Hello, Dan! 1)
This is a small reminder: --- ext/Encode/t/Aliases.t.orig Sat Mar 30 01:05:57 2002 +++ ext/Encode/t/Aliases.t Thu Apr 4 15:56:21 2002 @@ -29,6 +29,7 @@ 'arabic' => 'iso-8859-6', 'greek' => 'iso-8859-7', 'hebrew' => 'iso-8859-8', + 'Hebrew' => 'iso-8859-8', 'thai' => 'iso-8859-11', 'tis620' => 'iso-8859-11', 'WinLatin1' => 'cp1252', @@ -107,6 +108,9 @@ define_alias( qr/shift.*jis$/i => '"MacJapanese"', qr/sjis$/i => '"cp932"' ); @a2c{qw(Shift_JIS x-sjis)} = qw(MacJapanese cp932); + define_alias( 'hebrew' => 'UTF-8' ); + + @a2c{qw(hebrew Hebrew)} = qw(UTF-8 UTF-8); } print "# alias test with alias overrides\n"; still fails. There was a letter from me headed [PATCH] again! (was: Encode alias implementation fixed!) that had a proposed patch to Aliases.pm The patch is no longer directly applicable, but looks like the trouble that it tried to resolve has not gone yet. Did I catch your intention correctly? Should this test that I have added to Aliases.t succeeded? 2) The aliases are case-insensitive. The canonical names are not. (I can see 'koi8-r' works but not 'KOI8-R') This is not urgent, but is there an intention to empower 'KOI8-R'? - Anton