fetchmail-users — User questions and discussion

Re: [Fetchmail-users] Push Notification

[Andrzej M. <and...@gm...>]

Yes
try this https://otremba.net/wiki/Fetchmail_(Debian)

On Tue, Jul 25, 2023 at 1:51 PM <ns...@gm...> wrote:

> Hello,
>
> is fetchmail able to use the mechanism of push notifications to be anle to
> receive new E-Mails immeadiately from a server?
>
> Thanks,
>
> Thomas
>
>
> _______________________________________________
> Fetchmail-users mailing list
> Fet...@li...
> https://lists.sourceforge.net/lists/listinfo/fetchmail-users
>


-- 
Andrzej Milewski
and...@gm...
tel. 0603957324

[Fetchmail-users] Push Notification

[<ns...@gm...>]

Hello,

is fetchmail able to use the mechanism of push notifications to be anle to receive new E-Mails immeadiately from a server?

Thanks,

Thomas

Re: [Fetchmail-users] force plain / login authentication

[Ede W. <li...@ne...>]

auth password does indeed work, if one gets enough sleep and therefore 
stays clear from typos. Fixed.

Am 06.07.23 um 08:18 schrieb Ede Wolf:
> Hello,
> 
> I am dealing with a buggy pop3 server, that announces cram-md5 
> authentication, but in fact does not do so (any more). Confirmed by the 
> support.
> 
> As I will not be able to fix the remote server, I am wondering, are 
> there a means to tell fetchmail to only use either plain or login? Both 
> are advertised as well.
> 
> Adding "authenticate password" to fetchmailrc does not do the trick.
> 
> The connection itself is TLS secured
> 
> Thanks
> 
> Ede
> 
> 
> _______________________________________________
> Fetchmail-users mailing list
> Fet...@li...
> https://lists.sourceforge.net/lists/listinfo/fetchmail-users

[Fetchmail-users] force plain / login authentication

[Ede W. <li...@ne...>]

Hello,

I am dealing with a buggy pop3 server, that announces cram-md5 
authentication, but in fact does not do so (any more). Confirmed by the 
support.

As I will not be able to fix the remote server, I am wondering, are 
there a means to tell fetchmail to only use either plain or login? Both 
are advertised as well.

Adding "authenticate password" to fetchmailrc does not do the trick.

The connection itself is TLS secured

Thanks

Ede

Re: [Fetchmail-users] XCLIENT and CLIENTID extensions to SMTP

[Matthias A. <mat...@gm...>]

Am 23.03.23 um 11:55 schrieb Andrew C Aitchison:
> I have become aware of two extensions to SMTP (and one to IMAP too)
> that may be of interest to fetchmail developers and users.
>
> The first is https://www.postfix.org/XCLIENT_README.html
> which says that
>    The XCLIENT command targets the following problems:
>    ...        ...        ...
>    2. Client software that downloads mail from an up-stream mail server and
>    injects it into a local MTA via SMTP. In order to take advantage of the
>    local MTA's SMTP server access rules, the client software needs the
>    ability to override the SMTP server's idea of the remote client name,
>    client address and other information. Such information can typically be
>    extracted from the up-stream mail server's Received: message header.
>
> XCLIENT is implemented in postfix and is now a wishlist item for Exim:
> https://bugs.exim.org/show_bug.cgi?id=2702 as well as being available
> in a fork of Exim (
> https://github.com/SpamExperts/exim/commit/3798d48d73c89f7835726d31f096851f7f7fca2a ).
>
> The second is CLIENTID, a pair of draft RFCs
> https://datatracker.ietf.org/doc/draft-storey-smtp-client-id/
> https://datatracker.ietf.org/doc/draft-yu-imap-client-id/
> which allows the mail client to tell the server more about itself
> and the user, allowing a server to restrict access on a device+user basis;
> eg to block a spammer who has got hold of your login details, whilst
> continuing to allowing you to send and receive emails from your phone.
>
> CLIENTID is supported by Thunderbird, emClient, and BlueMail clients
> and at least MagicMail "the Top Selling Carrier Grade Email platform for
> the ISP, Telco, and Cable industry in North America".
> I am currently attempting to write CLIENTID support for Exim.

