charset: improve arabic inheritance

[sheet.git] / charset.plp

diff --git a/charset.plp b/charset.plp
index e1efd83c0880e8885f06569486751206fb63e784..51c391c88c9b603d1e5f87efd4c3c251eb7f96b1 100644 (file)
--- a/charset.plp
+++ b/charset.plp
@@ -39,19 +39,23 @@ sub tabinput {
                                   cp737 cp869 cp866 cp855 cp862 cp864 )],
                aix        => [qw( cp1006 )],
                win        => [qw( cp1252 cp1250 cp1254 cp1257 cp1258 cp1253 cp1251 cp1255 cp1256 cp874 )],
+               mac        => [qw( MacRoman MacRomanian MacRumanian MacCroatian MacCentralEurRoman MacTurkish MacIcelandic MacSami
+                                  MacGreek MacCyrillic MacHebrew MacArabic MacFarsi MacThai )],
 
+               # languages
                westeur    => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )],
                centeur    => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB
                turkish    => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )],
                baltic     => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
                nordic     => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
                cyrillic   => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855 )], # MacUkrainian is broken
-               arabic     => [qw( iso-8859-6 cp1006 cp864 cp1256 MacArabic )], # MacFarsi same as MacArabic?
+               arabic     => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi )],
                greek      => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 )],
                hebrew     => [qw( iso-8859-8 cp1255 MacHebrew cp862 )],
                thai       => [qw( iso-8859-11 cp874 MacThai )],
                vietnamese => [qw( viscii cp1258 MacVietnamese )],
                symbol     => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )],
+
                # iso-code shorthand
                1 => 'westeur',
                2 => 'centeur',
@@ -83,10 +87,10 @@ sub tabinput {
                        'gsm0338'     => ['ascii' => '0-127'],
 
                        'iso-8859-2'  => ['iso-8859-1' => '160'],
-                       'iso-8859-3'  => ['iso-8859-1' => '160'],
+                       'iso-8859-3'  => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9
                        'iso-8859-4'  => ['iso-8859-2' => '160'],
                        'iso-8859-5'  => ['iso-8859-1' => '160'],
-                       'iso-8859-6'  => ['iso-8859-1' => '160'],
+                       'iso-8859-6'  => ['cp1256' => '128', 'iso-8859-1' => '160'],
                        'iso-8859-7'  => ['iso-8859-1' => '160'],
                        'iso-8859-8'  => ['iso-8859-1' => '160'],
                        'iso-8859-9'  => ['iso-8859-1' => '208-223+240'],
@@ -119,21 +123,27 @@ sub tabinput {
                        'cp775'       => ['cp850' => '128'],  # partial cp437
                        'cp866'       => ['cp437' => '128-175+224'],
                        'cp855'       => ['cp437' => '128'],
-                       'cp1006'      => ['cp437' => '128'],
+                       'cp1006'      => ['iso-8859-6' => '160', 'cp437' => '128'],
                        'cp737'       => ['cp437' => '128-175+224'],
                        'cp869'       => ['cp437' => '128'],
                        'cp862'       => ['cp437' => '128-159'],
-                       'cp864'       => ['cp437' => '128'],
+                       'cp864'       => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants
 
                        'koi8-u'      => ['koi8-r' => 128],
                        'koi8-f'      => ['koi8-u' => 128],
 
-                       'MacRomanian' => ['MacRoman' => '160'],
+                       'MacRomanian' => ['MacRoman' => '160-191+208-223'],
+                       'MacRumanian' => ['MacRomanian' => '160-191+208-223', 'MacRoman' => '160-191+208-223'],
                        'MacCroatian' => ['MacRoman' => '160'],
                        'MacCentralEurRoman' => ['MacRoman' => '128'],
+                       'MacIcelandic'=> ['MacRoman' => '160-175+208-239'], #TODO: gaps at C/E
                        'MacTurkish'  => ['MacRoman' => '208-223'], # F5 is unassigned
+                       'MacSami'     => ['MacIcelandic' => '144', 'MacRoman' => '144'],
+                       'MacGreek'    => ['MacRoman' => '128'],
                        'MacCyrillic' => ['MacRoman' => '128'],
-                       'MacHebrew'   => ['MacRoman' => '128'],
+                       'MacHebrew'   => ['iso-8859-8' => '128', 'MacRoman' => '128-143+160'], # partial ascii
+                       'MacArabic'   => ['iso-8859-6' => '128', 'cp864' => '128', 'MacRoman' => '128'], #TODO: multiple parents
+                       'MacFarsi'    => ['MacArabic' => '176-191', 'MacRoman' => '128'],
                };
 
                my @parents = @{ $INHERIT->{$input} || [] };
@@ -204,7 +214,7 @@ sub tabinput {
                        $row{endpoint} = 255;
                }
                else {
-                       if ($row{set} eq 'MacHebrew') {
+                       if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') {
                                # array of possibly multiple characters per code point
                                $row{table} = [
                                        map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $endpoint