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 && "/$a?compare"), $a eq $Request);
55 $tablist[0] eq 'default' ? () : ('' => 'common'),
60 norteur => 'North European',
72 use Shiar_Sheet::FormatChar;
73 my $glyphs = Shiar_Sheet::FormatChar->new;
77 # generate character table(s)
78 my $input = shift or return;
81 default => [qw( u+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
82 unicode => [qw( uu+cols=32+realsize u+0-4095 u+4096-6319 u+6320-8191 )],
83 us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
84 ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
85 iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
86 dos => [qw( cp437 cp865 cp861 cp860 cp863 cp850 cp857 cp852 cp775
87 cp737 cp869 cp866 cp855 cp862 cp864 )],
88 aix => [qw( cp1006 )],
89 win => [qw( cp1252 cp1250 cp1254 cp1257 cp1258 cp1253 cp1251 cp1255 cp1256 cp874 )],
90 mac => [qw( MacRoman MacRomanian MacRumanian MacCroatian MacCentralEurRoman MacTurkish MacIcelandic MacSami
91 MacGreek MacCyrillic MacHebrew MacArabic MacFarsi MacThai )],
94 westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )],
95 centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB
96 turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )],
97 norteur => [qw( baltic nordic )],
98 baltic => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
99 nordic => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
100 cyrillic => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855
101 U4 U2DE-2DF UA64-A69 U50-52 )], # MacUkrainian is broken
102 arabic => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi
103 U6 U8A-8F+0-31+64 U75-77 )],
104 greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 U37-3F U1F )],
105 hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 U59-5F )],
106 thai => [qw( iso-8859-11 cp874 MacThai )],
107 vietnamese => [qw( viscii cp1258 MacVietnamese )],
108 symbols => [qw( symbol dingbats MacDingbats )],
123 if (my $follow = $ALIAS->{$input}) {
124 tabinput($_) for ref $follow ? @{$follow} : $follow;
128 state $visible = {'' => 1}; # all present tables
129 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'],
136 'dingbats' => ['' => '32-127+160'],
137 'MacDingbats' => ['dingbats' => '128-159'],
138 'AdobeZdingbat'=> ['MacDingbats' => '128-159'], # should be identical but maps to private use
139 'symbol' => ['' => '32-127+160'],
140 'AdobeSymbol' => ['symbol' => '32-127+160', '' => '32-127+160'], # minor differences, irrelevant except for different '€'
142 'iso-8859-2' => ['iso-8859-1' => '160'],
143 'iso-8859-3' => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9
144 'iso-8859-4' => ['iso-8859-2' => '160'],
145 'iso-8859-5' => ['iso-8859-1' => '160'],
146 'iso-8859-6' => ['cp1256' => '128', 'iso-8859-1' => '160'],
147 'iso-8859-7' => ['iso-8859-1' => '160'],
148 'iso-8859-8' => ['iso-8859-1' => '160'],
149 'iso-8859-9' => ['iso-8859-1' => '208-223+240'],
150 'iso-8859-10' => ['iso-8859-4' => '160'],
151 'iso-8859-11' => ['iso-8859-1' => '160'],
152 'iso-8859-13' => ['iso-8859-4' => '160'],
153 'iso-8859-14' => ['iso-8859-1' => '160'],
154 'iso-8859-15' => ['iso-8859-1' => '160-191'],
155 'iso-8859-16' => ['iso-8859-2' => '160'],
156 'hp-roman8' => ['iso-8859-1' => '160'],
158 'cp1252' => ['iso-8859-1' => '128-159'],
159 'cp1250' => ['iso-8859-2' => '128-191', 'cp1252' => '128'],
160 'cp1254' => ['iso-8859-9' => '128-159', 'cp1252' => '128-159+208'],
161 'cp874' => ['iso-8859-11' => '128-159', 'cp1252' => '128'], # windows-874 actually cp1162
162 'cp1257' => ['iso-8859-13' => '128-159+255', 'cp1252' => '128'],
