3 my $mode = exists $get{compare};
4 my @tablist = split m{/+}, $Request || 'default';
7 title => 'charset cheat sheet',
10 "Reference sheet with all glyphs in common character encoding tables,",
11 "and an overview of Unicode ranges and UTF-8 bytes.",
14 charset codepage unicode ascii utf8 latin glyph character encoding
15 reference common overview table
17 stylesheet => [qw'light'],
19 charset-encoding.inc.pl
20 charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl
24 use List::Util qw( first pairmap pairfirst pairs );
27 <h1>Character encodings</h1>
31 if ($tablist[0] eq 'default') {
32 say "Overview of Unicode allocation and common latin code pages.";
33 say "Compare alternate charsets:";
36 say "Charset comparison:";
40 my ($title, $href, $selected) = @_;
42 $selected ? '<strong>%s</strong>' : '<a href="%2$s">%s</a>',
43 EscapeHTML($title), $href
49 join " ·\n", pairmap {
50 optionlink($b || ucfirst $a, '/charset'.($a && "/$a?compare"), $a eq $Request);
59 $tablist[0] eq 'default' ? () : ('' => 'common'),
64 norteur => 'North European',
76 use Shiar_Sheet::FormatChar;
77 my $glyphs = Shiar_Sheet::FormatChar->new;
80 my $charsets = do 'charset-encoding.inc.pl'
81 or Alert('Encoding metadata could not be read', $@ || $!);
84 # generate character table(s)
85 my $input = shift or return;
88 default => [qw( u+0-27F utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
89 unicode => [qw( uu+cols=32+realsize u+0-FFF u+1000-18AF u+18B0-1FFF )],
90 us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
91 ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
92 iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
93 dos => [qw( cp437 cp865 cp861 cp860 cp863 cp850 cp857 cp852 cp775
94 cp737 cp869 cp866 cp855 cp862 cp864 )],
95 aix => [qw( cp1006 )],
96 win => [qw( cp1252 cp1250 cp1254 cp1257 cp1258 cp1253 cp1251 cp1255 cp1256 cp874 )],
97 mac => [qw( MacRoman MacRomanian MacRumanian MacCroatian MacCentralEurRoman MacTurkish MacIcelandic MacSami
98 MacGreek MacCyrillic MacHebrew MacArabic MacFarsi MacThai )],
101 westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )],
102 centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB
103 turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )],
104 norteur => [qw( baltic nordic )],
105 baltic => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
106 nordic => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
107 cyrillic => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855
108 +400 +2DE0 +A640-A69F +500-52F )], # MacUkrainian is broken
109 arabic => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi
110 +600 +8A0-8BF+8E0 +750-77F )],
111 greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 +370 +1F00 )],
112 hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 +590 )],
113 thai => [qw( iso-8859-11 cp874 MacThai )],
114 vietnamese => [qw( viscii cp1258 MacVietnamese )],
115 symbols => [qw( symbol dingbats MacDingbats wingdings wingdings2 wingdings3 webdings )],
130 if (my $follow = $ALIAS->{$input}) {
131 tabinput($_) for ref $follow ? @{$follow} : $follow;
135 state $visible = {'' => 1}; # all present tables
136 my %row = (offset => 0, cols => 16);
137 my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
139 if (not defined $params) {
140 my $charset = $charsets->{$input} || {};
141 my @parents = @{ $charset->{inherit} || [] };
143 if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
144 $row{parent} = $parent;
146 $params = 80 unless $visible->{$parent}
147 or ($input eq 'MacCroatian' and defined $visible->{MacRomanian});
149 elsif (defined $visible->{ascii}) {
150 $row{parent} = $parents[0];
151 $params = $parents[1] // 80;
152 $params = 80 if hex $params >= 0x80; # ascii offset at most
155 $row{parent} = $parents[0];
156 $params = $parents[1] if hex $parents[1] == 0; # apply ascii end
158 $visible->{$_} //= 0 for $row{parent} || ();
161 for my $param (split /[+]+/, $params // '') {
