From 2876b76b7b654f43dd1668c4a963a049450ed66c Mon Sep 17 00:00:00 2001
From: Mischa POSLAWSKY <perl@shiar.org>
Date: Sat, 21 Feb 2015 05:33:12 +0100
Subject: [PATCH] unicode: mark missing digraphs in multiple characters

---
 Shiar_Sheet/FormatChar.pm | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/Shiar_Sheet/FormatChar.pm b/Shiar_Sheet/FormatChar.pm
index 8b00318..ee0c310 100644
--- a/Shiar_Sheet/FormatChar.pm
+++ b/Shiar_Sheet/FormatChar.pm
@@ -3,6 +3,7 @@ package Shiar_Sheet::FormatChar;
 use 5.010;
 use strict;
 use warnings;
+use utf8;
 
 use Data::Dump 'pp';
 use PLP::Functions 'EscapeHTML';
@@ -51,7 +52,7 @@ sub glyphs_html {
 		EscapeHTML($_[0]), # cell
 		join(' | ', map { $_->[1] } @chars), # title
 		$chars[0][2], # class
-		join(' ',  grep { defined } map { $_->[3] } @chars), # digraph
+		join(' ',  map { $_->[3] // '…' } @chars), # digraph
 	);
 }
 
@@ -80,7 +81,7 @@ sub cell {
 
 		$input =~ s/^\\//;  # escaped char
 		($cell, $title, my $class, $mnem, $entity) = $self->glyphs_html($input);
-		my $codepoint = ord(substr $input, 0, 1);
+		my $codepoint = ord $input;
 
 		if ($self->{style} eq 'univer') {
 			if ($input =~ /\p{age=unassigned}/) {
@@ -108,7 +109,10 @@ sub cell {
 		}
 
 		if ($self->{style} eq 'di') {
-			if ($class =~ /\bu-di\b/) {
+			if ($mnem =~ /…/) {
+				# incomplete representation, usually partial
+			}
+			elsif ($class =~ /\bu-di\b/) {
 				push @class, ('l4', 'u-di'); # standard digraph
 			}
 			elsif ($class =~ /\bu-prop\b/) {
-- 
2.30.0