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{/+}, $Request || 'default';
20 use List::Util qw( first pairmap pairfirst pairs );
23 <h1>Character encodings</h1>
27 if ($tablist[0] eq 'default') {
28 say "Overview of Unicode allocation and common latin code pages.";
29 say "Compare alternate charsets:";
32 say "Charset comparison:";
36 my ($title, $href, $selected) = @_;
38 $selected ? '<strong>%s</strong>' : '<a href="%2$s">%s</a>',
39 EscapeHTML($title), $href
45 join " ·\n", pairmap {
46 optionlink($b || ucfirst $a, "/charset/$a?compare", $a eq $Request);
58 norteur => 'North European',
71 use Shiar_Sheet::FormatChar;
72 my $glyphs = Shiar_Sheet::FormatChar->new;
76 # generate character table(s)
77 my $input = shift or return;
80 default => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
81 uc => [qw( U++ unicode+0-4095 unicode+4096-6319 unicode+6320-8191 )],
82 us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
83 ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
84 iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
85 dos => [qw( cp437 cp865 cp861 cp860 cp863 cp850 cp857 cp852 cp775
86 cp737 cp869 cp866 cp855 cp862 cp864 )],
87 aix => [qw( cp1006 )],
88 win => [qw( cp1252 cp1250 cp1254 cp1257 cp1258 cp1253 cp1251 cp1255 cp1256 cp874 )],
89 mac => [qw( MacRoman MacRomanian MacRumanian MacCroatian MacCentralEurRoman MacTurkish MacIcelandic MacSami
90 MacGreek MacCyrillic MacHebrew MacArabic MacFarsi MacThai )],
93 westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )],
94 centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB
95 turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )],
96 norteur => [qw( baltic nordic )],
97 baltic => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
98 nordic => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
99 cyrillic => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855
100 U4 U2DE-2DF UA64-A69 U50-52 )], # MacUkrainian is broken
101 arabic => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi
102 U6 U8A-8F+0-31+64 U75-77 )],
103 greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 U37-3F U1F )],
104 hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 U59-5F )],
105 thai => [qw( iso-8859-11 cp874 MacThai )],
106 vietnamese => [qw( viscii cp1258 MacVietnamese )],
107 symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )],
122 if (my $follow = $ALIAS->{$input}) {
123 tabinput($_) for ref $follow ? @{$follow} : $follow;
127 state $visible = {}; # all present tables
128 my %row = (offset => 0, cols => 16);
130 my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
132 if (not defined $params) {
134 'cp437' => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later
135 'gsm0338' => ['ascii' => '0-127'],
137 'iso-8859-2' => ['iso-8859-1' => '160'],
138 'iso-8859-3' => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9
139 'iso-8859-4' => ['iso-8859-2' => '160'],
140 'iso-8859-5' => ['iso-8859-1' => '160'],
141 'iso-8859-6' => ['cp1256' => '128', 'iso-8859-1' => '160'],
142 'iso-8859-7' => ['iso-8859-1' => '160'],
143 'iso-8859-8' => ['iso-8859-1' => '160'],
144 'iso-8859-9' => ['iso-8859-1' => '208-223+240'],
145 'iso-8859-10' => ['iso-8859-4' => '160'],
146 'iso-8859-11' => ['iso-8859-1' => '160'],
147 'iso-8859-13' => ['iso-8859-4' => '160'],