163 'cp1251' => ['cp1252' => '128'],
164 'cp1253' => ['cp1252' => '128'],
165 'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'],
166 'cp1256' => ['cp1252' => '128'],
167 'cp1258' => ['cp1252' => '128-159+192'],
169 'cp850' => ['cp437' => '144'],
170 'cp860' => ['cp437' => '128-175'],
171 'cp861' => ['cp865' => '128-175'],
172 'cp863' => ['cp437' => '128-175'],
173 'cp865' => ['cp437' => '144-175'],
174 'cp852' => ['cp850' => '128', 'cp437' => '128'],
175 'cp857' => ['cp850' => '128-175+208-239', 'cp437' => '128'],
176 'cp775' => ['cp850' => '128'], # partial cp437
177 'cp866' => ['cp437' => '128-175+224'],
178 'cp855' => ['cp437' => '128'],
179 'cp1006' => ['iso-8859-6' => '160', 'cp437' => '128'],
180 'cp737' => ['cp437' => '128-175+224'],
181 'cp869' => ['cp437' => '128'],
182 'cp862' => ['cp437' => '128-159'],
183 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants
185 'koi8-u' => ['koi8-r' => '144-191'],
186 'koi8-f' => ['koi8-u' => '144-191'],
188 'MacRomanian' => ['MacRoman' => '160-191+208-223'],
189 'MacRumanian' => ['MacRomanian' => '160-191+208-223', 'MacRoman' => '160-191+208-223'],
190 'MacCroatian' => ['MacRoman' => '160'],
191 'MacCentralEurRoman' => ['MacRoman' => '128'],
192 'MacIcelandic'=> ['MacRoman' => '160-175+208-239'],
193 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned
194 'MacSami' => ['MacIcelandic' => '144-191+208-223+240', 'MacRoman' => '144'],
195 'MacGreek' => ['MacRoman' => '128'],
196 'MacCyrillic' => ['MacRoman' => '128'],
197 'MacHebrew' => ['iso-8859-8' => '128', 'MacRoman' => '128-143+160'], # partial ascii
198 'MacArabic' => ['iso-8859-6' => '128', 'cp864' => '128', 'MacRoman' => '128'], #TODO: multiple parents
199 'MacFarsi' => ['MacArabic' => '176-191', 'MacRoman' => '128'],
201 'cp37' => ['posix-bc' => '0'],
202 'posix-bc' => ['cp1047' => '64'],
203 'cp500' => ['cp37' => '64-95+176-191'],
204 'cp1047' => ['cp37' => '16-47+80-95+160-191'],
205 'cp1026' => ['cp37' => '64'],
206 'cp875' => ['cp37' => '48'],
210 my @parents = @{ $INHERIT->{$input} || [] };
212 if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
213 $row{parent} = $parent;
215 $params = 128 unless $visible->{$parent}
216 or ($input eq 'MacCroatian' and defined $visible->{MacRomanian});
218 elsif (defined $visible->{ascii}) {
219 $row{parent} = $parents[0];
220 $params = $parents[1] // 128;
221 $params = 128 if $params >= 128; # ascii offset at most
224 $row{parent} = $parents[0];
225 $params = $parents[1] if $parents[1] == 0; # apply ascii end
227 $visible->{$_} //= 0 for $row{parent} || ();
230 for my $param (split /[+]+/, $params // '') {
231 if ($param eq 'realsize') {
234 elsif ($param =~ m{ \A cols = (\d+) \z }x) {
237 elsif ($param =~ m{ \A (?<offset> \d+) (?: [-] (?<endpoint> \d+) )? \z }x) {
238 if (defined $row{endpoint}) {
239 # extend earlier range
240 my $skip = int(($row{endpoint} || $row{offset}) / $row{cols});
241 for ($skip + 1 .. ($+{offset} / $row{cols}) - 1) {
242 $row{skip}->{ $_ * $row{cols} - $row{offset} }++;
246 $row{offset} = $+{offset};
248 $row{endpoint} = $+{endpoint} || 0;
251 Alert("Unknown option <q>$param</q> for charset $input");
255 if ($input =~ m{ \A u ([0-9a-f]+) (?:-([0-9a-f]+))? \z }ix) {
256 my $start = hex($1) << ($2 ? 4 : 8);
257 my $end = $2 ? (hex($2) << 4) + $row{cols} - 1 : $start + 255;
258 $row{table} = join '', map { chr } $start .. $end;
259 utf8::upgrade($row{table}); # prevent latin1 output
260 $row{endpoint} = $end - $start;
261 $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
262 $row{offset} = $start % 256;
264 elsif (lc $input eq 'uu') {
265 $row{set} = 'Unicode planes';
