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);
129 my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
131 if (not defined $params) {
133 'cp437' => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later
134 'gsm0338' => ['ascii' => '0-127'],
136 'iso-8859-2' => ['iso-8859-1' => '160'],
137 'iso-8859-3' => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9
138 'iso-8859-4' => ['iso-8859-2' => '160'],
139 'iso-8859-5' => ['iso-8859-1' => '160'],
140 'iso-8859-6' => ['cp1256' => '128', 'iso-8859-1' => '160'],
141 'iso-8859-7' => ['iso-8859-1' => '160'],
142 'iso-8859-8' => ['iso-8859-1' => '160'],
143 'iso-8859-9' => ['iso-8859-1' => '208-223+240'],
144 'iso-8859-10' => ['iso-8859-4' => '160'],
145 'iso-8859-11' => ['iso-8859-1' => '160'],
146 'iso-8859-13' => ['iso-8859-4' => '160'],
147 'iso-8859-14' => ['iso-8859-1' => '160'],
148 'iso-8859-15' => ['iso-8859-1' => '160-191'],
149 'iso-8859-16' => ['iso-8859-2' => '160'],
150 'hp-roman8' => ['iso-8859-1' => '160'],
152 'cp1252' => ['iso-8859-1' => '128-159'],
153 'cp1250' => ['iso-8859-2' => '128-191', 'cp1252' => '128'],
154 'cp1254' => ['iso-8859-9' => '128-159', 'cp1252' => '128-159+208'],
155 'cp874' => ['iso-8859-11' => '128-159', 'cp1252' => '128'], # windows-874 actually cp1162
156 'cp1257' => ['iso-8859-13' => '128-159+255', 'cp1252' => '128'],
157 'cp1251' => ['cp1252' => '128'],
158 'cp1253' => ['cp1252' => '128'],
159 'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'],
160 'cp1256' => ['cp1252' => '128'],
161 'cp1258' => ['cp1252' => '128-159+192'],
163 'cp850' => ['cp437' => '144'],
164 'cp860' => ['cp437' => '128-175'],
165 'cp861' => ['cp865' => '128-175'],
166 'cp863' => ['cp437' => '128-175'],
167 'cp865' => ['cp437' => '144-175'],
168 'cp852' => ['cp850' => '128', 'cp437' => '128'],
169 'cp857' => ['cp850' => '128-175+208-239', 'cp437' => '128'],
170 'cp775' => ['cp850' => '128'], # partial cp437
171 'cp866' => ['cp437' => '128-175+224'],
172 'cp855' => ['cp437' => '128'],
173 'cp1006' => ['iso-8859-6' => '160', 'cp437' => '128'],
174 'cp737' => ['cp437' => '128-175+224'],
175 'cp869' => ['cp437' => '128'],
176 'cp862' => ['cp437' => '128-159'],
177 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants
179 'koi8-u' => ['koi8-r' => '144-191'],
180 'koi8-f' => ['koi8-u' => '144-191'],
182 'MacRomanian' => ['MacRoman' => '160-191+208-223'],
183 'MacRumanian' => ['MacRomanian' => '160-191+208-223', 'MacRoman' => '160-191+208-223'],
184 'MacCroatian' => ['MacRoman' => '160'],
185 'MacCentralEurRoman' => ['MacRoman' => '128'],
186 'MacIcelandic'=> ['MacRoman' => '160-175+208-239'],
187 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned
188 'MacSami' => ['MacIcelandic' => '144-191+208-223+240', 'MacRoman' => '144'],
189 'MacGreek' => ['MacRoman' => '128'],
190 'MacCyrillic' => ['MacRoman' => '128'],
191 'MacHebrew' => ['iso-8859-8' => '128', 'MacRoman' => '128-143+160'], # partial ascii
192 'MacArabic' => ['iso-8859-6' => '128', 'cp864' => '128', 'MacRoman' => '128'], #TODO: multiple parents
193 'MacFarsi' => ['MacArabic' => '176-191', 'MacRoman' => '128'],
195 'cp37' => ['posix-bc' => '0'],
196 'posix-bc' => ['cp1047' => '64'],
197 'cp500' => ['cp37' => '64-95+176-191'],
198 'cp1047' => ['cp37' => '16-47+80-95+160-191'],
199 'cp1026' => ['cp37' => '64'],
200 'cp875' => ['cp37' => '48'],
204 my @parents = @{ $INHERIT->{$input} || [] };
206 if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
207 $row{parent} = $parent;
209 $params = 128 unless $visible->{$parent}
210 or ($input eq 'MacCroatian' and defined $visible->{MacRomanian});
212 elsif (defined $visible->{ascii}) {
213 $row{parent} = $parents[0];
