Get Email Body decodes URLs with issues

[bobmeliev]

URLs are decoded with issues, for instance URL

<a href="https://example.com/verify-account?uid=5dd50bc5d3b1f228474551d7&t=98e1ee0c-a159-475e-a0e3-f74a533d48c8">

is decoded as

<a href="https://example.com/verify-account?uid]d50bc5d3b1f228474551d7&t�e1ee0c-a159-475e-a0e3-f74a533d48c8">

Here is Python code of function get_email_body():

        if self._is_walking_multipart(email_index):
            body = self.get_multipart_payload(decode=True)
        else:
            encoded_body = self._imap.uid('fetch', email_index, '(BODY[TEXT])')[1][0][1]
            try:
                body = decode(encoded_body, 'quopri_codec').decode('UTF-8')
            except:
                body = decode(encoded_body, 'quopri_codec').decode('ISO-8859-1')
        return body

[frirasyidi]

I stumbled on the same issue here.

I noticed that this issue has been open for close to 2 years. Have you figured a workaround on this problem @bobmeliev?

[lasselindqvist]

So it seems like the issues happen when there is the "equals sign" in the text.
It is because there is a usage of https://en.wikipedia.org/wiki/Quoted-printable decoding. It breaks the text.

[frirasyidi]

@lasselindqvist Sorry, I'm not as familiar with this, but is there a way to fix it?

I do try to look at the code and it says quopri_codec, which I assume is the one that you've mentioned. But I was not sure which part to update. I've tried to look at the encoding lists here: https://docs.python.org/3/library/email.message.html which does contain quopri_codec, but could not find the necessary q-encoding.

[lasselindqvist]

Sadly I am also not too familiar with it. It is fully possible that the quopri_codec decoding part is unnecessary, but then again it must be there for a reason. I think I need to setup a real test situation for this and see with different kind of messages.