_compat: Change the to_native_str into a to_unicode_str function.

The to_native_str function was supposed to produce Unicode decoded
native strings for python2 and python3.

Upon further consideration this doesn't make much sense since under
python2 it would need to decode the bytes into a Unicode string and turn
it back into a python2 str.

The ability to use the replacement character requires us to use a
Unicode string under python2 as well.

python/olm/_compat.py

@@ -48,8 +48,11 @@ def to_bytes(string):
     raise TypeError("Invalid type {}".format(type(string)))
 
 
-def to_native_str(byte_string, errors="replace"):
+def to_unicode_str(byte_string, errors="replace"):
-    """Turn a byte string into a native string decoding it as UTF-8.
+    """Turn a byte string into a unicode string.
+
+    Should be used everywhere where the input byte string might not be trusted
+    and may contain invalid unicode values.
 
     Args:
         byte_string (bytes): The bytestring that will be converted to a native
@@ -63,7 +66,4 @@ def to_native_str(byte_string, errors="replace"):
 
     Returns the decoded native string.
     """
-    try:
+    return byte_string.decode(errors=errors)
-        return native_str(byte_string, errors=errors)
-    except TypeError:
-        return bytes(byte_string).decode(errors=errors)

python/olm/group_session.py

@@ -33,7 +33,7 @@ from future.utils import bytes_to_native_str
 # pylint: disable=no-name-in-module
 from _libolm import ffi, lib  # type: ignore
 
-from ._compat import URANDOM, to_bytearray, to_bytes, to_native_str
+from ._compat import URANDOM, to_bytearray, to_bytes, to_unicode_str
 from ._finalize import track_for_finalization
 
 
@@ -230,7 +230,7 @@ class InboundGroupSession(object):
 
         self._check_error(plaintext_length)
 
-        plaintext = to_native_str(
+        plaintext = to_unicode_str(
             ffi.unpack(plaintext_buffer, plaintext_length),
             errors=errors
         )

python/olm/pk.py

@@ -40,7 +40,7 @@ from future.utils import bytes_to_native_str
 
 from _libolm import ffi, lib  # type: ignore
 
-from ._compat import URANDOM, to_bytearray, to_native_str
+from ._compat import URANDOM, to_bytearray, to_unicode_str
 from ._finalize import track_for_finalization
 
 
@@ -361,7 +361,7 @@ class PkDecryption(object):
         # clear out copies of the plaintext
         lib.memset(plaintext_buffer, 0, max_plaintext_length)
 
-        return to_native_str(plaintext, errors=errors)
+        return to_unicode_str(plaintext, errors=errors)
 
 
 def _clear_pk_signing(pk_struct):

python/olm/session.py

@@ -40,7 +40,7 @@ from future.utils import bytes_to_native_str
 # pylint: disable=no-name-in-module
 from _libolm import ffi, lib  # type: ignore
 
-from ._compat import URANDOM, to_bytearray, to_bytes, to_native_str
+from ._compat import URANDOM, to_bytearray, to_bytes, to_unicode_str
 from ._finalize import track_for_finalization
 
 # This is imported only for type checking purposes
@@ -318,7 +318,7 @@ class Session(object):
             plaintext_buffer, max_plaintext_length
         )
         self._check_error(plaintext_length)
-        plaintext = to_native_str(
+        plaintext = to_unicode_str(
             ffi.unpack(plaintext_buffer, plaintext_length),
             errors=errors
         )