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?compare", $a eq $Request);
+ } @{$_}
+ }
+ [
+ iso => 'ISO',
+ dos => 'DOS',
+ mac => 'Apple',
+ ebcdic => 'EBCDIC',
+ ],
+ [
+ westeur => 'West',
+ centeur => 'Central',
+ norteur => 'North European',
+ turkish => 0,
+ greek => 0,
+ cyrillic => 0,
+ hebrew => 0,
+ ],
+ [ uc => 'Unicode' ],
+);
+:>.
+</p>
<:
use POSIX qw( ceil );
my $input = shift or return;
state $ALIAS = {
- default => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
+ default => [qw( unicode+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
+ uc => [qw( U+cols=32 unicode+0-4095 unicode+4096-6319 unicode+6320-8191 )],
us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
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
state $visible = {}; # 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' => '0-31+128'], # ascii range overridden later
'gsm0338' => ['ascii' => '0-127'],
'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' => '160-191+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'], #TODO: gaps at C/E
+ 'MacIcelandic'=> ['MacRoman' => '160-175+208-239'],
'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned
- 'MacSami' => ['MacIcelandic' => '144', 'MacRoman' => '144'],
+ 'MacSami' => ['MacIcelandic' => '144-191+208-223+240', 'MacRoman' => '144'],
'MacGreek' => ['MacRoman' => '128'],
'MacCyrillic' => ['MacRoman' => '128'],
'MacHebrew' => ['iso-8859-8' => '128', 'MacRoman' => '128-143+160'], # partial ascii
'cp37' => ['posix-bc' => '0'],
'posix-bc' => ['cp1047' => '64'],
'cp500' => ['cp37' => '64-95+176-191'],
- 'cp1047' => ['cp37' => '16-95+160-191'], #TODO: gap at 3/4
+ 'cp1047' => ['cp37' => '16-47+80-95+160-191'],
'cp1026' => ['cp37' => '64'],
'cp875' => ['cp37' => '48'],
}
$visible->{$_} //= 0 for $row{parent} || ();
}
- else {
- # manual option to double table width
- $row{cols} *= 2 if $params =~ s/[+]\z//;
- }
- if (length $params) {
- $params =~ m{
- \A (?<offset> \d+)
- (?: (?: [-] (?<stop> \d+) )? (?: [+] (?<restart> \d+) ) )?
- (?: [-] (?<endpoint> \d+) )? \z
- }x or Alert([
- "Unknown range parameters for $input",
- "<q>$params</q> is not in format start(-stop)(+restart(-end))",
- ]);
-
- $row{offset} = $+{offset};
- $endpoint = $+{endpoint} if $+{endpoint};
- if (my $restart = $+{restart}) {
- my $skip = int(($+{stop} || $row{offset}) / $row{cols});
- for ($skip + 1 .. ($restart / $row{cols}) - 1) {
- $row{skip}->{ $_ * $row{cols} - $row{offset} }++;
+ 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 (?<offset> \d+) (?: [-] (?<endpoint> \d+) )? \z }x) {
+ if (defined $row{endpoint}) {
+ # extend earlier range
+ my $skip = int(($row{endpoint} || $row{offset}) / $row{cols});
+ for ($skip + 1 .. ($+{offset} / $row{cols}) - 1) {
+ $row{skip}->{ $_ * $row{cols} - $row{offset} }++;
+ }
+ }
+ else {
+ $row{offset} = $+{offset};
}
+ $row{endpoint} = $+{endpoint} || 0;
+ }
+ else {
+ Alert("Unknown option <q>$param</q> for charset $input");
}
}
$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 * $row{cell}->{colsize};
}
elsif ($row{set} = Encode::resolve_alias($input)) {
if ($row{set} eq 'Internal') {
- $row{set} = 'Unicode BMP';
$row{cell} = do 'charset-unicode.inc.pl'
or Alert('Table data could not be read', $@ || $!);
- $row{endpoint} = ($endpoint || 8191) * $row{cell}->{colsize};
+
+ $row{endpoint} ||= 8191;
+ $row{endpoint} *= $row{cell}->{colsize};
+ $row{startpoint} = $row{cell}->{colsize} * $row{offset};
+ $row{offset} = 0;
+ $row{set} = 'Unicode ' . (
+ $row{startpoint} < 0x10000 && $row{endpoint} < 0x10000 ? 'BMP' :
+ $row{startpoint} >= 0x10000 && $row{endpoint} < 0x20000 ? 'SMP' :
+ 'allocations'
+ );
}
elsif ($row{set} eq 'utf-8-strict') {
$row{set} = 'UTF-8';
$row{endpoint} = 255;
}
else {
+ $row{endpoint} ||= 255;
if ($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} .. $endpoint
+ map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $row{endpoint}
];
}
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);
+ $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint});
}
- $row{endpoint} = $endpoint - $row{offset};
+ $row{endpoint} -= $row{offset};
if ($row{set} eq 'cp437' and !$row{offset}) {
substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
$class .= ' joind';
}
- # coalesce multiple rows
- while ($rows > 3) {
- $info->{skip}->{$offset += $rowsize}++;
- $rows--;
- }
- if ($rows > 2) {
- $info->{skip}->{$offset += $rowsize} = 0;
+ unless ($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;
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 $offset = $row->{startpoint} || 0;
printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
my $title = $row->{set};