-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 );
+}