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'],
16 charset-encoding.inc.pl
17 charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl
21 my @tablist = split m{/+}, $Request || 'default';
23 use List::Util qw( first pairmap pairfirst pairs );
26 <h1>Character encodings</h1>
30 if ($tablist[0] eq 'default') {
31 say "Overview of Unicode allocation and common latin code pages.";
32 say "Compare alternate charsets:";
35 say "Charset comparison:";
39 my ($title, $href, $selected) = @_;
41 $selected ? '<strong>%s</strong>' : '<a href="%2$s">%s</a>',
42 EscapeHTML($title), $href
48 join " ·\n", pairmap {
49 optionlink($b || ucfirst $a, '/charset'.($a && "/$a?compare"), $a eq $Request);
58 $tablist[0] eq 'default' ? () : ('' => 'common'),
63 norteur => 'North European',
75 use Shiar_Sheet::FormatChar;
76 my $glyphs = Shiar_Sheet::FormatChar->new;
79 my $INHERIT = do 'charset-encoding.inc.pl'
80 or Alert('Encoding metadata could not be read', $@ || $!);
83 # generate character table(s)
84 my $input = shift or return;
87 default => [qw( u+0-27F utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
88 unicode => [qw( uu+cols=32+realsize u+0-FFF u+1000-18AF u+18B0-1FFF )],
89 us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
90 ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
91 iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
92 dos => [qw( cp437 cp865 cp861 cp860 cp863 cp850 cp857 cp852 cp775
93 cp737 cp869 cp866 cp855 cp862 cp864 )],
94 aix => [qw( cp1006 )],
95 win => [qw( cp1252 cp1250 cp1254 cp1257 cp1258 cp1253 cp1251 cp1255 cp1256 cp874 )],
96 mac => [qw( MacRoman MacRomanian MacRumanian MacCroatian MacCentralEurRoman MacTurkish MacIcelandic MacSami
97 MacGreek MacCyrillic MacHebrew MacArabic MacFarsi MacThai )],
100 westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )],
101 centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB
102 turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )],
103 norteur => [qw( baltic nordic )],
104 baltic => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
105 nordic => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
106 cyrillic => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855
107 +400 +2DE0 +A640-A69F +500-52F )], # MacUkrainian is broken
108 arabic => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi
109 +600 +8A0-8BF+8E0 +750-77F )],
110 greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 +370 +1F00 )],
111 hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 +590 )],
112 thai => [qw( iso-8859-11 cp874 MacThai )],
113 vietnamese => [qw( viscii cp1258 MacVietnamese )],
114 symbols => [qw( symbol dingbats MacDingbats wingdings wingdings2 wingdings3 webdings )],
129 if (my $follow = $ALIAS->{$input}) {
130 tabinput($_) for ref $follow ? @{$follow} : $follow;
134 state $visible = {'' => 1}; # all present tables
135 my %row = (offset => 0, cols => 16);
136 my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
138 if (not defined $params) {
139 my @parents = @{ $INHERIT->{$input} || [] };
141 if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
142 $row{parent} = $parent;
144 $params = 80 unless $visible->{$parent}
145 or ($input eq 'MacCroatian' and defined $visible->{MacRomanian});
147 elsif (defined $visible->{ascii}) {
148 $row{parent} = $parents[0];
149 $params = $parents[1] // 80;
150 $params = 80 if hex $params >= 0x80; # ascii offset at most
153 $row{parent} = $parents[0];
154 $params = $parents[1] if hex $parents[1] == 0; # apply ascii end
156 $visible->{$_} //= 0 for $row{parent} || ();
159 for my $param (split /[+]+/, $params // '') {
160 if ($param eq 'realsize') {
163 elsif ($param =~ m{ \A cols = (\d+) \z }x) {
166 elsif ($param =~ m{ \A (?<start> \p{AHex}+) (?: [-] (?<end> \p{AHex}+) )? \z }x) {
167 if (defined $row{endpoint}) {
168 # extend earlier range
169 my $skip = int(($row{endpoint} || $row{startpoint}) / $row{cols});