Andrew,

Thank you. I had thought about this in the past, but not added it, and I
find your context presentation valuable. I have thus filed this under
https://gitlab.com/fetchmail/fetchmail/-/issues/58

...so it's with the other feature requests.

Thanks for your suggestion!

Cheers,
Matthias

Re: [Fetchmail-users] email plus addressing?

[Matthias A. <mat...@gm...>]

Am 08.04.23 um 11:28 schrieb Andrew C Aitchison:
>
>> Am 08.04.23 um 04:19 schrieb Howard Spindel:
>>> Experimenting today with plus addressing on email.
>>>
>>> What I found is that fetchmail issues a warning that the recipient
>>> address+test doesn't match any local name (because it doesn't!).
>>> Everything still works because fetchmail passes the email off to
>>> sendmail which understands to strip off the plus addressing.
>>>
>>> Should fetchmail support plus addressing without issuing a warning?
>
> One page explaining what "plus addressing" is:
> https://www.codetwo.com/admins-blog/plus-addressing/

Thanks. I know what it is, I have been using it for a quarter century
for direct ingress mail organization with qmail (briefly, as minus
addressing) and then Postfix. The lists I host myself use it in VERP.

> Not sure I agree wit you Matthias.

You do not need to.

> Assuming you do know what Howard means by "plus addressing",
> his request is clear enough. Howard did not mean to prescribe
> *how* the issue is addressed - others on the list may have ideas
> - but it is reasonable to suggest that it works without producing
> warnings.

I am saying I need to see the specific warnings and the configuration,
and please not made up beyond recognition. This depends on too many
options to look in the right place in the code. No passwords please, and
feel free to replace domains by example.org.

Everything else is a waste of my time,
and everyone wants me to spend time on implementation, not on hunting a
cursorily described cosmetic bug and asking questions and then still
fixing something else that doesn't resolve the issue.

Re: [Fetchmail-users] email plus addressing?

[Matthias A. <mat...@gm...>]

Am 08.04.23 um 04:19 schrieb Howard Spindel:
> Experimenting today with plus addressing on email.
>
> What I found is that fetchmail issues a warning that the recipient
> address+test doesn't match any local name (because it doesn't!).
> Everything still works because fetchmail passes the email off to
> sendmail which understands to strip off the plus addressing.
>
> Should fetchmail support plus addressing without issuing a warning?


Howard,

the lady's make-up is a bit strange. Should it be different?


What lady? What in particular? Lid shadow? Rouge? Mascara?


What I am saying with this figure is: be specific, and show me.
→ https://www.fetchmail.info/fetchmail-FAQ.html#G3

→ https://www.chiark.greenend.org.uk/~sgtatham/bugs.html

→ http://www.catb.org/~esr/faqs/smart-questions.html

Regards,
Matthias

[Fetchmail-users] email plus addressing?

[Howard S. <ho...@sc...>]

Experimenting today with plus addressing on email.

What I found is that fetchmail issues a warning that the recipient 
address+test doesn't match any local name (because it doesn't!). 
Everything still works because fetchmail passes the email off to 
sendmail which understands to strip off the plus addressing.

Should fetchmail support plus addressing without issuing a warning?

[Fetchmail-users] XCLIENT and CLIENTID extensions to SMTP

[Andrew C A. <fet...@ai...>]

I have become aware of two extensions to SMTP (and one to IMAP too)
that may be of interest to fetchmail developers and users.

The first is https://www.postfix.org/XCLIENT_README.html
which says that
   The XCLIENT command targets the following problems:
   ...		...		...
   2. Client software that downloads mail from an up-stream mail server and
   injects it into a local MTA via SMTP. In order to take advantage of the
   local MTA's SMTP server access rules, the client software needs the
   ability to override the SMTP server's idea of the remote client name,
   client address and other information. Such information can typically be
   extracted from the up-stream mail server's Received: message header.

