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'],
18 my @tablist = split /[^\w-]+/, $Request || 'default';
21 <h1>Character encoding</h1>
25 use Shiar_Sheet::FormatChar;
26 my $glyphs = Shiar_Sheet::FormatChar->new;
30 # generate character table(s)
31 my $input = shift or return;
34 default => [qw( unicode+0-192 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
35 us => [qw( cp437 cp863 AdobeStandardEncoding gsm0338 )],
36 ebcdic => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )],
37 iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
38 dos => [qw( cp437 cp865 cp861 cp860 cp863 cp850 cp857 cp852 cp775
39 cp737 cp869 cp866 cp855 cp862 cp864 )],
40 aix => [qw( cp1006 )],
41 win => [qw( cp1252 cp1250 cp1254 cp1257 cp1258 cp1253 cp1251 cp1255 cp1256 cp874 )],
43 westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )],
44 centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB
45 turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )],
46 baltic => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
47 nordic => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
48 cyrillic => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855 )], # MacUkrainian is broken
49 arabic => [qw( iso-8859-6 cp1006 cp864 cp1256 MacArabic )], # MacFarsi same as MacArabic?
50 greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 )],
51 hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 )],
52 thai => [qw( iso-8859-11 cp874 MacThai )],
53 vietnamese => [qw( viscii cp1258 MacVietnamese )],
54 symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )],
68 if (my $follow = $ALIAS->{$input}) {
69 tabinput($_) for ref $follow ? @{$follow} : $follow;
73 state $visible = {}; # all present tables
74 my %row = (offset => 0, cols => 16);
76 my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
78 if (not defined $params) {
79 use List::Util qw( first pairfirst pairs );
82 'cp437' => ['cp850' => 0], # ascii range overridden later
83 'gsm0338' => ['ascii' => '0-127'],
85 'iso-8859-2' => ['iso-8859-1' => '160'],
86 'iso-8859-3' => ['iso-8859-1' => '160'],
87 'iso-8859-4' => ['iso-8859-2' => '160'],
88 'iso-8859-5' => ['iso-8859-1' => '160'],
89 'iso-8859-6' => ['iso-8859-1' => '160'],
90 'iso-8859-7' => ['iso-8859-1' => '160'],
91 'iso-8859-8' => ['iso-8859-1' => '160'],
92 'iso-8859-9' => ['iso-8859-1' => '208'],
93 'iso-8859-10' => ['iso-8859-4' => '160'],
94 'iso-8859-11' => ['iso-8859-1' => '160'],
95 'iso-8859-13' => ['iso-8859-4' => '160'],
96 'iso-8859-14' => ['iso-8859-1' => '160'],
97 'iso-8859-15' => ['iso-8859-1' => '160-191'],
98 'iso-8859-16' => ['iso-8859-2' => '160'],
99 'hp-roman8' => ['iso-8859-1' => '160'],
101 'cp1252' => ['iso-8859-1' => '128-159'],
102 'cp1250' => ['iso-8859-2' => '128-191'],
103 'cp1254' => ['iso-8859-9' => '128-159'],
104 'cp874' => ['iso-8859-11' => '128-159'], # windows-874 actually cp1162
105 'cp1257' => ['iso-8859-13' => '128'],
106 'cp1251' => ['cp1252' => '128'],
107 'cp1253' => ['cp1252' => '128'],
108 'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'],
109 'cp1256' => ['cp1252' => '128'],
111 'cp850' => ['cp437' => '144'],
112 'cp860' => ['cp437' => '128-175'],
113 'cp861' => ['cp865' => '128-175'],
114 'cp863' => ['cp437' => '128-175'],
115 'cp865' => ['cp437' => '144-175'],
116 'cp852' => ['cp850' => '128', 'cp437' => '128'],
117 'cp857' => ['cp850' => '128-239', 'cp437' => '128'],
118 'cp775' => ['cp850' => '128'], # partial cp437
119 'cp866' => ['cp437' => '128'],
120 'cp855' => ['cp437' => '128'],
121 'cp1006' => ['cp437' => '128'],
122 'cp737' => ['cp437' => '128'],
123 'cp869' => ['cp437' => '128'],
124 'cp862' => ['cp437' => '128'],
125 'cp864' => ['cp437' => '128'],
127 'koi8-u' => ['koi8-r' => 128],
128 'koi8-f' => ['koi8-u' => 128],
130 'MacRomanian' => ['MacRoman' => '160'],
131 'MacCroatian' => ['MacRoman' => '160'],
132 'MacCentralEurRoman' => ['MacRoman' => '128'],
133 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned
134 'MacCyrillic' => ['MacRoman' => '128'],
135 'MacHebrew' => ['MacRoman' => '128'],
138 my @parents = @{ $INHERIT->{$input} || [] };
140 if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
141 $row{parent} = $parent;
142 $params = $visible->{$parent} ? $part : 128;
144 elsif (defined $visible->{ascii}) {
145 $row{parent} = $parents[0] || 'ascii';
146 $params = $parents[1] || 128;
147 $params = 128 if $params >= 128; # ascii offset at most
150 $row{parent} = $parents[0];
152 $visible->{$_} //= 0 for $row{parent};
155 if (defined $params and $params =~ m/^ (\d+) (-\d+)? /x) {
157 $endpoint = -$2 if $2;
160 if ($input =~ /^U([0-9a-f]+)(?:-([0-9a-f]+))?/) {
161 my $start = hex($1) << ($2 ? 4 : 8);
162 my $end = $2 ? hex($2) << 4 : $start + 240;
163 $row{table} = join '', map { chr } $start .. $end+15;
164 utf8::upgrade($row{table}); # prevent latin1 output
165 $row{len} = length $row{table};
166 $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
168 elsif ($input eq 'U') {
