Bug #43494 for Encode-IMAPUTF7: National chars are encoded incorrectly with module ver. 1.01

I don't think this is a bug in Encode::IMAPUTF7 but an utf8 error on you side. All the below examples returns the expected '&AWE-' encode('IMAP-UTF-7', "\x{0161}"); # Unicode small letter s with caron encode("IMAP-UTF-7", decode("ISO-8859-2", "\xB9")); encode("IMAP-UTF-7", decode("ISO-8859-15", "\xA8")); To get you result I have to do something like encode('IMAP-UTF-7', "\xC5\xA1"); # Raw UTF8 small letter s with caron with an octet string consisting of the utf8 encoded letter. Upgrading to a real perl unicode-string makes encode work as expected: encode('IMAP-UTF-7', decode_utf8("\xC5\xA1")); I'm not really sure how the changes between 1.00 and 1.01 should explain why it worked otherwise before. I don't have easy access to a pre-5.10 perl where Encode::IMAPUTF7 worked, so I havn't tested the old version.

Yes, I agree it was bad example in my case. I was missing "use utf8", because the command contains char encoded in utf8. My original problem was in this command: encode('IMAP-UTF-7', encode("utf-8", $folder_name)) I used it for encoding of folder names get from MS Outlook via Win32::OLE. I'm using localised Outlook and it sometimes returns badly UTF-8 encoded folder names. Therefore I had to use workaround 'encode("utf-8", $string)' for such strings. Whole command worked well with IMAPUTF7 ver 1.00. Since 1.01 it returns another string. But I have found that with 1.01 I can use pure command 'encode('IMAP-UTF-7', $string)' and it works well as before. Therefore I think you can close this bug as NOTABUG - removing of my workaround fixes this problem and works well with IMAPUTF7 ver 1.01