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 m{/+}, $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 $input = shift or return;
34 default => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
35 us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
36 ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
37 iso => [map {"iso-8859-$_"} 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 )],
42 mac => [qw( MacRoman MacRomanian MacRumanian MacCroatian MacCentralEurRoman MacTurkish MacIcelandic MacSami
43 MacGreek MacCyrillic MacHebrew MacArabic MacFarsi MacThai )],
46 westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )],
47 centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB
48 turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )],
49 baltic => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
50 nordic => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
51 cyrillic => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855
52 U4 U2DE-2DF UA64-A69 U50-52 )], # MacUkrainian is broken
53 arabic => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi
54 U6 U8A-8F+0-31+64 U75-77 )],
55 greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 U37-3F U1F )],
56 hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 U59-5F )],
57 thai => [qw( iso-8859-11 cp874 MacThai )],
58 vietnamese => [qw( viscii cp1258 MacVietnamese )],
59 symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )],
74 if (my $follow = $ALIAS->{$input}) {
75 tabinput($_) for ref $follow ? @{$follow} : $follow;
79 state $visible = {}; # all present tables
80 my %row = (offset => 0, cols => 16);
82 my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
84 if (not defined $params) {
85 use List::Util qw( first pairfirst pairs );
88 'cp437' => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later
89 'gsm0338' => ['ascii' => '0-127'],
91 'iso-8859-2' => ['iso-8859-1' => '160'],
92 'iso-8859-3' => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9
93 'iso-8859-4' => ['iso-8859-2' => '160'],
94 'iso-8859-5' => ['iso-8859-1' => '160'],
95 'iso-8859-6' => ['cp1256' => '128', 'iso-8859-1' => '160'],
96 'iso-8859-7' => ['iso-8859-1' => '160'],
97 'iso-8859-8' => ['iso-8859-1' => '160'],
98 'iso-8859-9' => ['iso-8859-1' => '208-223+240'],
99 'iso-8859-10' => ['iso-8859-4' => '160'],
100 'iso-8859-11' => ['iso-8859-1' => '160'],
101 'iso-8859-13' => ['iso-8859-4' => '160'],
102 'iso-8859-14' => ['iso-8859-1' => '160'],
103 'iso-8859-15' => ['iso-8859-1' => '160-191'],
104 'iso-8859-16' => ['iso-8859-2' => '160'],
105 'hp-roman8' => ['iso-8859-1' => '160'],
107 'cp1252' => ['iso-8859-1' => '128-159'],
108 'cp1250' => ['iso-8859-2' => '128-191', 'cp1252' => '128'],
109 'cp1254' => ['iso-8859-9' => '128-159', 'cp1252' => '128-159+208'],
110 'cp874' => ['iso-8859-11' => '128-159', 'cp1252' => '128'], # windows-874 actually cp1162
111 'cp1257' => ['iso-8859-13' => '128-159+255', 'cp1252' => '128'],
112 'cp1251' => ['cp1252' => '128'],
113 'cp1253' => ['cp1252' => '128'],
114 'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'],
115 'cp1256' => ['cp1252' => '128'],
116 'cp1258' => ['cp1252' => '128-159+192'],
118 'cp850' => ['cp437' => '144'],
119 'cp860' => ['cp437' => '128-175'],
120 'cp861' => ['cp865' => '128-175'],
121 'cp863' => ['cp437' => '128-175'],
122 'cp865' => ['cp437' => '144-175'],
123 'cp852' => ['cp850' => '128', 'cp437' => '128'],
124 'cp857' => ['cp850' => '128-175+208-239', 'cp437' => '128'],
125 'cp775' => ['cp850' => '128'], # partial cp437
126 'cp866' => ['cp437' => '128-175+224'],
127 'cp855' => ['cp437' => '128'],
128 'cp1006' => ['iso-8859-6' => '160', 'cp437' => '128'],
129 'cp737' => ['cp437' => '128-175+224'],
130 'cp869' => ['cp437' => '128'],
131 'cp862' => ['cp437' => '128-159'],
132 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants
134 'koi8-u' => ['koi8-r' => 128],
135 'koi8-f' => ['koi8-u' => 128],
137 'MacRomanian' => ['MacRoman' => '160-191+208-223'],
138 'MacRumanian' => ['MacRomanian' => '160-191+208-223', 'MacRoman' => '160-191+208-223'],
139 'MacCroatian' => ['MacRoman' => '160'],
