};
}
-my @chargroups = qw(
- N Z Math
- Assigned
- Latin Greek Cyrillic Georgian Arabic Thai Hangul Han
-);
+my %charlist;
+
+my $chartables = do 'unicode-table.inc.pl' or warn $@ || $!;
+if ($chartables) {
+ while (my ($tablegroup, $grouprow) = each %{$chartables}) {
+ while (my ($tablename, $chars) = each %{$grouprow}) {
+ next if $tablename =~ /^-/;
+ my $includerows; # ignore rows before body row
+ for (@{$chars}) {
+ $includerows ||= m/^[.]/ or next;
+ next if /^[.-]/;
+ next if $_ eq '>' or $_ eq '=';
+ s/^\\//; # escape
+ push @{ $charlist{table}->{"$tablegroup/$tablename"} }, $_;
+ push @{ $charlist{table}->{$tablegroup} }, $_;
+ }
+ }
+# if ($tablegroup eq 'ipa') {
+# @chars = grep { !m/[a-zA-Z]/ } @chars;
+# }
+ }
+}
-say 'use utf8;';
-say '+{';
-for my $name (@chargroups) {
- my $match = qr/\A\p{$name}\z/;
- my @chars = eval {
- grep { m/$match/ } map { chr $_ }
- 0..0xD7FF, 0xE000..0xFDCF, 0xFDF0..0xFFFD,
- } or do {
- warn $@;
- next;
- };
+eval {
+ require HTML::Entities;
+ our %char2entity;
+ HTML::Entities->import('%char2entity');
+ while (my ($char, $entity) = each %char2entity) {
+ $entity =~ /[a-zA-Z]/ or next; # only actual aliases
+ push @{ $charlist{table}->{html} }, $char;
+ }
+ 1;
+} or warn "Could not include count for html entities: $@";
+
+use Unicode::UCD 'charinfo';
+for my $code (0 .. 256**2) {
+ my $charinfo = charinfo($code) or next;
+ next if $charinfo->{category} =~ /^[MC]/; # ignore Marks and "other" Control chars
+ push @{ $charlist{$_}->{ $charinfo->{$_} } }, chr $code
+ for qw( script category block );
+}
- my %cover = map {
+for (values %charlist) {
+for my $chars (values %{$_}) {
+ my %row = map {
my $fontcover = $font{$_};
- ($_ => scalar grep { $fontcover->{$_} } @chars);
+ ($_ => scalar grep { $fontcover->{$_} } @{$chars});
} keys %font;
- $cover{-count} = scalar @chars;
- $cover{-chars} = [ map { ord } sort @chars ];
+ $row{-count} = scalar @{$chars};
+
+ $row{-query} = eval {
+ my @query = map { ord } sort @{$chars};
+ my $i = 0;
+ while ($i < @query) {
+ my $j = $i + 1;
+ my $v = $query[$i];
+ while ($j < @query) {
+ $v++;
+ last if $query[$j] != $v;
+ $j++;
+ }
+ if ($j - $i > 2) {
+ splice(@query, $i, $j - $i, "$query[$i]-$query[$j-1]");
+ }
+ $i++;
+ }
+ return join '+', @query;
+ };
- say $name.' => '.pp(\%cover).',';
+ $chars = \%row;
+}
}
-say '}';
+
+say "# automatically generated by $0";
+say 'use utf8;';
+say '+'.pp(\%charlist);
__END__