170 $row{set} = 'Unicode planes';
171 $row{cell} = do 'charset-ucplanes.inc.pl'
172 or Alert('Table data could not be read', $@ || $!);
175 elsif ($row{set} = Encode::resolve_alias($input)) {
176 if ($row{set} eq 'Internal') {
177 $row{len} = $endpoint < 255 ? 640 : 8192;
178 $row{set} = 'Unicode BMP';
179 $row{cell} = do 'charset-unicode.inc.pl'
180 or Alert('Table data could not be read', $@ || $!);
182 elsif ($row{set} eq 'utf-8-strict') {
185 $row{cell} = do 'charset-utf8.inc.pl'
186 or Alert('Table data could not be read', $@ || $!);
189 if ($row{set} eq 'MacHebrew') {
190 # array of possibly multiple characters per code point
192 map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $endpoint
196 # ~16x faster than decoding in loop;
197 # substr strings is twice as fast as splitting to an array
198 $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
200 $row{len} = $endpoint - $row{offset};
202 if ($row{set} eq 'cp437' and !$row{offset}) {
203 substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
204 substr($row{table}, 0, 32) = pack 'U*', map {hex} qw(
205 2007 263A 263B 2665 2666 2663 2660 2022
206 25D8 25CB 25D9 2642 2640 266A 266B 263C
207 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8
208 2191 2193 2192 2190 221F 2194 25B2 25BC
212 $visible->{ascii} = # assume common base
213 $visible->{ $row{set} } = 1;
217 Alert("Encoding <q>$input</q> unknown");
220 push @request, \%row;
222 tabinput($_) for @tablist;
224 my $NOCHAR = chr 0xFFFD;
227 my ($info, $offset) = @_;
228 my $table = $info->{cell} or return;
229 my $def = $table->{$offset} or return;
230 my ($len, $class, $name, $title) = @{$def};
232 my $cols = $info->{cols};
233 my $colsize = $table->{colsize} || 1;
236 $name //= $len <= 2 ? 'res' : 'reserved';
238 if (my $part = $offset/$colsize % $cols) {
240 my $rest = $cols - $part; # remaining
241 $rest = $len if $len < $rest; #TODO: optimise
243 # continued on new row
244 my @next = ($len * $colsize, "$class joinu");
247 push @next, $name, $title;
252 # minority on next row
253 push @next, '"', $title || $name;
255 $table->{$offset + $colsize*$rest} //= \@next;
260 elsif (my $rows = int($len / $cols)) {
262 if ($len -= $rows * $cols) {
263 # partial row remains
264 $table->{$offset + $colsize*$rows * $cols} //= [$len*$colsize, "$class joinu", '', $title];
267 $attr .= sprintf ' rowspan=%d', $rows;
271 $attr .= sprintf ' colspan=%d', $len unless $len == 1;
272 $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
273 $attr .= sprintf ' class="%s"', $class if $class;
274 $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
275 return "<td$attr>$name";
278 for my $row (@request) {
279 my $cols = $row->{cols};
280 my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1;
281 my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header
282 my $rowdiv = 16 ** $coldigits; # row divide for column digits
283 $rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched
285 my $endpoint = $offset + $row->{len} * $colsize;
287 printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
288 my $title = $row->{set};
289 $title .= " <aside>(over $_)</aside>"
290 for grep { $_ ne 'iso-8859-1' } $row->{parent} // ();
291 printf '<caption>%s</caption>', $title;
292 print '<col>' x ($cols + 1);
293 for my $section (qw{thead}) {
294 print "<$section><tr><th>", $rowdiv == 1 ? '+' : '↱';
295 printf '<th>%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
299 while ($offset < $endpoint - 1) {
302 if (my $rowmod = $offset % $rowdiv) {
303 # offset in column units
304 printf '<small>+%X</small>', $rowmod;
308 printf '%X', ($offset + $row->{offset}) / $rowdiv;
313 print range_cell($row, $offset);
317 my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
318 substr $row->{table}, $offset, 1;
319 if ($glyph eq $NOCHAR) {
324 print "\n".$glyphs->glyph_cell($glyph);
331 say '</table></div>';
338 <table class="glyphs"><tr>
339 <td class="X Cc">control
340 <td class="X Zs"><span>whitespace</span>
341 <td class="X Mn">diacritic<table class="glyphs"><tr>
342 <td class="X Sk">letter
344 <td class="X Po">punctuation<table class="glyphs"><tr>
345 <td class="X Pf">quote
347 <td class="X So">symbol<table class="glyphs"><tr>
348 <td class="X Sm">math
349 <td class="X Sc">currency
351 <td class="X No">numeric
352 <td class="X Greek">greek<table class="glyphs"><tr>
353 <td class="X Latin">latin
354 <td class="X Cyrillic">cyrillic
356 <td class="X Aramaic">aramaic<table class="glyphs"><tr>
357 <td class="X Brahmic">brahmic
358 <td class="X Arabic">arabic
360 <td class="X Syllabic">syllabic<table class="glyphs"><tr>
361 <td class="X African">african
362 <td class="X Hiragana">japanese
363 <td class="X Han">cjk
364 <td class="X Bopomofo">chinese
366 <td class="X Alpha">alphabetic
369 <table class="glyphs"><tr>
370 <td class="X">unicode 7.0
371 <td class="X Xr">proposed
372 <td class="X Xd">deprecated
373 <td class="">unassigned
374 <td class="X Xi">invalid