XCLIENT is implemented in postfix and is now a wishlist item for Exim:
   https://bugs.exim.org/show_bug.cgi?id=2702 as well as being available
in a fork of Exim ( https://github.com/SpamExperts/exim/commit/3798d48d73c89f7835726d31f096851f7f7fca2a ).

The second is CLIENTID, a pair of draft RFCs
   https://datatracker.ietf.org/doc/draft-storey-smtp-client-id/
   https://datatracker.ietf.org/doc/draft-yu-imap-client-id/
which allows the mail client to tell the server more about itself
and the user, allowing a server to restrict access on a device+user basis;
eg to block a spammer who has got hold of your login details, whilst
continuing to allowing you to send and receive emails from your phone.

CLIENTID is supported by Thunderbird, emClient, and BlueMail clients
and at least MagicMail "the Top Selling Carrier Grade Email platform 
for the ISP, Telco, and Cable industry in North America".
I am currently attempting to write CLIENTID support for Exim.

-- 
Andrew C. Aitchison                      Kendal, UK
                    an...@ai...

[Fetchmail-users] ANNOUNCE: fetchmail 6.4.37 is available (translation updates)

[Matthias A. <mat...@gm...>]

The 6.4.37 release of fetchmail is now available at the usual locations,
including <https://sourceforge.net/projects/fetchmail/files/branch_6.4/>.

The source archive is available at:
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.37.tar.xz/download>
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.37.tar.lz/download>

The detached GnuPG signature is available at:
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.37.tar.xz.asc/download>
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.37.tar.lz.asc/download>

The SHA256 hashes for the tarballs are:
SHA2-256(fetchmail-6.4.37.tar.xz)= 4a182e5d893e9abe6ac37ae71e542651fce6d606234fc735c2aaae18657e69ea
SHA2-256(fetchmail-6.4.37.tar.lz)= 2bc3c602bb967a8d7a7348a6e7bdffc54ab93c61187301431c63914372ec3c1b


Here are the release notes:
--------------------------------------------------------------------------------
fetchmail-6.4.37 (released 2023-02-26, 31710 LoC):

# TRANSLATIONS: language translations were updated by this fine person:
* sr:    Мирослав Николић (Miroslav Nikolić) [Serbian]

--------------------------------------------------------------------------------

[Fetchmail-users] ANNOUNCE: fetchmail 6.4.36 is available (translation updates)

[Matthias A. <mat...@gm...>]

The 6.4.36 release of fetchmail is now available at the usual locations,
including <https://sourceforge.net/projects/fetchmail/files/branch_6.4/>.

The source archive is available at:
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.36.tar.xz/download>
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.36.tar.lz/download>

The detached GnuPG signature is available at:
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.36.tar.xz.asc/download>
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.36.tar.lz.asc/download>

The SHA256 hashes for the tarballs are:
SHA2-256(fetchmail-6.4.36.tar.xz)= 700d433838d3e29e304452aec56b21874f538ec24113fdcbb25139c5f2edc23a
SHA2-256(fetchmail-6.4.36.tar.lz)= 5554fd033b461f255f83301985fe12b5bbf964d8fdcead87a339b7e0c726cd65


Here are the release notes:
--------------------------------------------------------------------------------
fetchmail-6.4.36 (released 2023-01-28, 31710 LoC):

# TRANSLATIONS: language translations were updated by these fine people:
(in alphabetical order of language codes):
* cs:    Petr Pisar [Czech]
* es:    Cristian Othón Martínez Vera [Spanish]
* fr:    Frédéric Marchal [French]
* ja:    Takeshi Hamasaki [Japanese]
* pl:    Jakub Bogusz [Polish]
* ro:    Remus-Gabriel Chelu [Romanian]
* sq:    Besnik Bleta [Albanian]
* sv:    Göran Uddeborg [Swedish]

--------------------------------------------------------------------------------

Re: [Fetchmail-users] ANNOUNCE: The 6.5.0.beta9 snapshot of fetchmail is available

[Christian E. <bc...@ph...>]

