7 use open OUT => ':utf8', ':std';
11 # import and combine various digraph data
12 my $rfc = do 'data/digraphs-rfc.inc.pl'
13 or die "error reading digraphs include: ", $@ // $!;
14 my $extra = do 'data/digraphs-shiar.inc.pl'
15 or warn "could not include shiar proposals: ", $@ // $!;
16 my $vim = do 'data/digraphs-vim.inc.pl'
17 or warn "could not include vim extensions ", $@ // $!;
18 my $vimold = do 'data/digraphs-vim-74.inc.pl'
19 or warn "could not include vim compatibility ", $@ // $!;
20 my $di = { %{$vim // {}}, %{$rfc}, %{$extra // {}} };
22 # optionally get unicode character information
23 my $uninfo = do 'unicode-char.inc.pl'
24 or warn "could not include unicode details: ", $@ // $!;
26 # output perl code of hash
27 # (assume no backslashes or curlies, so we can just q{} w/o escaping)
28 print "# automatically generated by $0\n";
31 printf '(map {$_=>0} qw{%s}),'."\n", join(' ',
32 grep { !defined $di->{$_} }
33 map { substr($_, 1, 1).substr($_, 0, 1) } sort keys %{$di}
35 printf "q{%s}=>[%s],\n", s/(?=[\\}])/\\/gr, join(',',
36 ord $di->{$_}, # original code point
38 $uninfo->{ $di->{$_} }->[1] // '', # name
41 ? $vim->{$_} ? 'l4' : 'l1' # vim+rfc or rfc only
42 : $vim->{$_} ? 'l3' : 'l2', # vim only or neither
43 $vimold && $vim->{$_} && !$vimold->{$_} ? 'ex' : (), # new vim feature
45 ($uninfo->{ $di->{$_} }->[0] // '') =~ s/ u-di| u-prop| ex//gr, # class
46 $uninfo->{ $di->{$_} }->[4] // (), # string
47 ) for sort keys %{$di};
54 mkdigraphlist - Output character list of combined digraph data
58 mkdigraphlist >digraphs.inc.pl
59 perl -e'$di = do "digraphs.inc.pl"; print chr $di->{DO}->[0]'
63 Combines precompiled digraph includes of rfc (1345), vim, and shiar
64 and outputs a complete map including character details and usage classes.
66 The value can either be a scalar string containing another
67 digraph which can be considered identical (usually inverted),
68 or an array ref containing at least the resulting character's
69 Unicode code point value. If available, the following UCD data
70 is appended: character name, usage classes, unicode classes,
71 and replacement output string.
75 AE => [198, 'LATIN CAPITAL LETTER AE', 'u-di', 'Latin Lu Xl u-v11'],
81 Mischa POSLAWSKY <perl@shiar.org>
85 Licensed under the GNU Affero General Public License version 3.