148 'iso-8859-14' => ['iso-8859-1' => '160'],
149 'iso-8859-15' => ['iso-8859-1' => '160-191'],
150 'iso-8859-16' => ['iso-8859-2' => '160'],
151 'hp-roman8' => ['iso-8859-1' => '160'],
153 'cp1252' => ['iso-8859-1' => '128-159'],
154 'cp1250' => ['iso-8859-2' => '128-191', 'cp1252' => '128'],
155 'cp1254' => ['iso-8859-9' => '128-159', 'cp1252' => '128-159+208'],
156 'cp874' => ['iso-8859-11' => '128-159', 'cp1252' => '128'], # windows-874 actually cp1162
157 'cp1257' => ['iso-8859-13' => '128-159+255', 'cp1252' => '128'],
158 'cp1251' => ['cp1252' => '128'],
159 'cp1253' => ['cp1252' => '128'],
160 'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'],
161 'cp1256' => ['cp1252' => '128'],
162 'cp1258' => ['cp1252' => '128-159+192'],
164 'cp850' => ['cp437' => '144'],
165 'cp860' => ['cp437' => '128-175'],
166 'cp861' => ['cp865' => '128-175'],
167 'cp863' => ['cp437' => '128-175'],
168 'cp865' => ['cp437' => '144-175'],
169 'cp852' => ['cp850' => '128', 'cp437' => '128'],
170 'cp857' => ['cp850' => '128-175+208-239', 'cp437' => '128'],
171 'cp775' => ['cp850' => '128'], # partial cp437
172 'cp866' => ['cp437' => '128-175+224'],
173 'cp855' => ['cp437' => '128'],
174 'cp1006' => ['iso-8859-6' => '160', 'cp437' => '128'],
175 'cp737' => ['cp437' => '128-175+224'],
176 'cp869' => ['cp437' => '128'],
177 'cp862' => ['cp437' => '128-159'],
178 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants
180 'koi8-u' => ['koi8-r' => 128],
181 'koi8-f' => ['koi8-u' => 128],
183 'MacRomanian' => ['MacRoman' => '160-191+208-223'],
184 'MacRumanian' => ['MacRomanian' => '160-191+208-223', 'MacRoman' => '160-191+208-223'],
185 'MacCroatian' => ['MacRoman' => '160'],
186 'MacCentralEurRoman' => ['MacRoman' => '128'],
187 'MacIcelandic'=> ['MacRoman' => '160-175+208-239'], #TODO: gaps at C/E
188 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned
189 'MacSami' => ['MacIcelandic' => '144', 'MacRoman' => '144'],
190 'MacGreek' => ['MacRoman' => '128'],
191 'MacCyrillic' => ['MacRoman' => '128'],
192 'MacHebrew' => ['iso-8859-8' => '128', 'MacRoman' => '128-143+160'], # partial ascii
193 'MacArabic' => ['iso-8859-6' => '128', 'cp864' => '128', 'MacRoman' => '128'], #TODO: multiple parents
194 'MacFarsi' => ['MacArabic' => '176-191', 'MacRoman' => '128'],
196 'cp37' => ['posix-bc' => '0'],
197 'posix-bc' => ['cp1047' => '64'],
198 'cp500' => ['cp37' => '64-95+176-191'],
199 'cp1047' => ['cp37' => '16-95+160-191'], #TODO: gap at 3/4
200 'cp1026' => ['cp37' => '64'],
201 'cp875' => ['cp37' => '48'],
205 my @parents = @{ $INHERIT->{$input} || [] };
207 if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
208 $row{parent} = $parent;
210 $params = 128 unless $visible->{$parent}
211 or ($input eq 'MacCroatian' and defined $visible->{MacRomanian});
213 elsif (defined $visible->{ascii}) {
214 $row{parent} = $parents[0];
215 $params = $parents[1] // 128;
216 $params = 128 if $params >= 128; # ascii offset at most
219 $row{parent} = $parents[0];
220 $params = $parents[1] if $parents[1] == 0; # apply ascii end
222 $visible->{$_} //= 0 for $row{parent} || ();
225 if (length $params) {
226 # manual option to double table width
227 $row{cols} *= 2 if $params =~ s/[+]\z//;
230 if (length $params) {
233 (?: (?: [-] (?<stop> \d+) )? (?: [+] (?<restart> \d+) ) )?