* Matthias Andree on Friday, January 13, 2023 at 23:49:02 +0100:
> Am 12.01.23 um 17:07 schrieb Christian Ebert:
>> -extern int getopt (void);
>> +extern int getopt ();
> 
> [...]
> 
> So, since fetchmail 6.5 will expect some degree of standards
> conformance, and unistd.h has been required to provide a getopt()
> declaration since before the dawn of ages, I will just strip this and
> the #ifdef stuff. This should be fixed in git 6b8fb508. Can you please
> try if that version works for you, or else, download from Gitlab, and
> see if you can still compile it on MacOS? Link:
> https://gitlab.com/fetchmail/fetchmail/-/raw/legacy_6x/getopt.h?inline=false

6b8fb508 builds fine on MacOS 12.6.2, fwiw also as universal. 
Also seems to do its job nicely.

Thanks for the quick fix.

cheers,

Christian
-- 
LAST SHIP HOME                   --->> https://lastshiphome.de/en
Official Selection                           DOK.fest Munich 2018
German Ocean Film Award                        CineMare Kiel 2019
Best Documentary Feature   Wales International Film Festival 2020

Re: [Fetchmail-users] ANNOUNCE: The 6.5.0.beta9 snapshot of fetchmail is available

[Matthias A. <mat...@gm...>]

Am 12.01.23 um 17:07 schrieb Christian Ebert:
> Hi Matthias,
>
> * Matthias Andree on Friday, January 06, 2023 at 22:09:32 +0100:
>> The 6.5.0.beta9 release of fetchmail is now available at the usual
>> locations,
>
> On MacOS 12.6.2 I had to revert getopt.h to beta8 state:
>
> --- getopt.h.orig       2023-01-12 15:53:05.000000000 +0000
> +++ getopt.h    2023-01-12 15:53:31.000000000 +0000
> @@ -101,7 +101,7 @@
>     errors, only prototype getopt for the GNU C library.  */
>  extern int getopt (int argc, char *const *argv, const char *shortopts);
>  #else /* not __GNU_LIBRARY__ */
> -extern int getopt (void);
> +extern int getopt ();

Hi Christian,

thanks for the report. My thinking is "Uh, right. Sorry. I know better."

I was fixing compiler warnings quickly, and I was so caught up in C++
implementation thinking where () indeed means (void) that I didn't
notice this was C, and getopt() without arguments does not make sense
either without looking at the #ifdef context and cleaning it up properly.

So, since fetchmail 6.5 will expect some degree of standards
conformance, and unistd.h has been required to provide a getopt()
declaration since before the dawn of ages, I will just strip this and
the #ifdef stuff. This should be fixed in git 6b8fb508. Can you please
try if that version works for you, or else, download from Gitlab, and
see if you can still compile it on MacOS? Link:
https://gitlab.com/fetchmail/fetchmail/-/raw/legacy_6x/getopt.h?inline=false

Thanks in advance.

Cheers,
Matthias

Re: [Fetchmail-users] ANNOUNCE: The 6.5.0.beta9 snapshot of fetchmail is available

[Christian E. <bc...@ph...>]

Hi Matthias,

* Matthias Andree on Friday, January 06, 2023 at 22:09:32 +0100:
> The 6.5.0.beta9 release of fetchmail is now available at the usual locations,

On MacOS 12.6.2 I had to revert getopt.h to beta8 state:

--- getopt.h.orig       2023-01-12 15:53:05.000000000 +0000
+++ getopt.h    2023-01-12 15:53:31.000000000 +0000
@@ -101,7 +101,7 @@
     errors, only prototype getopt for the GNU C library.  */
  extern int getopt (int argc, char *const *argv, const char *shortopts);
  #else /* not __GNU_LIBRARY__ */
-extern int getopt (void);
+extern int getopt ();
  #endif /* __GNU_LIBRARY__ */
  extern int getopt_long (int argc, char *const *argv, const char *shortopts,
                         const struct option *longopts, int *longind);

To avoid the following error:

