unicode: mark missing digraphs in multiple characters

[sheet.git] / Shiar_Sheet / FormatChar.pm

diff --git a/Shiar_Sheet/FormatChar.pm b/Shiar_Sheet/FormatChar.pm
index a3f4cc9c3725bdee642fecceb91ae5b79c1b29ca..ee0c31065f9c7e716dafb92b38960d8af0a8b911 100644 (file)
--- a/Shiar_Sheet/FormatChar.pm
+++ b/Shiar_Sheet/FormatChar.pm
@@ -1,7 +1,9 @@
 package Shiar_Sheet::FormatChar;
 
+use 5.010;
 use strict;
 use warnings;
+use utf8;
 
 use Data::Dump 'pp';
 use PLP::Functions 'EscapeHTML';
@@ -50,7 +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
        );
 }
 
@@ -70,7 +72,7 @@ sub cell {
                push @class, 'u-invalid';
                $cell = '';
        }
-       else {
+       else {{
                push @class, 'X';
 
                if ($input =~ s/^-//) {
@@ -79,37 +81,68 @@ sub cell {
 
                $input =~ s/^\\//;  # escaped char
                ($cell, $title, my $class, $mnem, $entity) = $self->glyphs_html($input);
+               my $codepoint = ord $input;
+
+               if ($self->{style} eq 'univer') {
+                       if ($input =~ /\p{age=unassigned}/) {
+                               # check include for assignments after unicode 6.0 (perl v5.14)
+                               state $agemap = do 'unicode-age.inc.pl';
+                               my $version = $agemap->{$codepoint};
+                               push @class, $version ? 'l2' : 'l1';
+                       }
+                       elsif ($input =~ /^\p{in=1.1}*$/) {
+                               push @class, 'l5';  # first release 1993
+                       }
+                       elsif ($input =~ /^\p{in=3.0}*$/) {
+                               push @class, 'l4';  # 20th century
+                       }
+                       elsif ($input =~ /^\p{in=4.1}*$/) {
+                               push @class, 'l4';  # over 10 years ago
+                       }
+                       elsif ($input =~ /^\p{in=6.0}*$/) {
+                               push @class, 'l3';  # before 2012
+                       }
+                       else {
+                               push @class, 'l2';  # more recent
+                       }
+                       next;
+               }
 
                if ($self->{style} eq 'di') {
-                       if ($class =~ /\bu-di\b/) {
-                               push @class, ('l3', 'u-di'); # standard digraph
+                       if ($mnem =~ /…/) {
+                               # incomplete representation, usually partial
+                       }
+                       elsif ($class =~ /\bu-di\b/) {
+                               push @class, ('l4', 'u-di'); # standard digraph
                        }
                        elsif ($class =~ /\bu-prop\b/) {
-                               push @class, ('l2', 'u-prop'); # unofficial
+                               push @class, ('l3', 'u-prop'); # unofficial
                        }
                }
                elsif ($self->{style} eq 'html') {
                        if (defined $entity) {
-                               push @class, ('l3', 'u-html');
+                               push @class, ($codepoint <= 0xFF ? 'l4' : 'l3', 'u-html');
                        }
                }
                else {
-                       my $codepoint = ord(substr $input, 0, 1);
                        if ($codepoint <= 0xFF) {
-                               push @class, 'l3', 'u-lat1';  # latin1
+                               push @class, 'l4', 'u-lat1';  # latin1
                        }
                        elsif ($codepoint <= 0xD7FF) {
-                               push @class, 'l2', 'u-bmp';  # bmp
+                               push @class, 'l3', 'u-bmp';  # bmp
                        }
                }
 
                if ($input =~ /[ -~]/) {
-                       push @class, 'l4', 'u-ascii'; # ascii
+                       push @class, 'l5', 'u-ascii'; # ascii
+               }
+               elsif ($input =~ /^\p{in=6.0}+$/) {
+                       push @class, 'l2'; # in unicode 6.0
                }
                else {
-                       push @class, 'l1'; # basic unicode
+                       push @class, 'l1'; # any unicode
                }
-       }
+       }}
 
        my $anno = '';
        if ($cell ne '') {