234 (?: [-] (?<endpoint> \d+) )? \z
236 "Unknown range parameters for $input",
237 "<q>$params</q> is not in format start(-stop)(+restart(-end))",
240 $row{offset} = $+{offset};
241 $endpoint = $+{endpoint} if $+{endpoint};
242 if (my $restart = $+{restart}) {
243 my $skip = int(($+{stop} || $row{offset}) / $row{cols});
244 for ($skip + 1 .. ($restart / $row{cols}) - 1) {
245 $row{skip}->{ $_ * $row{cols} - $row{offset} }++;
250 if ($input =~ /^U([0-9a-fA-F]+)(?:-([0-9a-fA-F]+))?/) {
251 my $start = hex($1) << ($2 ? 4 : 8);
252 my $end = $2 ? (hex($2) << 4) + $row{cols} - 1 : $start + 255;
253 $row{table} = join '', map { chr } $start .. $end;
254 utf8::upgrade($row{table}); # prevent latin1 output
255 $row{endpoint} = $end - $start;
256 $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
257 $row{offset} = $start % 256;
259 elsif ($input eq 'U') {
260 $row{set} = 'Unicode planes';
261 $row{cell} = do 'charset-ucplanes.inc.pl'
262 or Alert('Table data could not be read', $@ || $!);
263 $row{endpoint} = 1023 * $row{cell}->{colsize};
265 elsif ($row{set} = Encode::resolve_alias($input)) {
266 if ($row{set} eq 'Internal') {
267 $row{cell} = do 'charset-unicode.inc.pl'
268 or Alert('Table data could not be read', $@ || $!);
270 $row{endpoint } = $row{cell}->{colsize} * ($endpoint || 8191);
271 $row{startpoint} = $row{cell}->{colsize} * $row{offset};
273 $row{set} = 'Unicode ' . (
274 $row{startpoint} < 0x10000 && $row{endpoint} < 0x10000 ? 'BMP' :
275 $row{startpoint} >= 0x10000 && $row{endpoint} < 0x20000 ? 'SMP' :
279 elsif ($row{set} eq 'utf-8-strict') {
281 $row{cell} = do 'charset-utf8.inc.pl'
282 or Alert('Table data could not be read', $@ || $!);
283 $row{endpoint} = 255;
286 if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') {
287 # array of possibly multiple characters per code point
289 map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $endpoint
293 # ~16x faster than decoding in loop;
294 # substr strings is twice as fast as splitting to an array
295 $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
297 $row{endpoint} = $endpoint - $row{offset};
299 if ($row{set} eq 'cp437' and !$row{offset}) {
300 substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
301 substr($row{table}, 0, 32) = pack 'U*', map {hex} qw(
302 2007 263A 263B 2665 2666 2663 2660 2022
303 25D8 25CB 25D9 2642 2640 266A 266B 263C
304 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8
305 2191 2193 2192 2190 221F 2194 25B2 25BC
309 $visible->{ascii} = # assume common base
310 $visible->{ $row{set} } = 1;
314 Alert("Encoding <q>$input</q> unknown");
317 push @request, \%row;
319 tabinput($_) for @tablist;
321 my $NOCHAR = chr 0xFFFD;
324 my ($info, $offset) = @_;
325 my $table = $info->{cell} or return;
326 my $def = $table->{$offset} or return;
327 my ($len, $class, $name, $title) = @{$def};
329 my $cols = $info->{cols};
330 my $colsize = $table->{colsize} || 1;
333 $name //= $len <= 2 ? 'res' : 'reserved';
335 if (my $part = $offset/$colsize % $cols) {
337 my $rest = $cols - $part; # remaining
338 $rest = $len if $len < $rest; #TODO: optimise
340 # continued on new row
341 my @next = ($len * $colsize, "$class joinu");
344 push @next, $name, $title;
349 # minority on next row
350 push @next, '"', $title || $name;
352 $table->{$offset + $colsize*$rest} //= \@next;
357 elsif (my $rows = int($len / $cols)) {
359 my $rowsize = $colsize * $cols;
360 if ($len -= $rows * $cols) {
361 # partial row remains