In file included from fetchmail.c:37:
./getopt.h:104:12: error: conflicting types for 'getopt'
extern int getopt (void);
            ^
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/unistd.h:509:6: note: previous declaration is here
int      getopt(int, char * const [], const char *) __DARWIN_ALIAS(getopt);
          ^
1 error generated.
make[2]: *** [fetchmail.o] Error 1
make[1]: *** [all-recursive] Error 1
make: *** [all] Error 2

HTH

Christian

-- 
LAST SHIP HOME                   --->> https://lastshiphome.de/en
Official Selection                           DOK.fest Munich 2018
German Ocean Film Award                        CineMare Kiel 2019
Best Documentary Feature   Wales International Film Festival 2020

[Fetchmail-users] ANNOUNCE: The 6.5.0.beta9 snapshot of fetchmail is available

[Matthias A. <mat...@gm...>]

The 6.5.0.beta9 release of fetchmail is now available at the usual locations,
including <https://downloads.sourceforge.net/project/fetchmail/branch_6.5/>.

The source archive is available at:
<https://downloads.sourceforge.net/project/fetchmail/branch_6.5/fetchmail-6.5.0.beta9.tar.xz/download>

The detached GnuPG signature is available at:
<https://downloads.sourceforge.net/project/fetchmail/branch_6.5/fetchmail-6.5.0.beta9.tar.xz.asc/download>

The SHA256 hashes for the tarballs are:
SHA2-256(fetchmail-6.5.0.beta9.tar.xz)= ee6212d2aad80507f9128a0969ae1a2a8dbd3407c8bf8e5f9ec4fc3a6754b093


Here are the additions to the release notes since beta8:
+---------------------------------------------------------------------------
KNOWN ISSUES:
* Using ccache may trigger "implicit fallthrough" warnings because
  the comments that, for instance, GCC understands, are removed by ccache's
  separate preprocessing.  Fixing this portably requires C++17.
 
INCOMPATIBLE CHANGES:
* The ca, da, en_GB, id, it, nl, ru, zh_CN translations have been disabled,
  they are too far behind.
 
 ## CHANGED REQUIREMENTS
* fetchmail 6.5.0 requires a TLSv1.3-capable version of OpenSSL or wolfSSL,
  at a minimum OpenSSL v1.1.1 or wolfSSL v5.5.1.
 
 ## BUG FIXES
* Received: lines now return GMT time if the tzoffset cannot be represented
  as whole minutes. Reported by @rriddicc via Gitlab #49.
 
 ## CHANGES
* There is now a --moveto feature (only feasible in IMAP) that, instead of
  flushing mail, moves it to a user-specified folder. This is to assist with
  archiving, or when providers (G...) break the IMAP model.
  Courteously provided by Damjan Jovanovic.
* fetchmail can be built with meson 0.60 or newer <https://mesonbuild.com/>.
  fetchmail is not currently written in a way that supports unity
  (amalgamated) builds.
============================================================================

[Fetchmail-users] ANNOUNCE: fetchmail 6.4.35 is available (translation updates and raised SSL/TLS library requirements)

[Matthias A. <mat...@gm...>]

The 6.4.35 release of fetchmail is now available at the usual locations,
including <https://sourceforge.net/projects/fetchmail/files/branch_6.4/>.

The source archive is available at:
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.35.tar.xz/download>
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.35.tar.lz/download>

The detached GnuPG signature is available at:
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.35.tar.xz.asc/download>
<https://sourceforge.net/projects/fetchmail/files/branch_6.4/fetchmail-6.4.35.tar.lz.asc/download>

The SHA256 hashes for the tarballs are:
SHA2-256(fetchmail-6.4.35.tar.xz)= 7b0b56cbc0fca854504f167795fab532d5a54d5a7d3b6e3e36a33f34a0960a01
SHA2-256(fetchmail-6.4.35.tar.lz)= 8c8675b4655344d6d14143c4a16fa0ebcf5754c9e707d413dab9fec68d1826cd

Here are the release notes:
>-------------------------------------------------------------------------------
fetchmail-6.4.35 (released 2023-01-04, 31707 LoC):

