git.shiar.nl
/
sheet.git
/ blobdiff
commit
grep
author
committer
pickaxe
?
search:
re
summary
|
shortlog
|
log
|
commit
|
commitdiff
|
tree
raw
|
inline
| side by side
digraphs: include flag legend in include data
[sheet.git]
/
tools
/
mkdigraphlist
diff --git
a/tools/mkdigraphlist
b/tools/mkdigraphlist
index 2d1fdbd8380b27cb4d620e5d99bee3ac110bee45..075d564b07d2079abe09a9e84087f2ee1f4cf235 100755
(executable)
--- a/
tools/mkdigraphlist
+++ b/
tools/mkdigraphlist
@@
-4,9
+4,10
@@
use strict;
use warnings;
use utf8;
use warnings;
use utf8;
-use open OUT => ':utf8', ':std';
+use open OUT => ':encoding(utf-8)', ':std';
+use JSON ();
-our $VERSION = '1.0
6
';
+our $VERSION = '1.0
7
';
# import and combine various digraph data
push @INC, 'data';
# import and combine various digraph data
push @INC, 'data';
@@
-24,18
+25,14
@@
my $di = { %{$vim // {}}, %{$rfc}, %{$extra // {}} };
my $uninfo = do 'unicode-char.inc.pl'
or warn "could not include unicode details: ", $@ // $!;
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}
);
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
ord $di->{$_}, # original code point
- map {"'$_'"}
$uninfo->{ $di->{$_} }->[1] // '', # name
join(' ',
$rfc->{$_}
$uninfo->{ $di->{$_} }->[1] // '', # name
join(' ',
$rfc->{$_}
@@
-45,8
+42,18
@@
printf "q{%s}=>[%s],\n", s/(?=[\\}])/\\/gr, join(',',
),
($uninfo->{ $di->{$_} }->[0] // '') =~ s/ u-di| u-prop| ex//gr, # class
$uninfo->{ $di->{$_} }->[4] // (), # string
),
($uninfo->{ $di->{$_} }->[0] // '') =~ s/ u-di| u-prop| ex//gr, # class
$uninfo->{ $di->{$_} }->[4] // (), # string
-) for sort keys %{$di};
-print "}\n";
+] for sort keys %{$di};
+
+print JSON->new->ascii->canonical->encode({
+ key => \%table,
+ flag => {
+ l4 => 'full support',
+ l3 => 'vim extension',
+ 'l3 ex' => 'vim v8.0',
+ l2 => 'proposal',
+ l1 => 'not in vim',
+ },
+});
__END__
__END__
@@
-56,15
+63,14
@@
mkdigraphlist - Output character list of combined digraph data
=head1 SYNOPSIS
=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.
=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
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
@@
-72,9
+78,11
@@
is appended: character name, usage classes, unicode classes,
and replacement output string.
For example:
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
}
=head1 AUTHOR