162 if ($param eq 'realsize') {
165 elsif ($param =~ m{ \A cols = (\d+) \z }x) {
168 elsif ($param =~ m{ \A (?<start> \p{AHex}+) (?: [-] (?<end> \p{AHex}+) )? \z }x) {
169 if (defined $row{endpoint}) {
170 # extend earlier range
171 my $skip = int(($row{endpoint} || $row{startpoint}) / $row{cols});
172 for ($skip + 1 .. (hex($+{start}) / $row{cols}) - 1) {
173 $row{skip}->{ $_ * $row{cols} - $row{startpoint} }++;
177 $row{startpoint} = hex $+{start};
179 $row{endpoint} = hex($+{end} || 0);
182 Alert("Unknown option <q>$param</q> for charset $input");
186 if ($input =~ m{ \A (?:wing|web)dings \d* \z }ix) {
187 eval "require Encode::\u$input";
191 $row{offset} = delete $row{startpoint};
192 $row{set} = 'Unicode characters';
193 my $block = $row{offset} >> 8;
194 $row{endpoint} ||= ($block + 1 << 8) - 1;
195 $block == ($row{endpoint} >> 8) or undef $block;
197 $row{table} = join '', map { chr } $row{offset} .. $row{endpoint};
198 utf8::upgrade($row{table}); # prevent latin1 output
200 $row{endpoint} -= $row{offset};
202 if (defined $block) {
203 $row{set} = sprintf 'Unicode block U+%02Xxx', $block;
204 $row{offset} %= 0x100;
207 elsif (lc $input eq 'uu') {
208 $row{cell} = do 'charset-ucplanes.inc.pl'
209 or Alert('Table data could not be read', $@ || $!);
210 $row{endpoint} ||= 0x3FF;
211 $row{set} = 'Unicode planes';
213 elsif (lc $input eq 'u') {
214 $row{cell} = do 'charset-unicode.inc.pl'
215 or Alert('Table data could not be read', $@ || $!);
217 $row{endpoint} ||= 0x1FFF;
218 $row{set} = 'Unicode ' . (
219 $row{startpoint} < 0x1000 && $row{endpoint} < 0x1000 ? 'BMP' :
220 $row{startpoint} >= 0x1000 && $row{endpoint} < 0x2000 ? 'SMP' :
224 elsif ($input =~ m/^utf-*8$/i) {
226 $row{cell} = do 'charset-utf8.inc.pl'
227 or Alert('Table data could not be read', $@ || $!);
228 $row{endpoint} = 0xFF;
230 elsif ($row{set} = Encode::resolve_alias($input)) {
231 $row{offset} = delete $row{startpoint};
232 $row{endpoint} ||= 0xFF;
233 if ($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} .. $row{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} .. $row{endpoint});
245 if ($row{set} eq 'cp437') {
246 if ($row{offset} <= 0xED and $row{endpoint} >= 0xED) {
248 substr($row{table}, 0xED - $row{offset}, 1) = 'ϕ';
250 if ($row{offset} < 0x20) {
251 # replace control characters by visible variants
252 my $sub = substr ' ☺☻♥♦♣♠•◘○◙♂♀♪♫☼►◄↕‼¶§▬↨↑↓→←∟↔▲▼', $row{offset};
253 substr($row{table}, 0, length $sub) = $sub;
256 elsif ($row{set} eq 'symbol') {
257 if ($row{offset} <= 0x60 and $row{endpoint} >= 0x60) {
258 # replace radical extender by closest unicode equivalent
259 substr($row{table}, 0x60 - $row{offset}, 1) = '│';
261 if ($row{offset} <= 0xBD and $row{endpoint} >= 0xFF) {
262 substr($row{table}, 0xBD - $row{offset}, 2) = '⏐⎯'; # arrow extenders
263 substr($row{table}, 0xD2 - $row{offset}, 3) = '®©™'; # serif variants
264 substr($row{table}, 0xE0 - $row{offset}, 1) = '◊'; # replace lookalike, should match AdobeSymbol
265 substr($row{table}, 0xE2 - $row{offset}, 3) = '®©™'; # sans-serif variants
266 substr($row{table}, 0xE6 - $row{offset}, 10) = '⎛⎜⎝⎡⎢⎣⎧⎨⎩⎪';
267 substr($row{table}, 0xF0 - $row{offset}, 1) = '€';
268 substr($row{table}, 0xF4 - $row{offset}, 11) = '⎮⌡⎞⎟⎠⎤⎥⎦⎫⎬⎭';
272 $row{endpoint} -= $row{offset};
274 $visible->{ascii} = # assume common base
275 $visible->{ $row{set} } = 1;
278 Alert("Encoding <q>$input</q> unknown");
281 push @request, \%row;
283 tabinput($_) for @tablist;
285 my $NOCHAR = chr 0xFFFD;
288 my ($info, $offset) = @_;
289 my $table = $info->{cell} or return;
290 my $def = $table->{$offset} or return;
291 my ($len, $class, $name, $title) = @{$def};
293 my $cols = $info->{cols};
294 my $colsize = $table->{colsize} || 1;
297 $name //= $len <= 2 ? 'res' : 'reserved';
299 if (my $part = ($offset/$colsize - $info->{startpoint}) % $cols) {
301 my $rest = $cols - $part; # remaining
302 $rest = $len if $len < $rest; #TODO: optimise
