charset: replace private use glyphs in MacHebrew

[sheet.git] / charset-encoding.inc.pl

diff --git a/charset-encoding.inc.pl b/charset-encoding.inc.pl
index 71ba2bae277446e4a68666dab5081b1cdc3875ef..c402187a201d99acca878351470c15542a8767bd 100644 (file)
--- a/charset-encoding.inc.pl
+++ b/charset-encoding.inc.pl
@@ -131,7 +131,15 @@ use utf8;
        'macsami'      => {inherit => ['MacIcelandic' => '90-BF+D0-DF+F0', 'MacRoman' => '90']},
        'macgreek'     => {inherit => ['MacRoman' => '80']},
        'maccyrillic'  => {inherit => ['MacRoman' => '80']},
-       'machebrew'    => {inherit => ['iso-8859-8' => '80', 'MacRoman' => '80-8F+A0']}, # partial ascii
+       'machebrew'    => {
+               inherit => ['iso-8859-8' => '80', 'MacRoman' => '80-8F+A0'], # ignore partial ascii
+               replace => {
+                       # strip private use characters for unneeded roundtrip
+                       0xDE => chr(0x5C7), # qamats qatan
+                       0xC0 => 'לֹ', # lamed holam
+                       #TODO: private use for canoral codes (obsolete nikud positioning)
+               },
+       },
        'macarabic'    => {inherit => ['iso-8859-6' => '80', 'cp864' => '80', 'MacRoman' => '80']}, #TODO: multiple parents
        'macfarsi'     => {inherit => ['MacArabic' => 'B0-BF', 'MacRoman' => '80']},
 
@@ -188,7 +196,6 @@ use utf8;
                $row->{set} = 'UTF-8';
                $row->{cell} = do 'charset-utf8.inc.pl'
                        or Alert('Table data could not be read', $@ || $!);
-               $row->{endpoint} = 0xFF;
                return $row;
        }},
        'utf-8'        => 'utf8',