362 $table->{$offset + $rowsize * $rows} //= [$len*$colsize, "$class joinu", '', $title];
366 # coalesce multiple rows
368 $info->{skip}->{$offset += $rowsize}++;
372 $info->{skip}->{$offset += $rowsize} = 0;
375 $attr .= sprintf ' rowspan=%d', $rows;
379 $attr .= sprintf ' colspan=%d', $len unless $len == 1;
380 $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
381 $attr .= sprintf ' class="%s"', $class if $class;
382 $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
383 return "<td$attr>$name";
386 for my $row (@request) {
387 my $cols = $row->{cols};
388 my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1;
389 my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header
390 my $rowdiv = 16 ** $coldigits; # row divide for column digits
391 $rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched
392 my $offset = $row->{startpoint} || 0;
394 printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
395 my $title = $row->{set};
396 $title .= " <aside>(over $_)</aside>"
397 for grep { $_ ne 'iso-8859-1' } $row->{parent} // ();
398 printf '<caption>%s</caption>', $title;
399 print '<col>' x ($cols + 1);
400 for my $section (qw{thead}) {
401 print "<$section><tr><th>", $rowdiv == 1 ? '+' : '↱';
402 printf '<th>%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
406 while ($offset < $row->{endpoint}) {
407 if ($row->{skip}->{$offset}) {
408 $offset += $cols * $colsize;
413 if (defined $row->{skip}->{$offset}) {
417 if (my $rowmod = $offset % $rowdiv) {
418 # offset in column units
419 printf '<small>+%X</small>', $rowmod;
423 printf '%X', ($offset + $row->{offset}) / $rowdiv;
428 print range_cell($row, $offset);
432 my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
433 substr $row->{table}, $offset, 1;
434 if ($glyph eq $NOCHAR) {
439 if (exists $get{compare}) {
441 my $cp = $offset + $row->{offset};
442 printf '<td class="%s" title="%3$s">%2$s',
443 $cp == ord $glyph ? 'l4' :
444 $row->{parent} && $glyph eq
445 Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
446 $visible->{$glyph} ? 'l2' :
448 $glyphs->glyph_html($glyph);
449 $visible->{$glyph}++;
453 print "\n".$glyphs->glyph_cell($glyph);
460 say '</table></div>';
467 <table class="glyphs"><tr><: if (exists $get{compare}) { :>
468 <td class="X l4">unicode
469 <td class="X l3">inherited
470 <td class="X l2">existing
471 <td class="X l1">original
472 <td class="">unassigned
474 <td class="X Cc">control
475 <td class="X Zs"><span>whitespace</span>
476 <td class="X Mn">diacritic<table class="glyphs"><tr>
477 <td class="X Sk">letter
479 <td class="X Po">punctuation<table class="glyphs"><tr>
480 <td class="X Pf">quote
482 <td class="X So">symbol<table class="glyphs"><tr>
483 <td class="X Sm">math
484 <td class="X Sc">currency
486 <td class="X No">numeric
487 <td class="X Greek">greek<table class="glyphs"><tr>
488 <td class="X Latin">latin
489 <td class="X Cyrillic">cyrillic
491 <td class="X Aramaic">aramaic<table class="glyphs"><tr>
492 <td class="X Brahmic">brahmic
493 <td class="X Arabic">arabic
495 <td class="X Syllabic">syllabic<table class="glyphs"><tr>
496 <td class="X African">african
497 <td class="X Hiragana">japanese
498 <td class="X Han">cjk
499 <td class="X Bopomofo">chinese
501 <td class="X Alpha">alphabetic
504 <table class="glyphs"><tr>
505 <td class="X">unicode 7.0
506 <td class="X Xr">proposed
507 <td class="X Xd">deprecated
508 <td class="">unassigned
509 <td class="X Xi">invalid