140 'MacCentralEurRoman' => ['MacRoman' => '128'],
141 'MacIcelandic'=> ['MacRoman' => '160-175+208-239'], #TODO: gaps at C/E
142 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned
143 'MacSami' => ['MacIcelandic' => '144', 'MacRoman' => '144'],
144 'MacGreek' => ['MacRoman' => '128'],
145 'MacCyrillic' => ['MacRoman' => '128'],
146 'MacHebrew' => ['iso-8859-8' => '128', 'MacRoman' => '128-143+160'], # partial ascii
147 'MacArabic' => ['iso-8859-6' => '128', 'cp864' => '128', 'MacRoman' => '128'], #TODO: multiple parents
148 'MacFarsi' => ['MacArabic' => '176-191', 'MacRoman' => '128'],
150 'cp37' => ['posix-bc' => '0'],
151 'posix-bc' => ['cp1047' => '64'],
152 'cp500' => ['cp37' => '64-95+176-191'],
153 'cp1047' => ['cp37' => '16-95+160-191'], #TODO: gap at 3/4
154 'cp1026' => ['cp37' => '64'],
155 'cp875' => ['cp37' => '48'],
159 my @parents = @{ $INHERIT->{$input} || [] };
161 if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
162 $row{parent} = $parent;
164 $params = 128 unless $visible->{$parent}
165 or ($input eq 'MacCroatian' and defined $visible->{MacRomanian});
167 elsif (defined $visible->{ascii}) {
168 $row{parent} = $parents[0];
169 $params = $parents[1] // 128;
170 $params = 128 if $params >= 128; # ascii offset at most
173 $row{parent} = $parents[0];
174 $params = $parents[1] if $parents[1] == 0; # apply ascii end
176 $visible->{$_} //= 0 for $row{parent} || ();
179 # manual option to double table width
180 $row{cols} *= 2 if $params =~ s/[+]\z//;
183 if (length $params) {
186 (?: (?: [-] (?<stop> \d+) )? (?: [+] (?<restart> \d+) ) )?
187 (?: [-] (?<endpoint> \d+) )? \z
189 "Unknown range parameters for $input",
190 "<q>$params</q> is not in format start(-stop)(+restart(-end))",
193 $row{offset} = $+{offset};
194 $endpoint = $+{endpoint} if $+{endpoint};
195 if (my $restart = $+{restart}) {
196 my $skip = int(($+{stop} || $row{offset}) / $row{cols});
197 for ($skip + 1 .. ($restart / $row{cols}) - 1) {
198 $row{skip}->{ $_ * $row{cols} - $row{offset} }++;
203 if ($input =~ /^U([0-9a-fA-F]+)(?:-([0-9a-fA-F]+))?/) {
204 my $start = hex($1) << ($2 ? 4 : 8);
205 my $end = $2 ? (hex($2) << 4) + $row{cols} - 1 : $start + 255;
206 $row{table} = join '', map { chr } $start .. $end;
207 utf8::upgrade($row{table}); # prevent latin1 output
208 $row{endpoint} = $end - $start;
209 $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
210 $row{offset} = $start % 256;
212 elsif ($input eq 'U') {
213 $row{set} = 'Unicode planes';
214 $row{cell} = do 'charset-ucplanes.inc.pl'
215 or Alert('Table data could not be read', $@ || $!);
217 $row{endpoint} = 1023 * $row{cell}->{colsize};
219 elsif ($row{set} = Encode::resolve_alias($input)) {
220 if ($row{set} eq 'Internal') {
221 $row{set} = 'Unicode BMP';
222 $row{cell} = do 'charset-unicode.inc.pl'
223 or Alert('Table data could not be read', $@ || $!);
224 $row{endpoint} = ($endpoint || 8191) * $row{cell}->{colsize};
226 elsif ($row{set} eq 'utf-8-strict') {
228 $row{cell} = do 'charset-utf8.inc.pl'
229 or Alert('Table data could not be read', $@ || $!);
230 $row{endpoint} = 255;
233 if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') {
234 # array of possibly multiple characters per code point
236 map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $endpoint
240 # ~16x faster than decoding in loop;
241 # substr strings is twice as fast as splitting to an array
242 $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
244 $row{endpoint} = $endpoint - $row{offset};
246 if ($row{set} eq 'cp437' and !$row{offset}) {
247 substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
248 substr($row{table}, 0, 32) = pack 'U*', map {hex} qw(
249 2007 263A 263B 2665 2666 2663 2660 2022
250 25D8 25CB 25D9 2642 2640 266A 266B 263C
251 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8
252 2191 2193 2192 2190 221F 2194 25B2 25BC
256 $visible->{ascii} = # assume common base
257 $visible->{ $row{set} } = 1;
261 Alert("Encoding <q>$input</q> unknown");
264 push @request, \%row;
266 tabinput($_) for @tablist;
268 my $NOCHAR = chr 0xFFFD;
271 my ($info, $offset) = @_;
272 my $table = $info->{cell} or return;
