reference common overview table
'],
stylesheet => [qw'light'],
- data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'],
+ data => [qw'charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl'],
});
-my @tablist = split /[^\w-]+/, $Request || 'default';
+my @tablist = split m{/+}, $Request || 'default';
+
+use List::Util qw( first pairmap pairfirst pairs );
:>
-<h1>Character encoding</h1>
+<h1>Character encodings</h1>
+
+<p>
+<:
+if ($tablist[0] eq 'default') {
+ say "Overview of Unicode allocation and common latin code pages.";
+ say "Compare alternate charsets:";
+}
+else {
+ say "Charset comparison:";
+}
+
+sub optionlink {
+ my ($title, $href, $selected) = @_;
+ return sprintf(
+ $selected ? '<strong>%s</strong>' : '<a href="%2$s">%s</a>',
+ EscapeHTML($title), $href
+ );
+}
+
+print join " •\n", (
+ map {
+ join " ·\n", pairmap {
+ optionlink($b || ucfirst $a, '/charset'.($a && "/$a?compare"), $a eq $Request);
+ } @{$_}
+ }
+ [
+ iso => 'ISO',
+ win => 'Windows',
+ dos => 'DOS',
+ mac => 'Apple',
+ ebcdic => 'EBCDIC',
+ $tablist[0] eq 'default' ? () : ('' => 'common'),
+ ],
+ [
+ westeur => 'West',
+ centeur => 'Central',
+ norteur => 'North European',
+ turkish => 0,
+ greek => 0,
+ cyrillic => 0,
+ hebrew => 0,
+ ],
+);
+:>.
+</p>
<:
use POSIX qw( ceil );
my $input = shift or return;
state $ALIAS = {
- default => [qw( unicode+0-192 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
- us => [qw( cp437 cp863 AdobeStandardEncoding gsm0338 )],
- ebcdic => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )],
+ default => [qw( u+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
+ unicode => [qw( uu+cols=32+realsize u+0-4095 u+4096-6319 u+6320-8191 )],
+ us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
+ ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
dos => [qw( cp437 cp865 cp861 cp860 cp863 cp850 cp857 cp852 cp775
cp737 cp869 cp866 cp855 cp862 cp864 )],
aix => [qw( cp1006 )],
win => [qw( cp1252 cp1250 cp1254 cp1257 cp1258 cp1253 cp1251 cp1255 cp1256 cp874 )],
+ mac => [qw( MacRoman MacRomanian MacRumanian MacCroatian MacCentralEurRoman MacTurkish MacIcelandic MacSami
+ MacGreek MacCyrillic MacHebrew MacArabic MacFarsi MacThai )],
+ # languages
westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )],
centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB
turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )],
+ norteur => [qw( baltic nordic )],
baltic => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
nordic => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
- cyrillic => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855 )], # MacUkrainian is broken
- arabic => [qw( iso-8859-6 cp1006 cp864 cp1256 MacArabic )], # MacFarsi same as MacArabic?
- greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 )],
- hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 )],
+ cyrillic => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855
+ U4 U2DE-2DF UA64-A69 U50-52 )], # MacUkrainian is broken
+ arabic => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi
+ U6 U8A-8F+0-31+64 U75-77 )],
+ greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 U37-3F U1F )],
+ hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 U59-5F )],
thai => [qw( iso-8859-11 cp874 MacThai )],
vietnamese => [qw( viscii cp1258 MacVietnamese )],
- symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )],
+ symbols => [qw( symbol dingbats MacDingbats wingdings wingdings2 wingdings3 webdings )],
