Re: Is there any 'xoauth2' login wrapper support for v5 of current getmail for ms o365 in the works?
Celejar <[email protected]> Sun, 9 Jan 2022 21:22:21 -0500
| Newsgroups | gmane.mail.getmail.user |
|---|---|
| Message-ID | <[email protected]> |
On Sat, 8 Jan 2022 23:31:07 -0800 dmacdoug <[email protected]> wrote: > On Fri, Nov 12, 2021 at 12:06:45PM -0600, Charles Cazabon wrote: > > Urs Rau <[email protected]> wrote: > ... > > > > > > But for that we would need to find a xoauth2 ‘plugin’ that supports ms o365 > > > with modern authentication as well as running on an old and end of life > > > version of python. > > > > I don't actually use xoauth2, but getmail has some support for it. The > > getmail-gmail-xoauth-tokens helper script it uses should still work? I don't > > know for sure as I don't use Gmail. > > > > If you find xoauth2 support is broken, please report it. No one else has > > mentioned it IIRC. > > > ... > > > > > Has anybody worked out a way to expand the xoauth2 functionality from Stefan > > > Krah to work with ms o365 and not just gmail? > > > > No one has contributed support for it, I think. The docs ask people who can > > make it work with other providers provide that info to me for inclusion. > > > > Charles > > I'm wondering if the OP or anyone else has made any further headway on > using XOAUTH2 with Getmail to download email from MS Office365 IMAP server > since this inquiry was posted or, alternatively, found it to not be possible? > > I tried it a couple of years ago without success, but now my university is > going to require it starting Jan. 20th, so if I can't make it work I'll > have to find some alternative way to get my email. This project looks quite promising, if it actually works as described: https://github.com/simonrob/email-oauth2-proxy I don't currently use OAuth for email, so I haven't tried it. > Additionally they are going to require two factor authentication using a > company called Duo. I'm not sure how this could be made to work with > getmail, so perhaps all is futile. Any observations would be appreciated. Presumably something similar to the above could be written for any (open) 2FA system, but I don't know if one actually exists. Celejar