Figure out all possible alternative encodings for a unicode string.

possible_encodings.py

# coding=utf-8
import pkgutil
import encodings

# List of possible codecs.
# Multibyte unicode is unlikely. Along with other...
unlikely_codecs = {"bz2_codec", "uu_codec", "base64_codec", "hex_codec", "zlib_codec", "unicode_escape",
                   "unicode_internal", "utf_16", "utf_16_le", "utf_16_be", "utf_32", "utf_32_be", "utf_32_le",
                   "raw_unicode_escape","punycode","palmos","idna"}
found = set(name for imp, name, ispkg in pkgutil.iter_modules(encodings.__path__) if not ispkg)
found.difference_update(unlikely_codecs)


def find_unique_encodings(in_word):
    """
    Return all possible alternative encodings for a given string.
    :param in_word: utf_string
    :return: set of all possible encodings minus the unlikely_codecs
    In [6]: find_unique_encodings(u"Limón")
    Out[6]:
    {'Lim\x1b$(D+Q\x1b(Bn',
     'Lim\x1b$(O)i\x1b(Bn',
     'Lim\x1b$(Q)i\x1b(Bn',
     'Lim+APM-n',
     'Lim~{(.~}n',
     'Lim\x85\x89n',
     'Lim\x88un',
     'Lim\x8f\xab\xd1n',
     'Lim\x97n',
     'Lim\xa2n',
     'Lim\xa8\xaen',
     'Lim\xa9\xe9n',
     'Lim\xc3\xb3n',
     'Lim\xc6n',
     'Lim\xf3n',
     'Yvz\xf3a',
     '\xd3\x89\x94\xce\x95',
     '\xef\xbb\xbfLim\xc3\xb3n'}
    """
    out_encodings = set()
    for encoding in found:
        # Bad
        try:
            # print "{} {}".format(encoding, in_word.encode(encoding))
            out_encodings.add(in_word.encode(encoding))
        except:
            pass
    return out_encodings