+
# iso-code shorthand
1 => 'westeur',
2 => 'centeur',
return;
}
- state $visible = {}; # all present tables
+ state $visible = {'' => 1}; # all present tables
my %row = (offset => 0, cols => 16);
- my $endpoint = 255;
my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
if (not defined $params) {
- use List::Util qw( first pairfirst pairs );
-
state $INHERIT = {
- 'cp437' => ['cp850' => 0], # ascii range overridden later
+ 'ascii' => ['' => '0-127'],
+ 'cp437' => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later
'gsm0338' => ['ascii' => '0-127'],
+ 'dingbats' => ['' => '32-127+160'],
+ 'MacDingbats' => ['dingbats' => '128-159'],
+ 'AdobeZdingbat'=> ['MacDingbats' => '128-159'], # should be identical but maps to private use
+ 'symbol' => ['' => '32-127+160'],
+ 'AdobeSymbol' => ['symbol' => '32-127+160', '' => '32-127+160'], # minor differences, irrelevant except for different '€'
+ 'wingdings' => ['' => '32'],
+ 'wingdings2' => ['' => '32'],
+ 'wingdings3' => ['' => '32'],
+ 'webdings' => ['' => '32'],
'iso-8859-2' => ['iso-8859-1' => '160'],
- 'iso-8859-3' => ['iso-8859-1' => '160'],
+ 'iso-8859-3' => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9
'iso-8859-4' => ['iso-8859-2' => '160'],
'iso-8859-5' => ['iso-8859-1' => '160'],
- 'iso-8859-6' => ['iso-8859-1' => '160'],
+ 'iso-8859-6' => ['cp1256' => '128', 'iso-8859-1' => '160'],
'iso-8859-7' => ['iso-8859-1' => '160'],
'iso-8859-8' => ['iso-8859-1' => '160'],
- 'iso-8859-9' => ['iso-8859-1' => '208'],
+ 'iso-8859-9' => ['iso-8859-1' => '208-223+240'],
'iso-8859-10' => ['iso-8859-4' => '160'],
'iso-8859-11' => ['iso-8859-1' => '160'],
'iso-8859-13' => ['iso-8859-4' => '160'],
'hp-roman8' => ['iso-8859-1' => '160'],
'cp1252' => ['iso-8859-1' => '128-159'],
- 'cp1250' => ['iso-8859-2' => '128-191'],
- 'cp1254' => ['iso-8859-9' => '128-159'],
- 'cp874' => ['iso-8859-11' => '128-159'], # windows-874 actually cp1162
- 'cp1257' => ['iso-8859-13' => '128'],
+ 'cp1250' => ['iso-8859-2' => '128-191', 'cp1252' => '128'],
+ 'cp1254' => ['iso-8859-9' => '128-159', 'cp1252' => '128-159+208'],
+ 'cp874' => ['iso-8859-11' => '128-159', 'cp1252' => '128'], # windows-874 actually cp1162
+ 'cp1257' => ['iso-8859-13' => '128-159+255', 'cp1252' => '128'],
'cp1251' => ['cp1252' => '128'],
'cp1253' => ['cp1252' => '128'],
'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'],
'cp1256' => ['cp1252' => '128'],
+ 'cp1258' => ['cp1252' => '128-159+192'],
'cp850' => ['cp437' => '144'],
'cp860' => ['cp437' => '128-175'],
'cp863' => ['cp437' => '128-175'],
'cp865' => ['cp437' => '144-175'],
'cp852' => ['cp850' => '128', 'cp437' => '128'],
- 'cp857' => ['cp850' => '128-239', 'cp437' => '128'],
+ 'cp857' => ['cp850' => '128-175+208-239', 'cp437' => '128'],
'cp775' => ['cp850' => '128'], # partial cp437
- 'cp866' => ['cp437' => '128'],
+ 'cp866' => ['cp437' => '128-175+224'],
'cp855' => ['cp437' => '128'],
- 'cp1006' => ['cp437' => '128'],
- 'cp737' => ['cp437' => '128'],
+ 'cp1006' => ['iso-8859-6' => '160', 'cp437' => '128'],
+ 'cp737' => ['cp437' => '128-175+224'],
'cp869' => ['cp437' => '128'],
- 'cp862' => ['cp437' => '128'],
- 'cp864' => ['cp437' => '128'],
+ 'cp862' => ['cp437' => '128-159'],
+ 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants
- 'koi8-u' => ['koi8-r' => 128],
- 'koi8-f' => ['koi8-u' => 128],
+ 'koi8-u' => ['koi8-r' => '144-191'],
+ 'koi8-f' => ['koi8-u' => '144-191'],
- 'MacRomanian' => ['MacRoman' => '160'],
