Walk Multipart Email: initial_value must be str or None, not bytes

[bobmeliev]

Walk Multipart Email is failing with error:

TypeError: initial_value must be str or None, not bytes

Wait For Email is returning index as bytes but converting the index into string doesn't help.

    ${latest}=    Wait For Email    subject=Welcome!    timeout=60
    ${type}=    Evaluate    type($latest)    
    ${latest_str}=    Convert To String    ${latest}
    ${type}=    Evaluate    type($latest_str)    
    ${parts}=    Walk Multipart Email    ${latest_str}

20191121 10:17:59.267 : INFO : ${latest} = 12
20191121 10:17:59.284 : INFO : ${type} = <class 'bytes'>
20191121 10:17:59.288 : INFO : ${latest_str} = 12
20191121 10:17:59.292 : INFO : ${type} = <class 'str'>
20191121 10:17:59.533 : FAIL : TypeError: initial_value must be str or None, not bytes

[Tanmay-Agrawal]

@bobmeliev , @lasselindqvist have you been able to find a solution to the problem?

[lasselindqvist]

I haven't been looking into this, but will happily merge if there is a solution to a problem.

[ulkanti]

In init.py from email import message_from_bytes instead of from email import message_from_string and in def walk_multipart_email(self, email_index): the same change should fix the issue.

That is if you are using python 3. If you are using python2, the current setup would work. So the library now only for one python version only.

[lasselindqvist]

import message_from_string and in def walk_multipart_email(self, email_index)

If this is the case, I would prefer to fix it even if the quick fix meant Python 2 no longer works, since it is EOL anyway.

[lynhbn]

@lasselindqvist This one can easily be fixed by changing:
data = self._imap.uid('fetch', email_index, '(RFC822)')[1][0][1]
into:
data = self._imap.uid('fetch', email_index, '(RFC822)')[1][0][1].decode('utf-8')
in init.py line 348
Sorry I'm unable to help you to create a pull request

[lynhbn]

One another point, the fuction open_link_from_email also does not work with Python 3.
Please help to fix it at line 255:
content_type = resp.headers.getheader('content-type')
change into:
content_type = resp.headers.get('content-type')
Then everything will be okay
Looking forward to your answer soon!

[lasselindqvist]

One another point, the fuction open_link_from_email also does not work with Python 3.
Please help to fix it at line 255:
content_type = resp.headers.getheader('content-type')
change into:
content_type = resp.headers.get('content-type')
Then everything will be okay
Looking forward to your answer soon!

I'm happy to merge and release, if you can provide a PR for it.

[azad-DH]

Since I was trying to fix the same issue, I created a PR now.
I'd be very grateful if you could review and approve it, so we can finally use ImapLibary with Python 3. 🙏

I also added lynhbn's suggestion in a second commit.

[lasselindqvist]

Can you make the PR in this repository? You made it in rickpc original.

[azad-DH]

~~Oh ok, yeah sure I'll make the PR here as well.~~~

Actually, when I try to create a PR in this repository, the PR is just created in rickypc again automatically.

@lasselindqvist Can you merge the PR over there?

[lasselindqvist]

I have no rights to that repository, and the owner of it seems to be inactive, so I cannot do anything to those PRs.

[azad-DH]

Ah that's what I feared. I'll create a PR in your repo now.
Nope, I got no access rights to create a PR in your repo. I can only do a fork, or can you give me access rights for your repo?

[azad-DH]

Ok I managed to create a PR in your repo:
#13

I don't know why I can't assign a reviewer, but I'm assuming you should be able to merge it from there. :)