From b3b71de70835c9edbdf320035d5bb54d4c31db46 Mon Sep 17 00:00:00 2001
From: Mischa POSLAWSKY <perl@shiar.org>
Date: Thu, 25 Jan 2024 22:02:20 +0100
Subject: [PATCH] keyboard: split character formatting fallback method

Prepare uncached Shiar_Sheet/FormatChar::glyph_info code path for future use
in similar tools/mkcharinfo generation.
---
 Shiar_Sheet/FormatChar.pm | 25 ++++++++++++++++++-------
 tools/mkcharinfo          |  1 +
 2 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/Shiar_Sheet/FormatChar.pm b/Shiar_Sheet/FormatChar.pm
index 3d9b6fb..f471497 100644
--- a/Shiar_Sheet/FormatChar.pm
+++ b/Shiar_Sheet/FormatChar.pm
@@ -8,7 +8,7 @@ use utf8;
 use Data::Dump 'pp';
 use PLP::Functions 'EscapeHTML';
 
-our $VERSION = '1.09';
+our $VERSION = '1.10';
 
 our $uc = do 'data/unicode-char.inc.pl';
 
@@ -17,14 +17,25 @@ sub new {
 	bless { anno => ['di', 0], style => 'di' }, $class;
 }
 
-sub glyph_info {
+sub glyph_mkinfo {
 	my ($self, $codepoint) = @_;
-	return $uc->{chr $codepoint} || eval {
+	# attempt to get unicode character information
+	my $info = eval {
 		require Unicode::UCD;
-		if (my $fullinfo = Unicode::UCD::charinfo($codepoint)) {
-			return [@$fullinfo{qw/category name - string/}];
-		}
-	} || [];
+		Unicode::UCD::charinfo($codepoint)
+			|| { category => 'Xn', name => '' };
+	} or return;
+	my $string;
+	if ($info->{combining}) {
+		# overlay combining diacritics
+		$string = chr(9676) . chr($codepoint);
+	}
+	return [@$info{qw( category name )}, undef, $string];
+}
+
+sub glyph_info {
+	my ($self, $codepoint) = @_;
+	return $uc->{chr $codepoint} || $self->glyph_mkinfo($codepoint) || [];
 }
 
 sub glyph_html {
diff --git a/tools/mkcharinfo b/tools/mkcharinfo
index ace9dbe..da6aa87 100755
--- a/tools/mkcharinfo
+++ b/tools/mkcharinfo
@@ -84,6 +84,7 @@ eval {
 
 for my $chr (keys %info) {
 	my $cp = ord $chr;
+	#my $info = glyph_mkinfo($cp) or next;
 	# attempt to get unicode character information
 	my $info = eval {
 		require Unicode::UCD;
-- 
2.30.0