+ 'MacRomanian' => ['MacRoman' => '160-191+208-223'],
+ 'MacRumanian' => ['MacRomanian' => '160-191+208-223', 'MacRoman' => '160-191+208-223'],
'MacCroatian' => ['MacRoman' => '160'],
'MacCentralEurRoman' => ['MacRoman' => '128'],
+ 'MacIcelandic'=> ['MacRoman' => '160-175+208-239'],
'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned
+ 'MacSami' => ['MacIcelandic' => '144-191+208-223+240', 'MacRoman' => '144'],
+ 'MacGreek' => ['MacRoman' => '128'],
'MacCyrillic' => ['MacRoman' => '128'],
- 'MacHebrew' => ['MacRoman' => '128'],
+ 'MacHebrew' => ['iso-8859-8' => '128', 'MacRoman' => '128-143+160'], # partial ascii
+ 'MacArabic' => ['iso-8859-6' => '128', 'cp864' => '128', 'MacRoman' => '128'], #TODO: multiple parents
+ 'MacFarsi' => ['MacArabic' => '176-191', 'MacRoman' => '128'],
+
+ 'cp37' => ['posix-bc' => '0'],
+ 'posix-bc' => ['cp1047' => '64'],
+ 'cp500' => ['cp37' => '64-95+176-191'],
+ 'cp1047' => ['cp37' => '16-47+80-95+160-191'],
+ 'cp1026' => ['cp37' => '64'],
+ 'cp875' => ['cp37' => '48'],
+
};
my @parents = @{ $INHERIT->{$input} || [] };
}
elsif (defined $visible->{ascii}) {
$row{parent} = $parents[0];
- $params = $parents[1] || 128;
+ $params = $parents[1] // 128;
$params = 128 if $params >= 128; # ascii offset at most
}
elsif (@parents) {
$row{parent} = $parents[0];
+ $params = $parents[1] if $parents[1] == 0; # apply ascii end
+ }
+ $visible->{$_} //= 0 for $row{parent} || ();
+ }
+
+ for my $param (split /[+]+/, $params // '') {
+ if ($param eq 'realsize') {
+ $row{realsize}++;
+ }
+ elsif ($param =~ m{ \A cols = (\d+) \z }x) {
+ $row{cols} = $1;
+ }
+ elsif ($param =~ m{ \A (?<start> \d+) (?: [-] (?<end> \d+) )? \z }x) {
+ if (defined $row{endpoint}) {
+ # extend earlier range
+ my $skip = int(($row{endpoint} || $row{startpoint}) / $row{cols});
+ for ($skip + 1 .. ($+{start} / $row{cols}) - 1) {
+ $row{skip}->{ $_ * $row{cols} - $row{startpoint} }++;
+ }
+ }
+ else {
+ $row{startpoint} = $+{start};
+ }
+ $row{endpoint} = $+{end} || 0;
+ }
+ else {
+ Alert("Unknown option <q>$param</q> for charset $input");
}
- $visible->{$_} //= 0 for $row{parent};
}
- if (defined $params and $params =~ m/^ (\d+) (-\d+)? /x) {
- $row{offset} = $1;
- $endpoint = -$2 if $2;
+ if ($input =~ m{ \A (?:wing|web)dings \d* \z }ix) {
+ eval "require Encode::\u$input";
}
- if ($input =~ /^U([0-9a-f]+)(?:-([0-9a-f]+))?/) {
+ if ($input =~ m{ \A u ([0-9a-f]+) (?:-([0-9a-f]+))? \z }ix) {
my $start = hex($1) << ($2 ? 4 : 8);
- my $end = $2 ? hex($2) << 4 : $start + 240;
- $row{table} = join '', map { chr } $start .. $end+15;
+ my $end = $2 ? (hex($2) << 4) + $row{cols} - 1 : $start + 255;
+ $row{table} = join '', map { chr } $start .. $end;
utf8::upgrade($row{table}); # prevent latin1 output
- $row{len} = length $row{table};
+ $row{endpoint} = $end - $start;
$row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
+ $row{offset} = $start % 256;
+ $row{startpoint} = 0;
}
- elsif ($input eq 'U') {
- $row{len} = 1024;
+ elsif (lc $input eq 'uu') {
$row{set} = 'Unicode planes';
$row{cell} = do 'charset-ucplanes.inc.pl'
or Alert('Table data could not be read', $@ || $!);
- $row{cols} *= 2;
+ $row{endpoint} ||= 1023;
+ }
+ elsif (lc $input eq 'u') {
+ $row{cell} = do 'charset-unicode.inc.pl'
+ or Alert('Table data could not be read', $@ || $!);
+
+ $row{endpoint} ||= 8191;
+ $row{set} = 'Unicode ' . (
+ $row{startpoint} < 0x1000 && $row{endpoint} < 0x1000 ? 'BMP' :