304 # continued on new row
305 my @next = ($len * $colsize, "$class joinu");
306 my $separate = $cols - $len > $rest; # columns not on next row
309 push @next, $name, $title;
311 $name = $separate && '…';
314 # minority on next row
315 push @next, $separate && '"', $title || $name;
317 $table->{$offset + $colsize*$rest} //= \@next;
322 elsif (my $rows = int($len / $cols)) {
324 my $rowsize = $colsize * $cols;
325 if ($len -= $rows * $cols) {
326 # partial row remains
327 $table->{$offset + $rowsize * $rows} //= [$len*$colsize, "$class joinu", '', $title];
331 unless ($info->{realsize}) {
332 # coalesce multiple rows
334 $info->{skip}->{$offset += $rowsize}++;
338 $info->{skip}->{$offset += $rowsize} = 0;
342 $attr .= sprintf ' rowspan=%d', $rows;
346 $attr .= sprintf ' colspan=%d', $len unless $len == 1;
347 $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
348 $attr .= sprintf ' class="%s"', $class if $class;
349 $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
350 return "<td$attr>$name\n";
353 for my $row (@request) {
354 my $cols = $row->{cols};
355 my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1;
356 my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header
357 my $rowdiv = 16 ** $coldigits; # row divide for column digits
358 $rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched
359 my $offset = $row->{startpoint} * $colsize || 0;
361 printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
362 my $title = $row->{set};
363 $title .= " <aside>(over $_)</aside>"
364 for $row->{parent} || ();
365 printf '<caption>%s</caption>', $title;
366 print '<col>' x ($cols + 1);
367 for my $section (qw{thead}) {
368 print "<$section><tr><th>", $rowdiv == 1 ? '+' : '↱';
369 printf '<th>%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
374 while ($offset <= $row->{endpoint} * $colsize) {
375 if ($row->{skip}->{$offset}) {
376 $offset += $cols * $colsize;
381 if (defined $row->{skip}->{$offset}) {
385 if (my $rowmod = $offset % $rowdiv) {
386 # offset in column units
387 printf '<small>+%X</small>', $rowmod;
391 printf '%X', ($offset + $row->{offset}) / $rowdiv;
398 print range_cell($row, $offset);
402 my $cp = $offset + $row->{offset};
403 my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
404 substr $row->{table}, $offset, 1;
405 my ($cell, $name, $class) = $glyph eq $NOCHAR ? () :
406 $glyphs->glyph_html($glyph);
411 $cp == ord $glyph ? 'l4' :
412 $row->{parent} && $glyph eq
413 Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
415 $visible->{$glyph} ? 'l2' :
418 $visible->{$glyph}++;
421 say sprintf $class ? '<td title="%s" class="X %s">%s' : '<td title="%s">',
422 $name, $class, $cell;
428 say '</table></div>';
435 <table class="glyphs"><tr><: if ($mode) { :>
436 <td class="X l4">unicode
437 <td class="X l3">inherited
438 <td class="X l2">existing
439 <td class="X l1">original
440 <td class="">unassigned
442 <td class="X Cc">control
443 <td class="X Zs"><span>whitespace</span>
444 <td class="X Mn">diacritic<table class="glyphs"><tr>
445 <td class="X Sk">letter
447 <td class="X Po">punctuation<table class="glyphs"><tr>
448 <td class="X Pf">quote
450 <td class="X So">symbol<table class="glyphs"><tr>
451 <td class="X Sm">math
452 <td class="X Sc">currency
454 <td class="X No">numeric
455 <td class="X Greek">greek<table class="glyphs"><tr>
456 <td class="X Latin">latin
457 <td class="X Cyrillic">cyrillic
459 <td class="X Aramaic">aramaic<table class="glyphs"><tr>
460 <td class="X Brahmic">brahmic
461 <td class="X Arabic">arabic
463 <td class="X Syllabic">syllabic<table class="glyphs"><tr>
464 <td class="X African">african
465 <td class="X Hiragana">japanese
466 <td class="X Han">cjk
467 <td class="X Bopomofo">chinese
469 <td class="X Alpha">alphabetic
472 <table class="glyphs"><tr>
473 <td class="X">unicode 7.0
474 <td class="X Xr">proposed
475 <td class="X Xd">deprecated
476 <td class="">unassigned
477 <td class="X Xi">invalid