index: release v1.18 with only altgr index linked

[sheet.git] / tools / mkdigraphlist

diff --git a/tools/mkdigraphlist b/tools/mkdigraphlist
index 38978b62d563279f3ca0e5e9311256f6fa1e733f..5117b7690117cb12bed52963260d0d885a621195 100755 (executable)
--- a/tools/mkdigraphlist
+++ b/tools/mkdigraphlist
@@ -4,48 +4,71 @@ use strict;
 use warnings;
 use utf8;
 
-use open OUT => ':utf8', ':std';
+use open OUT => ':encoding(utf-8)', ':std';
+use JSON ();
 
-our $VERSION = '1.06';
+our $VERSION = '1.07';
 
 # import and combine various digraph data
-my $rfc = do 'data/digraphs-rfc.inc.pl'
+push @INC, 'data';
+my $rfc = do 'digraphs-rfc.inc.pl'
        or die "error reading digraphs include: ", $@ // $!;
-my $extra = do 'data/digraphs-shiar.inc.pl'
+my $extra = do 'digraphs-shiar.inc.pl'
        or warn "could not include shiar proposals: ", $@ // $!;
-my $vim = do 'data/digraphs-vim.inc.pl'
+my $vim = do 'digraphs-vim.inc.pl'
        or warn "could not include vim extensions ", $@ // $!;
-my $vimold = do 'data/digraphs-vim-74.inc.pl'
+my $vimold = do 'digraphs-vim-74.inc.pl'
        or warn "could not include vim compatibility ", $@ // $!;
 my $di = { %{$vim // {}}, %{$rfc}, %{$extra // {}} };
 
 # optionally get unicode character information
-my $uninfo = do 'data/unicode-char.inc.pl'
+my $uninfo = do 'unicode-char.inc.pl'
        or warn "could not include unicode details: ", $@ // $!;
 
-# output perl code of hash
-# (assume no backslashes or curlies, so we can just q{} w/o escaping)
-print "# automatically generated by $0\n";
-print "use utf8;\n";
-print "+{\n";
-printf '(map {$_=>0} qw{%s}),'."\n", join(' ',
+# output json map of character info
+my %table;
+$table{$_} = 0 for (
        grep { !defined $di->{$_} }
        map { substr($_, 1, 1).substr($_, 0, 1) } sort keys %{$di}
 );
-printf "q{%s}=>[%s],\n", s/(?=[\\}])/\\/gr, join(',',
+$table{$_} = [
        ord $di->{$_},   # original code point
-       map {"'$_'"}
        $uninfo->{ $di->{$_} }->[1] // '',  # name
-       join(' ',
+       (
                $rfc->{$_}
-                       ? $vim->{$_} ? 'l4' : 'l1'  # vim+rfc or rfc only
+                       ? $vim->{$_} ? 'l5' : 'l1'  # vim+rfc or rfc only
+                       : $vimold && $vimold->{$_} ? 'l4'  # compat vim if known
                        : $vim->{$_} ? 'l3' : 'l2', # vim only or neither
-               $vimold && $vim->{$_} && !$vimold->{$_} ? 'ex' : (), # new vim feature
        ),
-       ($uninfo->{ $di->{$_} }->[0] // '') =~ s/ u-di| u-prop| ex//gr,  # class
+       ($uninfo->{ $di->{$_} }->[0] // '') =~ s/ u-di| u-prop//gr,  # class
        $uninfo->{ $di->{$_} }->[4] // (),  # string
-) for sort keys %{$di};
-print "}\n";
+] for sort keys %{$di};
+
+print JSON->new->ascii->canonical->encode({
+       title => 'RFC-1345',
+       key  => \%table,
+       intro => join("\n",
+               'Character mnemonics following compose key ⎄:',
+               'i^k in <a href="/vi">Vim</a>,',
+               '^u^\ in <a href="/readline">Emacs</a>,',
+               '^a^v in <a href="/screen">Screen</a>.',
+               'Similar but different from <a href="/digraphs/xorg">X.Org</a>.',
+               'Also see <a href="/unicode">common Unicode</a>.</p>',
+               '<p class="aside">Unofficial <span class="u-l2">proposals</span>',
+               'are available as <a href="/digraphs.vim">ex commands</a>.',
+       ),
+       flag => {
+               l5 => 'full support',
+               l4 => 'vim extension',
+               l3 => 'vim v8.0',
+               l2 => 'proposal',
+               l1 => 'not in vim',
+       },
+       flagclass => {
+               l5 => '', # common
+               l3 => 'u-l5', # rare
+       },
+});
 
 __END__
 
@@ -55,15 +78,14 @@ mkdigraphlist - Output character list of combined digraph data
 
 =head1 SYNOPSIS
 
-    mkdigraphlist >digraphs.inc.pl
-    perl -e'$di = do "digraphs.inc.pl"; print chr $di->{DO}->[0]'
+    mkdigraphlist | jq -r '.key."DO"[0]' | perl -nE 'say chr' # $
 
 =head1 DESCRIPTION
 
 Combines precompiled digraph includes of rfc (1345), vim, and shiar
 and outputs a complete map including character details and usage classes.
 
-The value can either be a scalar string containing another
+The C<key> values can either be a scalar string containing another
 digraph which can be considered identical (usually inverted),
 or an array ref containing at least the resulting character's
 Unicode code point value.  If available, the following UCD data
@@ -71,9 +93,11 @@ is appended: character name, usage classes, unicode classes,
 and replacement output string.
 For example:
 
- +{
-   AE => [198, 'LATIN CAPITAL LETTER AE', 'u-di', 'Latin Lu Xl u-v11'],
-   EA => 'AE',
+  {
+   "key": {
+    "AE" => [198, "LATIN CAPITAL LETTER AE", "u-di", "Latin Lu Xl u-v11"],
+    "EA" => "AE",
+   }
   }
 
 =head1 AUTHOR