+ $row{startpoint} >= 0x1000 && $row{endpoint} < 0x2000 ? 'SMP' :
+ 'allocations'
+ );
+ }
+ elsif ($input =~ m/^utf-*8$/i) {
+ $row{set} = 'UTF-8';
+ $row{cell} = do 'charset-utf8.inc.pl'
+ or Alert('Table data could not be read', $@ || $!);
+ $row{endpoint} = 255;
}
elsif ($row{set} = Encode::resolve_alias($input)) {
- if ($row{set} eq 'Internal') {
- $row{len} = $endpoint < 255 ? 640 : 8192;
- $row{set} = 'Unicode BMP';
- $row{cell} = do 'charset-unicode.inc.pl'
- or Alert('Table data could not be read', $@ || $!);
- }
- elsif ($row{set} eq 'utf-8-strict') {
- $row{len} = 256;
- $row{set} = 'UTF-8';
- $row{cell} = do 'charset-utf8.inc.pl'
- or Alert('Table data could not be read', $@ || $!);
+ $row{offset} = delete $row{startpoint};
+ $row{endpoint} ||= 255;
+ if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') {
+ # array of possibly multiple characters per code point
+ $row{table} = [
+ map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $row{endpoint}
+ ];
}
else {
- if ($row{set} eq 'MacHebrew') {
- # array of possibly multiple characters per code point
- $row{table} = [
- map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $endpoint
- ];
+ # ~16x faster than decoding in loop;
+ # substr strings is twice as fast as splitting to an array
+ $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint});
+ }
+
+ if ($row{set} eq 'cp437') {
+ if ($row{offset} <= 0xED and $row{endpoint} >= 0xED) {
+ # replace phi glyph
+ substr($row{table}, 0xED - $row{offset}, 1) = 'ϕ';
}
- else {
- # ~16x faster than decoding in loop;
- # substr strings is twice as fast as splitting to an array
- $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
+ if ($row{offset} < 0x20) {
+ # replace control characters by visible variants
+ my $sub = substr ' ☺☻♥♦♣♠•◘○◙♂♀♪♫☼►◄↕‼¶§▬↨↑↓→←∟↔▲▼', $row{offset};
+ substr($row{table}, 0, length $sub) = $sub;
}
- $row{len} = $endpoint - $row{offset};
-
- if ($row{set} eq 'cp437' and !$row{offset}) {
- substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
- substr($row{table}, 0, 32) = pack 'U*', map {hex} qw(
- 2007 263A 263B 2665 2666 2663 2660 2022
- 25D8 25CB 25D9 2642 2640 266A 266B 263C
- 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8
- 2191 2193 2192 2190 221F 2194 25B2 25BC
- );
+ }
+ elsif ($row{set} eq 'symbol') {
+ if ($row{offset} <= 0x60 and $row{endpoint} >= 0x60) {
+ # replace radical extender by closest unicode equivalent
+ substr($row{table}, 0x60 - $row{offset}, 1) = '│';
+ }
+ if ($row{offset} <= 0xBD and $row{endpoint} >= 0xFF) {
+ substr($row{table}, 0xBD - $row{offset}, 2) = '⏐⎯'; # arrow extenders
+ substr($row{table}, 0xD2 - $row{offset}, 3) = '®©™'; # serif variants
+ substr($row{table}, 0xE0 - $row{offset}, 1) = '◊'; # replace lookalike, should match AdobeSymbol
+ substr($row{table}, 0xE2 - $row{offset}, 3) = '®©™'; # sans-serif variants
+ substr($row{table}, 0xE6 - $row{offset}, 10) = '⎛⎜⎝⎡⎢⎣⎧⎨⎩⎪';
+ substr($row{table}, 0xF0 - $row{offset}, 1) = '€';
+ substr($row{table}, 0xF4 - $row{offset}, 11) = '⎮⌡⎞⎟⎠⎤⎥⎦⎫⎬⎭';
}
-
- $visible->{ascii} = # assume common base
- $visible->{ $row{set} } = 1;
}
+
+ $row{endpoint} -= $row{offset};
+
+ $visible->{ascii} = # assume common base
+ $visible->{ $row{set} } = 1;
}
else {
Alert("Encoding <q>$input</q> unknown");
$len /= $colsize;
$name //= $len <= 2 ? 'res' : 'reserved';
- if (my $part = $offset/$colsize % $cols) {
+ if (my $part = ($offset/$colsize - $info->{startpoint}) % $cols) {