266 $row{cell} = do 'charset-ucplanes.inc.pl'
267 or Alert('Table data could not be read', $@ || $!);
268 $row{endpoint} = 1023 * $row{cell}->{colsize};
270 elsif (lc $input eq 'u') {
271 $row{cell} = do 'charset-unicode.inc.pl'
272 or Alert('Table data could not be read', $@ || $!);
274 $row{endpoint} ||= 8191;
275 $row{endpoint} *= $row{cell}->{colsize};
276 $row{startpoint} = $row{cell}->{colsize} * $row{offset};
278 $row{set} = 'Unicode ' . (
279 $row{startpoint} < 0x10000 && $row{endpoint} < 0x10000 ? 'BMP' :
280 $row{startpoint} >= 0x10000 && $row{endpoint} < 0x20000 ? 'SMP' :
284 elsif ($input =~ m/^utf-*8$/i) {
286 $row{cell} = do 'charset-utf8.inc.pl'
287 or Alert('Table data could not be read', $@ || $!);
288 $row{endpoint} = 255;
290 elsif ($row{set} = Encode::resolve_alias($input)) {
291 $row{endpoint} ||= 255;
292 if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') {
293 # array of possibly multiple characters per code point
295 map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $row{endpoint}
299 # ~16x faster than decoding in loop;
300 # substr strings is twice as fast as splitting to an array
301 $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint});
303 $row{endpoint} -= $row{offset};
305 if ($row{set} eq 'cp437') {
306 for my $phipos (237 - $row{offset}) {
307 next if $phipos < 0 or $phipos > $row{endpoint};
309 substr($row{table}, $phipos, 1) = pack 'U*', 0x3D5;
311 if ($row{offset} == 0) {
312 # replace control characters by visible variants
313 substr($row{table}, 0, 32) = pack 'U*', map {hex} qw(
314 2007 263A 263B 2665 2666 2663 2660 2022
315 25D8 25CB 25D9 2642 2640 266A 266B 263C
316 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8
317 2191 2193 2192 2190 221F 2194 25B2 25BC
321 elsif ($row{set} eq 'symbol') {
322 if ($row{offset} <= 0x60 and $row{endpoint} >= 0x60 - $row{offset}) {
323 # replace radical extender by closest unicode equivalent
324 substr($row{table}, 0x60 - $row{offset}, 1) = '│';
326 if ($row{offset} <= 0xBD and $row{endpoint} >= 0xFF - $row{offset}) {
327 substr($row{table}, 0xBD - $row{offset}, 2) = '⏐⎯'; # arrow extenders
328 substr($row{table}, 0xD2 - $row{offset}, 3) = '®©™'; # serif variants
329 substr($row{table}, 0xE0 - $row{offset}, 1) = '◊'; # replace lookalike, should match AdobeSymbol
330 substr($row{table}, 0xE2 - $row{offset}, 3) = '®©™'; # sans-serif variants
331 substr($row{table}, 0xE6 - $row{offset}, 10) = '⎛⎜⎝⎡⎢⎣⎧⎨⎩⎪';
332 substr($row{table}, 0xF0 - $row{offset}, 1) = '€';
333 substr($row{table}, 0xF4 - $row{offset}, 11) = '⎮⌡⎞⎟⎠⎤⎥⎦⎫⎬⎭';
337 $visible->{ascii} = # assume common base
338 $visible->{ $row{set} } = 1;
341 Alert("Encoding <q>$input</q> unknown");
344 push @request, \%row;
346 tabinput($_) for @tablist;
348 my $NOCHAR = chr 0xFFFD;
351 my ($info, $offset) = @_;
352 my $table = $info->{cell} or return;
353 my $def = $table->{$offset} or return;
354 my ($len, $class, $name, $title) = @{$def};
356 my $cols = $info->{cols};
357 my $colsize = $table->{colsize} || 1;
360 $name //= $len <= 2 ? 'res' : 'reserved';
362 if (my $part = $offset/$colsize % $cols) {
364 my $rest = $cols - $part; # remaining
365 $rest = $len if $len < $rest; #TODO: optimise
367 # continued on new row
368 my @next = ($len * $colsize, "$class joinu");
369 my $separate = $cols - $len > $rest; # columns not on next row
372 push @next, $name, $title;
374 $name = $separate && '…';
377 # minority on next row
378 push @next, $separate && '"', $title || $name;
380 $table->{$offset + $colsize*$rest} //= \@next;
385 elsif (my $rows = int($len / $cols)) {