214 $params = $parents[1] // 128;
215 $params = 128 if $params >= 128; # ascii offset at most
218 $row{parent} = $parents[0];
219 $params = $parents[1] if $parents[1] == 0; # apply ascii end
221 $visible->{$_} //= 0 for $row{parent} || ();
224 if (length $params) {
225 # manual option to double table width
226 $row{cols} *= 2 if $params =~ s/[+]\z//;
229 for my $param (split /[+]+/, $params // '') {
230 if ($param =~ m{ \A (?<offset> \d+) (?: [-] (?<endpoint> \d+) )? \z }x) {
232 my $skip = int(($row{endpoint} || $row{offset}) / $row{cols});
233 for ($skip + 1 .. ($+{offset} / $row{cols}) - 1) {
234 $row{skip}->{ $_ * $row{cols} - $row{offset} }++;
238 $row{offset} = $+{offset};
240 $row{endpoint} = $+{endpoint};
243 Alert("Unknown option <q>$param</q> for charset $input");
247 if ($input =~ /^U([0-9a-fA-F]+)(?:-([0-9a-fA-F]+))?/) {
248 my $start = hex($1) << ($2 ? 4 : 8);
249 my $end = $2 ? (hex($2) << 4) + $row{cols} - 1 : $start + 255;
250 $row{table} = join '', map { chr } $start .. $end;
251 utf8::upgrade($row{table}); # prevent latin1 output
252 $row{endpoint} = $end - $start;
253 $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
254 $row{offset} = $start % 256;
256 elsif ($input eq 'U') {
257 $row{set} = 'Unicode planes';
258 $row{cell} = do 'charset-ucplanes.inc.pl'
259 or Alert('Table data could not be read', $@ || $!);
260 $row{endpoint} = 1023 * $row{cell}->{colsize};
262 elsif ($row{set} = Encode::resolve_alias($input)) {
263 if ($row{set} eq 'Internal') {
264 $row{cell} = do 'charset-unicode.inc.pl'
265 or Alert('Table data could not be read', $@ || $!);
267 $row{endpoint} ||= 8191;
268 $row{endpoint} *= $row{cell}->{colsize};
269 $row{startpoint} = $row{cell}->{colsize} * $row{offset};
271 $row{set} = 'Unicode ' . (
272 $row{startpoint} < 0x10000 && $row{endpoint} < 0x10000 ? 'BMP' :
273 $row{startpoint} >= 0x10000 && $row{endpoint} < 0x20000 ? 'SMP' :
277 elsif ($row{set} eq 'utf-8-strict') {
279 $row{cell} = do 'charset-utf8.inc.pl'
280 or Alert('Table data could not be read', $@ || $!);
281 $row{endpoint} = 255;
284 $row{endpoint} ||= 255;
285 if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') {
286 # array of possibly multiple characters per code point
288 map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $row{endpoint}
292 # ~16x faster than decoding in loop;
293 # substr strings is twice as fast as splitting to an array
294 $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint});
296 $row{endpoint} -= $row{offset};
298 if ($row{set} eq 'cp437' and !$row{offset}) {
299 substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
300 substr($row{table}, 0, 32) = pack 'U*', map {hex} qw(
301 2007 263A 263B 2665 2666 2663 2660 2022
302 25D8 25CB 25D9 2642 2640 266A 266B 263C
303 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8
304 2191 2193 2192 2190 221F 2194 25B2 25BC
308 $visible->{ascii} = # assume common base
309 $visible->{ $row{set} } = 1;
313 Alert("Encoding <q>$input</q> unknown");
316 push @request, \%row;
318 tabinput($_) for @tablist;
320 my $NOCHAR = chr 0xFFFD;
323 my ($info, $offset) = @_;
324 my $table = $info->{cell} or return;
325 my $def = $table->{$offset} or return;
326 my ($len, $class, $name, $title) = @{$def};
328 my $cols = $info->{cols};
329 my $colsize = $table->{colsize} || 1;
332 $name //= $len <= 2 ? 'res' : 'reserved';
334 if (my $part = $offset/$colsize % $cols) {
336 my $rest = $cols - $part; # remaining
337 $rest = $len if $len < $rest; #TODO: optimise
339 # continued on new row
340 my @next = ($len * $colsize, "$class joinu");
343 push @next, $name, $title;
348 # minority on next row