# continued row
my $rest = $cols - $part; # remaining
$rest = $len if $len < $rest; #TODO: optimise
if ($len -= $rest) {
# continued on new row
my @next = ($len * $colsize, "$class joinu");
+ my $separate = $cols - $len > $rest; # columns not on next row
if ($len > $rest) {
# minority remains
push @next, $name, $title;
$title ||= $name;
- $name = '';
+ $name = $separate && '…';
}
else {
# minority on next row
- push @next, '"', $title || $name;
+ push @next, $separate && '"', $title || $name;
}
$table->{$offset + $colsize*$rest} //= \@next;
$class .= ' joind';
}
elsif (my $rows = int($len / $cols)) {
# multiple full rows
+ my $rowsize = $colsize * $cols;
if ($len -= $rows * $cols) {
# partial row remains
- $table->{$offset + $colsize*$rows * $cols} //= [$len*$colsize, "$class joinu", '', $title];
+ $table->{$offset + $rowsize * $rows} //= [$len*$colsize, "$class joinu", '', $title];
$class .= ' joind';
}
+
+ unless ($info->{realsize}) {
+ # coalesce multiple rows
+ while ($rows > 3) {
+ $info->{skip}->{$offset += $rowsize}++;
+ $rows--;
+ }
+ if ($rows > 2) {
+ $info->{skip}->{$offset += $rowsize} = 0;
+ }
+ }
+
$attr .= sprintf ' rowspan=%d', $rows;
$len = $cols;
}
$attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
$attr .= sprintf ' class="%s"', $class if $class;
$attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
- return "<td$attr>$name";
+ return "<td$attr>$name\n";
}
for my $row (@request) {
my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header
my $rowdiv = 16 ** $coldigits; # row divide for column digits
$rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched
- my $offset = 0;
- my $endpoint = $offset + $row->{len} * $colsize;
+ my $offset = $row->{startpoint} * $colsize || 0;
printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
my $title = $row->{set};
$title .= " <aside>(over $_)</aside>"
- for grep { $_ ne 'iso-8859-1' } $row->{parent} // ();
+ for $row->{parent} || ();
printf '<caption>%s</caption>', $title;
print '<col>' x ($cols + 1);
for my $section (qw{thead}) {
print "\n";
}
print '<tbody>';
- while ($offset < $endpoint - 1) {
+ while ($offset < $row->{endpoint} * $colsize) {
+ if ($row->{skip}->{$offset}) {
+ $offset += $cols * $colsize;
+ next;
+ }
+
print '<tr><th>';
- {
+ if (defined $row->{skip}->{$offset}) {
+ print '⋮';
+ }
+ else {
if (my $rowmod = $offset % $rowdiv) {
# offset in column units
printf '<small>+%X</small>', $rowmod;
printf '%X', ($offset + $row->{offset}) / $rowdiv;
}
}
+ say '';
+
for (1 .. $cols) {
if ($row->{cell}) {
print range_cell($row, $offset);
next;
}
+ my $cp = $offset + $row->{offset};
my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
substr $row->{table}, $offset, 1;
- if ($glyph eq $NOCHAR) {
- print '<td>';
- next;
- }
+ my ($cell, $name, $class) = $glyph eq $NOCHAR ? () :
+ $glyphs->glyph_html($glyph);
if (exists $get{compare}) {
state $visible = {};
- my $cp = $offset + $row->{offset};
- printf '<td class="%s" title="%3$s">%2$s',
+ $class = (
$cp == ord $glyph ? 'l4' :
$row->{parent} && $glyph eq
Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
+ !$class ? undef :
$visible->{$glyph} ? 'l2' :
- 'l1',
- $glyphs->glyph_html($glyph);
+ 'l1'
+ );
$visible->{$glyph}++;
- next;
}
- print "\n".$glyphs->glyph_cell($glyph);
+ say sprintf $class ? '<td title="%s" class="X %s">%s' : '<td title="%s">',
+ $name, $class, $cell;
}
continue {
$offset += $colsize;
}
- print "\n";
}
say '</table></div>';
}
git.shiar.nl / sheet.git / blobdiff