--------------------------------------------------------------------------------
<pre>
$ cat /etc/redhat-release
CentOS Linux release 7.3.1611 (Core)

$ irb
1.9.3-p551 :001 > str = "test".force_encoding("UTF-7")
=> "\x74\x65\x73\x74"
1.9.3-p551 :002 > str = str.encode("UTF-8", :invalid => :replace,:undef => :replace, :replace => '?')
Encoding::ConverterNotFoundError: code converter not found (UTF-7 to UTF-8)
from (irb):2:in `encode'
from (irb):2
from /home/xxxx/.rvm/rubies/ruby-1.9.3-p551/bin/irb:12:in `<main>'
</pre>
--------------------------------------------------------------------------------
It seems Ruby does not have UTF-7 converter.

"Code converter not found (UTF-8 to EUC-TW)": https://bugs.ruby-lang.org/issues/6995
--------------------------------------------------------------------------------
This "encyclopedia":https://www.everipedia.com/UTF7/ article for UTF7 seems to be a good place to start
--------------------------------------------------------------------------------
@Toshi: utf-7 is relevant only for email processing. Ruby still supports it, but it's a little bit tricky. This will work:

<pre>
require "net/imap"
Net::IMAP.decode_utf7(mail_body)
</pre>

--------------------------------------------------------------------------------
The same problem with _Windows-1258_. Attached patch should cover all not supported code pages.
--------------------------------------------------------------------------------
@karel
could you add a more realistic test case with an email as a fixture?
https://github.com/redmine/redmine/blob/3ac569c4d8c8b04a9cc33e3c2a05f8553d99827e/test/unit/mail_handler_test.rb#L542

it should be already fixed by
https://github.com/mikel/mail/pull/1112
--------------------------------------------------------------------------------
A test with a real Vietnamese email added into the patch.
--------------------------------------------------------------------------------