349 push @next, '"', $title || $name;
351 $table->{$offset + $colsize*$rest} //= \@next;
356 elsif (my $rows = int($len / $cols)) {
358 my $rowsize = $colsize * $cols;
359 if ($len -= $rows * $cols) {
360 # partial row remains
361 $table->{$offset + $rowsize * $rows} //= [$len*$colsize, "$class joinu", '', $title];
365 # coalesce multiple rows
367 $info->{skip}->{$offset += $rowsize}++;
371 $info->{skip}->{$offset += $rowsize} = 0;
374 $attr .= sprintf ' rowspan=%d', $rows;
378 $attr .= sprintf ' colspan=%d', $len unless $len == 1;
379 $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
380 $attr .= sprintf ' class="%s"', $class if $class;
381 $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
382 return "<td$attr>$name";
385 for my $row (@request) {
386 my $cols = $row->{cols};
387 my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1;
388 my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header
389 my $rowdiv = 16 ** $coldigits; # row divide for column digits
390 $rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched
391 my $offset = $row->{startpoint} || 0;
393 printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
394 my $title = $row->{set};
395 $title .= " <aside>(over $_)</aside>"
396 for grep { $_ ne 'iso-8859-1' } $row->{parent} // ();
397 printf '<caption>%s</caption>', $title;
398 print '<col>' x ($cols + 1);
399 for my $section (qw{thead}) {
400 print "<$section><tr><th>", $rowdiv == 1 ? '+' : '↱';
401 printf '<th>%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
405 while ($offset < $row->{endpoint}) {
406 if ($row->{skip}->{$offset}) {
407 $offset += $cols * $colsize;
412 if (defined $row->{skip}->{$offset}) {
416 if (my $rowmod = $offset % $rowdiv) {
417 # offset in column units
418 printf '<small>+%X</small>', $rowmod;
422 printf '%X', ($offset + $row->{offset}) / $rowdiv;
427 print range_cell($row, $offset);
431 my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
432 substr $row->{table}, $offset, 1;
433 if ($glyph eq $NOCHAR) {
438 if (exists $get{compare}) {
440 my $cp = $offset + $row->{offset};
441 printf '<td class="%s" title="%3$s">%2$s',
442 $cp == ord $glyph ? 'l4' :
443 $row->{parent} && $glyph eq
444 Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
445 $visible->{$glyph} ? 'l2' :
447 $glyphs->glyph_html($glyph);
448 $visible->{$glyph}++;
452 print "\n".$glyphs->glyph_cell($glyph);
459 say '</table></div>';
466 <table class="glyphs"><tr><: if (exists $get{compare}) { :>
467 <td class="X l4">unicode
468 <td class="X l3">inherited
469 <td class="X l2">existing
470 <td class="X l1">original
471 <td class="">unassigned
473 <td class="X Cc">control
474 <td class="X Zs"><span>whitespace</span>
475 <td class="X Mn">diacritic<table class="glyphs"><tr>
476 <td class="X Sk">letter
478 <td class="X Po">punctuation<table class="glyphs"><tr>
479 <td class="X Pf">quote
481 <td class="X So">symbol<table class="glyphs"><tr>
482 <td class="X Sm">math
483 <td class="X Sc">currency
485 <td class="X No">numeric
486 <td class="X Greek">greek<table class="glyphs"><tr>
487 <td class="X Latin">latin
488 <td class="X Cyrillic">cyrillic
490 <td class="X Aramaic">aramaic<table class="glyphs"><tr>
491 <td class="X Brahmic">brahmic
492 <td class="X Arabic">arabic
494 <td class="X Syllabic">syllabic<table class="glyphs"><tr>
495 <td class="X African">african
496 <td class="X Hiragana">japanese
497 <td class="X Han">cjk
498 <td class="X Bopomofo">chinese
500 <td class="X Alpha">alphabetic
503 <table class="glyphs"><tr>
504 <td class="X">unicode 7.0
505 <td class="X Xr">proposed
506 <td class="X Xd">deprecated
507 <td class="">unassigned
508 <td class="X Xi">invalid