170 for ($skip + 1 .. (hex($+{start}) / $row{cols}) - 1) {
171 $row{skip}->{ $_ * $row{cols} - $row{startpoint} }++;
175 $row{startpoint} = hex $+{start};
177 $row{endpoint} = hex($+{end} || 0);
180 Alert("Unknown option <q>$param</q> for charset $input");
184 if ($input =~ m{ \A (?:wing|web)dings \d* \z }ix) {
185 eval "require Encode::\u$input";
189 $row{offset} = delete $row{startpoint};
190 $row{set} = 'Unicode characters';
191 my $block = $row{offset} >> 8;
192 $row{endpoint} ||= ($block + 1 << 8) - 1;
193 $block == ($row{endpoint} >> 8) or undef $block;
195 $row{table} = join '', map { chr } $row{offset} .. $row{endpoint};
196 utf8::upgrade($row{table}); # prevent latin1 output
198 $row{endpoint} -= $row{offset};
200 if (defined $block) {
201 $row{set} = sprintf 'Unicode block U+%02Xxx', $block;
202 $row{offset} %= 0x100;
205 elsif (lc $input eq 'uu') {
206 $row{cell} = do 'charset-ucplanes.inc.pl'
207 or Alert('Table data could not be read', $@ || $!);
208 $row{endpoint} ||= 0x3FF;
209 $row{set} = 'Unicode planes';
211 elsif (lc $input eq 'u') {
212 $row{cell} = do 'charset-unicode.inc.pl'
213 or Alert('Table data could not be read', $@ || $!);
215 $row{endpoint} ||= 0x1FFF;
216 $row{set} = 'Unicode ' . (
217 $row{startpoint} < 0x1000 && $row{endpoint} < 0x1000 ? 'BMP' :
218 $row{startpoint} >= 0x1000 && $row{endpoint} < 0x2000 ? 'SMP' :
222 elsif ($input =~ m/^utf-*8$/i) {
224 $row{cell} = do 'charset-utf8.inc.pl'
225 or Alert('Table data could not be read', $@ || $!);
226 $row{endpoint} = 0xFF;
228 elsif ($row{set} = Encode::resolve_alias($input)) {
229 $row{offset} = delete $row{startpoint};
230 $row{endpoint} ||= 0xFF;
231 if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') {
232 # array of possibly multiple characters per code point
234 map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $row{endpoint}
238 # ~16x faster than decoding in loop;
239 # substr strings is twice as fast as splitting to an array
240 $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint});
243 if ($row{set} eq 'cp437') {
244 if ($row{offset} <= 0xED and $row{endpoint} >= 0xED) {
246 substr($row{table}, 0xED - $row{offset}, 1) = 'ϕ';
248 if ($row{offset} < 0x20) {
249 # replace control characters by visible variants
250 my $sub = substr ' ☺☻♥♦♣♠•◘○◙♂♀♪♫☼►◄↕‼¶§▬↨↑↓→←∟↔▲▼', $row{offset};
251 substr($row{table}, 0, length $sub) = $sub;
254 elsif ($row{set} eq 'symbol') {
255 if ($row{offset} <= 0x60 and $row{endpoint} >= 0x60) {
256 # replace radical extender by closest unicode equivalent
257 substr($row{table}, 0x60 - $row{offset}, 1) = '│';
259 if ($row{offset} <= 0xBD and $row{endpoint} >= 0xFF) {
260 substr($row{table}, 0xBD - $row{offset}, 2) = '⏐⎯'; # arrow extenders
261 substr($row{table}, 0xD2 - $row{offset}, 3) = '®©™'; # serif variants
262 substr($row{table}, 0xE0 - $row{offset}, 1) = '◊'; # replace lookalike, should match AdobeSymbol
263 substr($row{table}, 0xE2 - $row{offset}, 3) = '®©™'; # sans-serif variants
264 substr($row{table}, 0xE6 - $row{offset}, 10) = '⎛⎜⎝⎡⎢⎣⎧⎨⎩⎪';
265 substr($row{table}, 0xF0 - $row{offset}, 1) = '€';
266 substr($row{table}, 0xF4 - $row{offset}, 11) = '⎮⌡⎞⎟⎠⎤⎥⎦⎫⎬⎭';
270 $row{endpoint} -= $row{offset};
272 $visible->{ascii} = # assume common base
273 $visible->{ $row{set} } = 1;
276 Alert("Encoding <q>$input</q> unknown");
279 push @request, \%row;
281 tabinput($_) for @tablist;
283 my $NOCHAR = chr 0xFFFD;
286 my ($info, $offset) = @_;
287 my $table = $info->{cell} or return;
288 my $def = $table->{$offset} or return;
289 my ($len, $class, $name, $title) = @{$def};
291 my $cols = $info->{cols};
292 my $colsize = $table->{colsize} || 1;
295 $name //= $len <= 2 ? 'res' : 'reserved';
297 if (my $part = ($offset/$colsize - $info->{startpoint}) % $cols) {
299 my $rest = $cols - $part; # remaining
300 $rest = $len if $len < $rest; #TODO: optimise
302 # continued on new row
303 my @next = ($len * $colsize, "$class joinu");
