font: aggregate counts for unicode versions

[sheet.git] / tools / mkfontinfo

diff --git a/tools/mkfontinfo b/tools/mkfontinfo
index 3be474c26c3f4d72fde4249694383021ed94441b..ed328cf8ea2c0c417fce1f987d25932b44371342 100755 (executable)
--- a/tools/mkfontinfo
+++ b/tools/mkfontinfo
@@ -18,6 +18,12 @@ for my $fontfile (glob 'ttfsupport/*'.$incsuffix) {
        my ($fontid) = basename($fontfile, $incsuffix);
        my ($fontmeta, @fontrange) = do $fontfile or next;
        $fontmeta->{file} = $fontid;
+       my $year = substr $fontmeta->{date}, 0, 4;
+       $fontmeta->{description} = join(' ',
+               (map { "version $_" } $fontmeta->{version} || ()),
+               $fontmeta->{version} && $fontmeta->{version} =~ /\Q$year/ ? () :
+               (map { "($_)" } $year || ()),
+       );
        push @fontlist, $fontmeta;
        $cover{$fontid} = { map { (chr $_ => 1) } @fontrange };
 }
@@ -57,13 +63,21 @@ eval {
        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 );
-}
+eval {
+       my $agemap = do 'unicode-age.inc.pl'
+               or warn "Could not include unicode version data: $!";
+
+       use Unicode::UCD 'charinfo';
+       for my $code (0 .. 256**2*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 );
+               push @{ $charlist{version}->{$_} }, (chr $code) x ($agemap->{$code} <= $_)
+                       for 11, 30, 63;
+       }
+       1;
+} or warn "Could not include unicode groups: $@";
 
 for (values %charlist) {
 for my $chars (values %{$_}) {
@@ -100,20 +114,29 @@ for my $chars (values %{$_}) {
 $charlist{fonts} = \@fontlist;
 
 my %osfonts = (
-       win95   => [qw( arial arialuni lucidau verdana timesnew couriernew )],  # microsoft
-       mac10   => [qw( helvetica lucida times garamond palatino )],  # apple
-       android => [qw( roboto noto )],  # google
-       oss     => [qw( dvsans c2k unifont opensans )],
+       win95   => [qw( arial.win95 arialuni lucidau verdana.win95 times.win95 cour.win95 )],  # microsoft
+       win8    => [map {"$_.win8"} qw( arial verdana times georgia pala cour )],
+       mac109  => [map {"$_.mac109"} qw( helv lucida times pala )],  # apple
+       android => [qw( roboto droidmono notosans )],  # google
+       oss     => [qw( dvsans freesans code2000 unifont )],
 );
+if (0) {
+       # copy rows to derive older os versions (same list with different trailing number)
+       s/8$/7/ for @{ $osfonts{  win7} = [@{ $osfonts{  win8} }] };
+       s/9$/7/ for @{ $osfonts{mac107} = [@{ $osfonts{mac109} }] };
+}
+
 my %fontnum = map { ($fontlist[$_]->{file} => $_) } 0 .. $#fontlist;
 while (my ($os, $fontids) = each %osfonts) {
        $charlist{os}->{$os} = [ map { $fontnum{$_} // () } @{$fontids} ];
 }
-$charlist{osdefault} = [qw( win95 mac10 oss android )];
+$charlist{osdefault} = [qw( win95 win8 mac109 android oss )];
 
 say "# automatically generated by $0";
 say 'use utf8;';
-say '+'.pp(\%charlist);
+say '+', pp(\%charlist) =~ s{
+       ( \[ \s* \d [^]]* ) ,\s* (?= \] )  # arrays of numbers, excluding trailing comma
+}{ $1 =~ s/\s+//gr }msxgre;  # strip whitespace
 
 __END__