note 96643 deleted from function.imap-utf8 by danbrown

Note Submitter: tomlove at gmail dot com ---- This function strictly observes parts of RFC 2047 to the extent that it may not decode encoded-words produced by non-compliant composing agents. 1. It is strict about a 76-character maximum length for encoded-words. Gmail may produce an encoded-word longer than this, and this will be passed through without decoding. 2. It is strict about the rule that encoded-words be separated by linear-white-space from adjacent text. Encoded-words not so separated will be ignored. 3. Invalid "B" encodings will cause this function to return without decoding ANY encoded-words in the string (including any valid ones). But... 3. Invalid "Q" encodings are decoded with invalid characters (e.g. > 127) still in place. The advantage of this function over mb_decode_mimeheader() is that it will decode any character set, including non-existent ones (by assuming UTF-8).