#!/usr/bin/env perl
-use 5.010;
-use strict;
+use 5.014;
use warnings;
use utf8;
no if $] >= 5.018, warnings => 'experimental::smartmatch';
use open OUT => ':utf8', ':std';
use Data::Dump 'pp';
-our $VERSION = '1.00';
+our $VERSION = '1.02';
my %info = (
# prepare presentational string for some control(lish) entries
"\xAD" => {string => '-'},
"\x{200E}" => {string => '→'},
"\x{200F}" => {string => '←'},
- "\x{200B}" => {string => '␣'},
+ "\x{200B}" => {string => '␣'}, # nbsp: ~ in TeX
"\x{200C}" => {string => '|'}, # ISO-9995-7-081 lookalike (alt: ∣ ⊺ ⟙)
"\x{200D}" => {string => '⁀'}, # join (alt: ∤ |ͯ ⨝)
);
$info{chr $_} //= {} for 32 .. 126;
eval {
- my $tables = do 'unicode-table.inc.pl' or die $@ || $!;
+ my $tables = do './unicode-table.inc.pl' or die $@ || $!;
for (values %$tables) {
for (values %$_) {
for (@$_) {
} or warn "Failed importing html entities: $@";
my %diinc = (
- 'data/digraphs-rfc.inc.pl' => 'u-di',
- 'data/digraphs-shiar.inc.pl' => 'u-prop Xz',
+ './data/digraphs-rfc.inc.pl' => 'u-di',
+ './data/digraphs-shiar.inc.pl' => 'u-prop',
+ './data/digraphs-vim.inc.pl' => 'u-vim',
);
-for (keys %diinc) {
+for (sort keys %diinc) {
-e $_ or next;
my $di = do $_ or die "Error reading digraphs file $_: ", $@ || $!;
- while (my ($mnem, $cp) = each %$di) {
+ for my $mnem (sort keys %{$di}) {
+ my $cp = $di->{$mnem};
length $mnem == 2 or next; # limit to digraphs
my $class = $diinc{$_};
- $info{chr $cp}->{di} //= $mnem;
- $info{chr $cp}->{class}->{$class}++;
+ $info{$cp}->{di} //= $mnem;
+ $info{$cp}->{class}->{$class}++;
}
}
eval {
# read introducing unicode versions for known characters
- my $agemap = do 'unicode-age.inc.pl' or die $@ || $!;
+ my $agemap = do './data/unicode-age.inc.pl' or die $@ || $!;
for my $chr (keys %info) {
my $version = $agemap->{ord $chr} or next;
$info{$chr}->{class}->{'u-v'.$version}++
}
1;
-} or warn "Failed including unicode version data $@";
+} or warn "Failed including unicode version data: $@";
for my $chr (keys %info) {
my $cp = ord $chr;
$info->{$_} = $info{$chr}->{$_} for keys %{ $info{$chr} };
+ # ignore vim flag in addition to rfc support, replace otherwise
+ $info->{class}->{'u-di'} or $info->{class}->{'u-prop'}++
+ if delete $info->{class}->{'u-vim'};
+
# categorise by unicode types and writing script
$info->{class}->{$_}++ for $info->{category};
$info->{class}->{$_}++ for $info->{script} || ();