data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'],
});
-my @tablist = split /[^\w-]+/, $Request || 'default';
+my @tablist = split m{/+}, $Request || 'default';
:>
<h1>Character encoding</h1>
my $input = shift or return;
state $ALIAS = {
- default => [qw( unicode+0-192 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
- us => [qw( cp437 cp863 AdobeStandardEncoding gsm0338 )],
- ebcdic => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )],
+ default => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
+ us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
+ ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
iso => [map {"iso-8859-$_"} 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 )],
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 )],
+
# iso-code shorthand
1 => 'westeur',
2 => 'centeur',
use List::Util qw( first pairfirst pairs );
state $INHERIT = {
- 'cp437' => ['cp850' => 0], # ascii range overridden later
+ 'cp437' => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later
'gsm0338' => ['ascii' => '0-127'],
'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],
- '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'], #TODO: gaps at C/E
'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned
+ 'MacSami' => ['MacIcelandic' => '144', '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' => '16-95+160-191'], #TODO: gap at 3/4
+ 'cp1026' => ['cp37' => '64'],
+ 'cp875' => ['cp37' => '48'],
+
};
my @parents = @{ $INHERIT->{$input} || [] };
if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
$row{parent} = $parent;
- $params = $visible->{$parent} ? $part : 128;
+ $params = $part;
+ $params = 128 unless $visible->{$parent}
+ or ($input eq 'MacCroatian' and defined $visible->{MacRomanian});
}
elsif (defined $visible->{ascii}) {
- $row{parent} = $parents[0] || 'ascii';
- $params = $parents[1] || 128;
+ $row{parent} = $parents[0];
+ $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};
+ $visible->{$_} //= 0 for $row{parent} || ();
+ }
+ else {
+ # manual option to double table width
+ $row{cols} *= 2 if $params =~ s/[+]\z//;
}
- if (defined $params and $params =~ m/^ (\d+) (-\d+)? /x) {
- $row{offset} = $1;
- $endpoint = -$2 if $2;
+ 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} - $row{offset} }++;
+ }
+ }
}
- if ($input =~ /^U([0-9a-f]+)(?:-([0-9a-f]+))?/) {
+ if ($input =~ /^U([0-9a-fA-F]+)(?:-([0-9a-fA-F]+))?/) {
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} = $start % 256;
}
elsif ($input eq 'U') {
- $row{len} = 1024;
$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{len} = $endpoint < 255 ? 640 : 8192;
$row{set} = 'Unicode BMP';
$row{cell} = do 'charset-unicode.inc.pl'
or Alert('Table data could not be read', $@ || $!);
+ $row{endpoint} = ($endpoint || 8191) * $row{cell}->{colsize};
}
elsif ($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', $@ || $!);
+ $row{endpoint} = 255;
}
else {
- if ($row{set} eq 'MacHebrew') {
+ 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
# substr strings is twice as fast as splitting to an array
$row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
}
- $row{len} = $endpoint - $row{offset};
+ $row{endpoint} = $endpoint - $row{offset};
if ($row{set} eq 'cp437' and !$row{offset}) {
substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
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} * $colsize;
printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
my $title = $row->{set};
print "\n";
}
print '<tbody>';
- while ($offset < $endpoint - 1) {
+ while ($offset < $row->{endpoint}) {
+ if ($row->{skip}->{$offset}) {
+ $offset += $cols * $colsize;
+ next;
+ }
+
print '<tr><th>';
{
if (my $rowmod = $offset % $rowdiv) {
next;
}
+ if (exists $get{compare}) {
+ state $visible = {};
+ my $cp = $offset + $row->{offset};
+ printf '<td class="%s" title="%3$s">%2$s',
+ $cp == ord $glyph ? 'l4' :
+ $row->{parent} && $glyph eq
+ Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
+ $visible->{$glyph} ? 'l2' :
+ 'l1',
+ $glyphs->glyph_html($glyph);
+ $visible->{$glyph}++;
+ next;
+ }
+
print "\n".$glyphs->glyph_cell($glyph);
}
continue {
<hr>
<div class="legend">
- <table class="glyphs"><tr>
+ <table class="glyphs"><tr><: if (exists $get{compare}) { :>
+ <td class="X l4">unicode
+ <td class="X l3">inherited
+ <td class="X l2">existing
+ <td class="X l1">original
+ <td class="">unassigned
+<: } else { :>
<td class="X Cc">control
<td class="X Zs"><span>whitespace</span>
<td class="X Mn">diacritic<table class="glyphs"><tr>
<td class="X Xd">deprecated
<td class="">unassigned
<td class="X Xi">invalid
- </table>
+<: } :> </table>
</div>
git.shiar.nl / sheet.git / blobdiff