273 my $def = $table->{$offset} or return;
274 my ($len, $class, $name, $title) = @{$def};
276 my $cols = $info->{cols};
277 my $colsize = $table->{colsize} || 1;
280 $name //= $len <= 2 ? 'res' : 'reserved';
282 if (my $part = $offset/$colsize % $cols) {
284 my $rest = $cols - $part; # remaining
285 $rest = $len if $len < $rest; #TODO: optimise
287 # continued on new row
288 my @next = ($len * $colsize, "$class joinu");
291 push @next, $name, $title;
296 # minority on next row
297 push @next, '"', $title || $name;
299 $table->{$offset + $colsize*$rest} //= \@next;
304 elsif (my $rows = int($len / $cols)) {
306 my $rowsize = $colsize * $cols;
307 if ($len -= $rows * $cols) {
308 # partial row remains
309 $table->{$offset + $rowsize * $rows} //= [$len*$colsize, "$class joinu", '', $title];
313 # coalesce multiple rows
315 $info->{skip}->{$offset += $rowsize}++;
319 $info->{skip}->{$offset += $rowsize} = 0;
322 $attr .= sprintf ' rowspan=%d', $rows;
326 $attr .= sprintf ' colspan=%d', $len unless $len == 1;
327 $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
328 $attr .= sprintf ' class="%s"', $class if $class;
329 $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
330 return "<td$attr>$name";
333 for my $row (@request) {
334 my $cols = $row->{cols};
335 my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1;
336 my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header
337 my $rowdiv = 16 ** $coldigits; # row divide for column digits
338 $rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched
341 printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
342 my $title = $row->{set};
343 $title .= " <aside>(over $_)</aside>"
344 for grep { $_ ne 'iso-8859-1' } $row->{parent} // ();
345 printf '<caption>%s</caption>', $title;
346 print '<col>' x ($cols + 1);
347 for my $section (qw{thead}) {
348 print "<$section><tr><th>", $rowdiv == 1 ? '+' : '↱';
349 printf '<th>%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
353 while ($offset < $row->{endpoint}) {
354 if ($row->{skip}->{$offset}) {
355 $offset += $cols * $colsize;
360 if (defined $row->{skip}->{$offset}) {
364 if (my $rowmod = $offset % $rowdiv) {
365 # offset in column units
366 printf '<small>+%X</small>', $rowmod;
370 printf '%X', ($offset + $row->{offset}) / $rowdiv;
375 print range_cell($row, $offset);
379 my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
380 substr $row->{table}, $offset, 1;
381 if ($glyph eq $NOCHAR) {
386 if (exists $get{compare}) {
388 my $cp = $offset + $row->{offset};
389 printf '<td class="%s" title="%3$s">%2$s',
390 $cp == ord $glyph ? 'l4' :
391 $row->{parent} && $glyph eq
392 Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
393 $visible->{$glyph} ? 'l2' :
395 $glyphs->glyph_html($glyph);
396 $visible->{$glyph}++;
400 print "\n".$glyphs->glyph_cell($glyph);
407 say '</table></div>';
414 <table class="glyphs"><tr><: if (exists $get{compare}) { :>
415 <td class="X l4">unicode
416 <td class="X l3">inherited
417 <td class="X l2">existing
418 <td class="X l1">original
419 <td class="">unassigned
421 <td class="X Cc">control
422 <td class="X Zs"><span>whitespace</span>
423 <td class="X Mn">diacritic<table class="glyphs"><tr>
424 <td class="X Sk">letter
426 <td class="X Po">punctuation<table class="glyphs"><tr>
427 <td class="X Pf">quote
429 <td class="X So">symbol<table class="glyphs"><tr>
430 <td class="X Sm">math
431 <td class="X Sc">currency
433 <td class="X No">numeric
434 <td class="X Greek">greek<table class="glyphs"><tr>
435 <td class="X Latin">latin
436 <td class="X Cyrillic">cyrillic
438 <td class="X Aramaic">aramaic<table class="glyphs"><tr>
439 <td class="X Brahmic">brahmic
440 <td class="X Arabic">arabic
442 <td class="X Syllabic">syllabic<table class="glyphs"><tr>
443 <td class="X African">african
444 <td class="X Hiragana">japanese
445 <td class="X Han">cjk
446 <td class="X Bopomofo">chinese
448 <td class="X Alpha">alphabetic
451 <table class="glyphs"><tr>
452 <td class="X">unicode 7.0
453 <td class="X Xr">proposed
454 <td class="X Xd">deprecated
455 <td class="">unassigned
456 <td class="X Xi">invalid