Friendly way to decode Envelope fields

[mjs]

Originally reported by: Vic Tor (Bitbucket: maxiimou)

---

Hello

I've been playing with imap and python, and use IMAPClient for little project. I'm trying to get out of encoding problems in Enveloppe structure. I wish I could have an easier way to access decoded string, here's what i have in mind.

output = parse_fetch_response([envelope_str])

subject = output[76920][b'ENVELOPE'].subject
assert subject == b'=?utf-8?B?VsOpbMO0VG91bG91c2U=?='
assert subject.decoded() == u'VélôToulouse'

One possible way to do that would be by subclassing the bytes type:

class imapbytes(bytes):
    def decoded(self, imap_encoding='us-ascii'):
        pass

with the following specifications so the current api is not broken:

somebytes = b"hey !"
someimapbytes = imapbytes(somebytes)
assert isinstance(somebytes, bytes)
assert not isinstance(somebytes, imapbytes)
assert isinstance(someimapbytes, bytes)
assert somebytes == someimapbytes

I forked your repo so i can test a bit further the idea. I'm not used to Mercurial so I couldn't push my changes on bitbuket but I managed to push them on github on the decode_imap_bytes branch. It seems to work the expected way on both python 2.7 and 3.4.

Could you tell me what you think of this improvement ? I wish I could contribute here =)

---

• Bitbucket: https://bitbucket.org/mjs0/imapclient/issue/152

[mjs]

Original comment by Vic Tor (Bitbucket: maxiimou):

---

Hello Menno, I'm sorry I didn't come back to you earlier. I did continue a bit this feature but I noticed some difficulty using it with (for example with an Envelope object when str method is called). I'm a bit busy for now but I'll explain my problem with few examples and unit test describing it better (some sample might already be visible on my github fork).

[mjs]

Original comment by Menno Smits (Bitbucket: mjs0):

---

Hi, have you had any more time to look at this?

[mjs]

Original comment by Vic Tor (Bitbucket: maxiimou):

---

I was waiting for your feedbacks before going further. I'll look to your comments and will come back to you soon !

[mjs]

Original comment by Menno Smits (Bitbucket: mjs0):

---

Hi again. Sorry it's taken me so longer to come back to this.

Thanks so much for all your work on this. I like the overall idea and I think the implementation is heading in the right direction. I've made some comments against some of your commits on Github. If you can address those then I think this is pretty close to being OK for merging.

No problems with using git/Github instead of Mercurial. I can take care of merging it from there back into the Bitbucket repo.

[mjs]

Original comment by Menno Smits (Bitbucket: mjs0):

---

Hi. Thanks for this. I'm in the middle of busy work trip at the moment but will hopefully get back to you properly later in the week.

[ataylor32]

I'm interested in this issue as well. I have an email from Example, Véronique <[email protected]>, so envelope.from_[0].name looks like this:

b'=?UTF-8?Q?Example,_V=C3=A9ronique?='

Thus, I need to run it through decode_header (from email.header import decode_header)