell($row->{cell}, $val);
176 my $glyph = substr $row->{table}, $val, 1;
177 if ($glyph eq $NOCHAR) {
182 print "\n".$glyphs->glyph_cell($glyph);
186 print "</table></div>\n";
193 <table class="glyphs"><tr>
194 <td class="X Cc">control
195 <td class="X Zs"><span>whitespace</span>
196 <td class="X Mn">diacritic<table class="glyphs"><tr>
197 <td class="X Sk">letter
199 <td class="X Po">punctuation<table class="glyphs"><tr>
200 <td class="X Pf">quote
202 <td class="X So">symbol<table class="glyphs"><tr>
203 <td class="X Sm">math
204 <td class="X Sc">currency
206 <td class="X No">numeric
207 <td class="X Greek">greek<table class="glyphs"><tr>
208 <td class="X Latin">latin
209 <td class="X Cyrillic">cyrillic
211 <td class="X Aramaic">aramaic<table class="glyphs"><tr>
212 <td class="X Brahmic">brahmic
213 <td class="X Arabic">arabic
215 <td class="X Syllabic">syllabic<table class="glyphs"><tr>
216 <td class="X African">african
217 <td class="X Hiragana">japanese
218 <td class="X Han">cjk
219 <td class="X Bopomofo">chinese
221 <td class="X Alpha">alphabetic
224 <table class="glyphs"><tr>
225 <td class="X">unicode 7.0
226 <td class="X Xr">proposed
227 <td class="X Xd">deprecated
228 <td class="">unassigned
229 <td class="X Xi">invalid