tools/mkcharinfo: comment alternative nbsp stringification

[sheet.git] / tools / mkcharinfo

diff --git a/tools/mkcharinfo b/tools/mkcharinfo
index b58f0c373f6215e828e4ad787683d1cfc23cd114..988d7b151ec86b4835849d124745fbb8373ec123 100755 (executable)
--- a/tools/mkcharinfo
+++ b/tools/mkcharinfo
@@ -8,14 +8,14 @@ no if $] >= 5.018, warnings => 'experimental::smartmatch';
 use open OUT => ':utf8', ':std';
 use Data::Dump 'pp';
 
-our $VERSION = '1.00';
+our $VERSION = '1.01';
 
 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: ∤ |ͯ ⨝)
 );
@@ -48,16 +48,18 @@ eval {
 
 my %diinc = (
        'data/digraphs-rfc.inc.pl' => 'u-di',
-       'data/digraphs-shiar.inc.pl' => 'u-prop Xz',
+       'data/digraphs-shiar.inc.pl' => 'u-prop ex',
+       'data/digraphs-vim.inc.pl' => 'u-prop',
 );
-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}++;
        }
 }
 
@@ -82,6 +84,10 @@ for my $chr (keys %info) {
 
        $info->{$_} = $info{$chr}->{$_} for keys %{ $info{$chr} };
 
+       # official digraphs either lose vim flag or gain experimental
+       delete $info->{class}->{'u-prop'} or $info->{class}->{ex}++
+               if $info->{class}->{'u-di'};
+
        # categorise by unicode types and writing script
        $info->{class}->{$_}++ for $info->{category};
        $info->{class}->{$_}++ for $info->{script} || ();
@@ -90,15 +96,15 @@ for my $chr (keys %info) {
        $info->{class}->{Xa}++ if $info->{block} eq 'Basic Latin';
        $info->{class}->{Xl}++ if $info->{block} eq 'Latin-1 Supplement';
 
-       given ($cp) {
-               when ($info->{string}) {
+       {
+               if ($info->{string}) {
                        # keep predefined presentational string
                }
-               when ($info->{combining}) {
+               elsif ($info->{combining}) {
                        # overlay combining accents
                        $info->{string} = chr(9676) . $chr;
                }
-               when (($cp & ~0b1001_1111) == 0 or $cp == 127) {
+               elsif (($cp & ~0b1001_1111) == 0 or $cp == 127) {
                        # control characters (first 32 chars from 0 and 128)
                        # rename to something more descriptive
                        $info->{name} = $info->{unicode10}