387 my $rowsize = $colsize * $cols;
388 if ($len -= $rows * $cols) {
389 # partial row remains
390 $table->{$offset + $rowsize * $rows} //= [$len*$colsize, "$class joinu", '', $title];
394 unless ($info->{realsize}) {
395 # coalesce multiple rows
397 $info->{skip}->{$offset += $rowsize}++;
401 $info->{skip}->{$offset += $rowsize} = 0;
405 $attr .= sprintf ' rowspan=%d', $rows;
409 $attr .= sprintf ' colspan=%d', $len unless $len == 1;
410 $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
411 $attr .= sprintf ' class="%s"', $class if $class;
412 $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
413 return "<td$attr>$name\n";
416 for my $row (@request) {
417 my $cols = $row->{cols};
418 my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1;
419 my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header
420 my $rowdiv = 16 ** $coldigits; # row divide for column digits
421 $rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched
422 my $offset = $row->{startpoint} || 0;
424 printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
425 my $title = $row->{set};
426 $title .= " <aside>(over $_)</aside>"
427 for grep { $_ ne 'iso-8859-1' } $row->{parent} // ();
428 printf '<caption>%s</caption>', $title;
429 print '<col>' x ($cols + 1);
430 for my $section (qw{thead}) {
431 print "<$section><tr><th>", $rowdiv == 1 ? '+' : '↱';
432 printf '<th>%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
436 while ($offset < $row->{endpoint}) {
437 if ($row->{skip}->{$offset}) {
438 $offset += $cols * $colsize;
443 if (defined $row->{skip}->{$offset}) {
447 if (my $rowmod = $offset % $rowdiv) {
448 # offset in column units
449 printf '<small>+%X</small>', $rowmod;
453 printf '%X', ($offset + $row->{offset}) / $rowdiv;
460 print range_cell($row, $offset);
464 my $cp = $offset + $row->{offset};
465 my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
466 substr $row->{table}, $offset, 1;
467 my ($cell, $name, $class) = $glyph eq $NOCHAR ? () :
468 $glyphs->glyph_html($glyph);
470 if (exists $get{compare}) {
473 $cp == ord $glyph ? 'l4' :
474 $row->{parent} && $glyph eq
475 Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
477 $visible->{$glyph} ? 'l2' :
480 $visible->{$glyph}++;
483 say sprintf $class ? '<td title="%s" class="X %s">%s' : '<td title="%s">',
484 $name, $class, $cell;
490 say '</table></div>';
497 <table class="glyphs"><tr><: if (exists $get{compare}) { :>
498 <td class="X l4">unicode
499 <td class="X l3">inherited
500 <td class="X l2">existing
501 <td class="X l1">original
502 <td class="">unassigned
504 <td class="X Cc">control
505 <td class="X Zs"><span>whitespace</span>
506 <td class="X Mn">diacritic<table class="glyphs"><tr>
507 <td class="X Sk">letter
509 <td class="X Po">punctuation<table class="glyphs"><tr>
510 <td class="X Pf">quote
512 <td class="X So">symbol<table class="glyphs"><tr>
513 <td class="X Sm">math
514 <td class="X Sc">currency
516 <td class="X No">numeric
517 <td class="X Greek">greek<table class="glyphs"><tr>
518 <td class="X Latin">latin
519 <td class="X Cyrillic">cyrillic
521 <td class="X Aramaic">aramaic<table class="glyphs"><tr>
522 <td class="X Brahmic">brahmic
523 <td class="X Arabic">arabic
525 <td class="X Syllabic">syllabic<table class="glyphs"><tr>
526 <td class="X African">african
527 <td class="X Hiragana">japanese
528 <td class="X Han">cjk
529 <td class="X Bopomofo">chinese
531 <td class="X Alpha">alphabetic
534 <table class="glyphs"><tr>
535 <td class="X">unicode 7.0
536 <td class="X Xr">proposed
537 <td class="X Xd">deprecated
538 <td class="">unassigned
539 <td class="X Xi">invalid