304 my $separate = $cols - $len > $rest; # columns not on next row
307 push @next, $name, $title;
309 $name = $separate && '…';
312 # minority on next row
313 push @next, $separate && '"', $title || $name;
315 $table->{$offset + $colsize*$rest} //= \@next;
320 elsif (my $rows = int($len / $cols)) {
322 my $rowsize = $colsize * $cols;
323 if ($len -= $rows * $cols) {
324 # partial row remains
325 $table->{$offset + $rowsize * $rows} //= [$len*$colsize, "$class joinu", '', $title];
329 unless ($info->{realsize}) {
330 # coalesce multiple rows
332 $info->{skip}->{$offset += $rowsize}++;
336 $info->{skip}->{$offset += $rowsize} = 0;
340 $attr .= sprintf ' rowspan=%d', $rows;
344 $attr .= sprintf ' colspan=%d', $len unless $len == 1;
345 $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
346 $attr .= sprintf ' class="%s"', $class if $class;
347 $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
348 return "<td$attr>$name\n";
351 for my $row (@request) {
352 my $cols = $row->{cols};
353 my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1;
354 my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header
355 my $rowdiv = 16 ** $coldigits; # row divide for column digits
356 $rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched
357 my $offset = $row->{startpoint} * $colsize || 0;
359 printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
360 my $title = $row->{set};
361 $title .= " <aside>(over $_)</aside>"
362 for $row->{parent} || ();
363 printf '<caption>%s</caption>', $title;
364 print '<col>' x ($cols + 1);
365 for my $section (qw{thead}) {
366 print "<$section><tr><th>", $rowdiv == 1 ? '+' : '↱';
367 printf '<th>%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
372 while ($offset <= $row->{endpoint} * $colsize) {
373 if ($row->{skip}->{$offset}) {
374 $offset += $cols * $colsize;
379 if (defined $row->{skip}->{$offset}) {
383 if (my $rowmod = $offset % $rowdiv) {
384 # offset in column units
385 printf '<small>+%X</small>', $rowmod;
389 printf '%X', ($offset + $row->{offset}) / $rowdiv;
396 print range_cell($row, $offset);
400 my $cp = $offset + $row->{offset};
401 my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
402 substr $row->{table}, $offset, 1;
403 my ($cell, $name, $class) = $glyph eq $NOCHAR ? () :
404 $glyphs->glyph_html($glyph);
406 if (exists $get{compare}) {
409 $cp == ord $glyph ? 'l4' :
410 $row->{parent} && $glyph eq
411 Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
413 $visible->{$glyph} ? 'l2' :
416 $visible->{$glyph}++;
419 say sprintf $class ? '<td title="%s" class="X %s">%s' : '<td title="%s">',
420 $name, $class, $cell;
426 say '</table></div>';
433 <table class="glyphs"><tr><: if (exists $get{compare}) { :>
434 <td class="X l4">unicode
435 <td class="X l3">inherited
436 <td class="X l2">existing
437 <td class="X l1">original
438 <td class="">unassigned
440 <td class="X Cc">control
441 <td class="X Zs"><span>whitespace</span>
442 <td class="X Mn">diacritic<table class="glyphs"><tr>
443 <td class="X Sk">letter
445 <td class="X Po">punctuation<table class="glyphs"><tr>
446 <td class="X Pf">quote
448 <td class="X So">symbol<table class="glyphs"><tr>
449 <td class="X Sm">math
450 <td class="X Sc">currency
452 <td class="X No">numeric
453 <td class="X Greek">greek<table class="glyphs"><tr>
454 <td class="X Latin">latin
455 <td class="X Cyrillic">cyrillic
457 <td class="X Aramaic">aramaic<table class="glyphs"><tr>
458 <td class="X Brahmic">brahmic
459 <td class="X Arabic">arabic
461 <td class="X Syllabic">syllabic<table class="glyphs"><tr>
462 <td class="X African">african
463 <td class="X Hiragana">japanese
464 <td class="X Han">cjk
465 <td class="X Bopomofo">chinese
467 <td class="X Alpha">alphabetic
470 <table class="glyphs"><tr>
471 <td class="X">unicode 7.0
472 <td class="X Xr">proposed
473 <td class="X Xd">deprecated
474 <td class="">unassigned
475 <td class="X Xi">invalid