db/utils: correctly handle 8bit encodings.

Because python's mail will use raw-unicode-escape for anything that
isn't ascii we need to encode back into the original raw bytes, then
decode into the proper encoding.

1 files changed, 6 insertions, 1 deletions

diff --git a/alot/db/utils.py b/alot/db/utils.py
index 7b2b2979..c49019ce 100644
--- a/alot/db/utils.py
+++ b/alot/db/utils.py
@@ -358,11 +358,16 @@ def extract_body(mail, types=None, field_key='copiousoutput'):
         enc = part.get_content_charset() or 'ascii'
         cte = str(part.get('content-transfer-encoding', '7bit')).lower()
         payload = part.get_payload()
-        if cte not in ['7bit', '8bit']:
+        if cte != '7bit':
             if cte == 'quoted-printable':
                 raw_payload = quopri.decodestring(payload.encode('ascii'))
             elif cte == 'base64':
                 raw_payload = base64.b64decode(payload)
+            elif cte == '8bit':
+                # Python's mail library will decode 8bit as raw-unicode-escape,
+                # so we need to encode that back to bytes so we can decode it
+                # using the correct encoding.
+                raw_payload = payload.encode('raw-unicode-escape')
             else:
                 raise Exception('Unknown Content-Transfer-Encoding {}'.format(cte))
             # message.get_payload(decode=True) also handles a number of unicode