6 no if $] >= 5.018, warnings => 'experimental::smartmatch';
8 use open OUT => ':utf8', ':std';
11 our $VERSION = '1.01';
14 # prepare presentational string for some control(lish) entries
15 "\xAD" => {string => '-'},
16 "\x{200E}" => {string => '→'},
17 "\x{200F}" => {string => '←'},
18 "\x{200B}" => {string => '␣'}, # nbsp: ~ in TeX
19 "\x{200C}" => {string => '|'}, # ISO-9995-7-081 lookalike (alt: ∣ ⊺ ⟙)
20 "\x{200D}" => {string => '⁀'}, # join (alt: ∤ |ͯ ⨝)
22 $info{chr $_} //= {} for 32 .. 126;
25 my $tables = do 'unicode-table.inc.pl' or die $@ || $!;
26 for (values %$tables) {
29 length $_ == 1 or next; # ignore meta values
36 } or warn "Failed reading unicode tables: $@";
39 require HTML::Entities;
41 HTML::Entities->import('%char2entity');
42 while (my ($char, $entity) = each %char2entity) {
43 $entity =~ /[a-zA-Z]/ or next; # only actual aliases
44 $info{$char}->{html} = substr($entity, 1, -1);
47 } or warn "Failed importing html entities: $@";
50 'data/digraphs-rfc.inc.pl' => 'u-di',
51 'data/digraphs-shiar.inc.pl' => 'u-prop ex',
52 'data/digraphs-vim.inc.pl' => 'u-prop',
54 for (sort keys %diinc) {
56 my $di = do $_ or die "Error reading digraphs file $_: ", $@ || $!;
57 for my $mnem (sort keys %{$di}) {
58 my $cp = $di->{$mnem};
59 length $mnem == 2 or next; # limit to digraphs
60 my $class = $diinc{$_};
61 $info{$cp}->{di} //= $mnem;
62 $info{$cp}->{class}->{$class}++;
67 # read introducing unicode versions for known characters
68 my $agemap = do 'unicode-age.inc.pl' or die $@ || $!;
69 for my $chr (keys %info) {
70 my $version = $agemap->{ord $chr} or next;
71 $info{$chr}->{class}->{'u-v'.$version}++
74 } or warn "Failed including unicode version data $@";
76 for my $chr (keys %info) {
78 # attempt to get unicode character information
81 Unicode::UCD::charinfo($cp)
82 || { block => '?', category => 'Xn', name => '', script => '' }
85 $info->{$_} = $info{$chr}->{$_} for keys %{ $info{$chr} };
87 # official digraphs either lose vim flag or gain experimental
88 delete $info->{class}->{'u-prop'} or $info->{class}->{ex}++
89 if $info->{class}->{'u-di'};
91 # categorise by unicode types and writing script
92 $info->{class}->{$_}++ for $info->{category};
93 $info->{class}->{$_}++ for $info->{script} || ();
95 # add custom categories for certain blocks
96 $info->{class}->{Xa}++ if $info->{block} eq 'Basic Latin';
97 $info->{class}->{Xl}++ if $info->{block} eq 'Latin-1 Supplement';
100 if ($info->{string}) {
101 # keep predefined presentational string
103 elsif ($info->{combining}) {
104 # overlay combining accents
105 $info->{string} = chr(9676) . $chr;
107 elsif (($cp & ~0b1001_1111) == 0 or $cp == 127) {
108 # control characters (first 32 chars from 0 and 128)
109 # rename to something more descriptive
110 $info->{name} = $info->{unicode10}
111 ? '<'.$info->{unicode10}.'>' # the old name was much more useful
112 : sprintf('<control U+%04X>', $cp); # at least identify by value
113 # show descriptive symbols instead of control chars themselves
114 $info->{string} = $cp < 32 ? chr($cp + 0x2400) :
115 $cp == 127 ? chr(0x2421) :
123 # output perl code of hash
124 say "# automatically generated by $0";
127 for my $cp (sort keys %info) {
128 $info{$cp}->{classstr} = join(' ', sort keys %{ $info{$cp}->{class} });
129 # convert info hashes into arrays of strings to output in display order
130 my $row = [ map { $info{$cp}->{$_} } qw/classstr name di html string/ ];
131 # strip off trailing missing values (especially string may be unknown)
132 defined $row->[-1] ? last : pop @$row for 1 .. @$row;
133 # final line (assume safe within single quotes)
134 say sprintf '"\x{%X}" => [%s],',
135 ord $cp, join(',', map { escapeq($_) } @$row);
141 return 'undef' if not defined;
150 mkcharinfo - Gather Unicode character details in Perl array
154 mkcharinfo > unicode-char.inc.pl
156 Test by printing the description of U+0041 (latin A):
158 perl -e'$u = do "unicode-char.inc.pl"; print $u->{A}->[1]'
162 Mischa POSLAWSKY <perl@shiar.org>
166 Licensed under the GNU Affero General Public License version 3.