# BREAKING CHANGES:
* Fetchmail now warns about OpenSSL before 1.1.1s or 3.0.7,
  and rejects wolfSSL older than 5.5.0.

# TRANSLATIONS: language translations were updated by these fine people:
(in reverse alphabetical order of language codes so as not to prefer people):
* sv:    Göran Uddeborg [Swedish]
* eo:    Keith Bowes [Esperanto]
>-------------------------------------------------------------------------------

Re: [Fetchmail-users] changing ~/.netrc does not restart fetchmail

[Matthias A. <mat...@gm...>]

Am 05.11.22 um 19:06 schrieb Francesco Potortì:
> I know that modifying (or touching) ~/.fetchmailrc forces fetchmail (daemon mode) to restart and reload it.  This is useful and intuitive, so I had assumed that the same happens when ~/.netrc is changed.
>
> However, I have a ~/.netrc whic is a symbolic link to ~/.authinfo, and apparently modifying ~/.authinfo does not force fetchmail to reread it.  I have to kill the fetchmail daemon and restart it to read the new data.
>
> Is this normal?  Intended?  Am I missing anything?  Should I file a feature request?
>
Hello Francesco,

sorry that the .netrc file is only read on startup, which kinda makes it
less useful in daemon mode.
It would make sense to re-read it whenever it has changed.

Feel free to file a feature request on Gitlab ->
https://gitlab.com/fetchmail/fetchmail/-/issues

Cheers,
Matthias

[Fetchmail-users] changing ~/.netrc does not restart fetchmail

[Francesco P. <Po...@is...>]

I know that modifying (or touching) ~/.fetchmailrc forces fetchmail (daemon mode) to restart and reload it.  This is useful and intuitive, so I had assumed that the same happens when ~/.netrc is changed.

However, I have a ~/.netrc whic is a symbolic link to ~/.authinfo, and apparently modifying ~/.authinfo does not force fetchmail to reread it.  I have to kill the fetchmail daemon and restart it to read the new data.

Is this normal?  Intended?  Am I missing anything?  Should I file a feature request?

-- 
Francesco Potortì (ricercatore)        Mobile: +39.348.8283.107	     
ISTI - Area della ricerca CNR          Skype:  wnlabisti	     
via G. Moruzzi 1, I-56124 Pisa         Web:    http://fly.isti.cnr.it
(gate 20, 1st floor, room C71)         ISPIN:  https://ieee-jispin.org/

Re: [Fetchmail-users] Working with Office 365; fetchmail_oauth2.py

[Matthias A. <mat...@gm...>]

Am 14.10.22 um 11:01 schrieb rbell--- via Fetchmail-users:
> 	A mail server I use uses Office 365, recently switched to its
> OAuth2.  I've gotten OAuth2 to work with gmail, which is different.
> 	I downloaded github.com/lubonbvba/fetchmail_oauth2 but see no
> instructions to make it work - does it?
> 	Another person mentions using fetchmail 7 - what does
> fetchmail say about that?

fetchmail 7 is an alpha version, without integrated OAuth2, and only
with a few hooks. Being alpha, I take the liberty to change interfaces
and requirements liberally and incompatibly all over the map, no
promises, no guarantees, no support.­  Not that that's different from
other free and open source software unless you have a maintenance
contract...

OAuth2 is "hooks and external scripts" at best because the big OAuth2
proponents (Google and Microsoft) have not only used it as
authentication scheme but where you can still just give your credentials
to your mail client, but **also** require applications to be registered,
possibly some "tenant" stuff with rulers (or their delegates) of your
organization needing to permit the application and/or application and a
particular user's access and all other sorts of political obstacles to
get OAuth2 use, which are usually a bigger issue than the technical
issues -- that is why OAuth2 is not fully integrated and will not
currently be. OAuth2 *in practice* is abused to control what
applications can access YOUR OWN mail.

Do you need to register your slippers or mailbox key with $YOURCOUNTRY's
Postal Service before you can open your mailbox and get your mail out?
No. But that is what Google and Microsoft do with OAuth2 in practice.

With mail providers requiring applications to be registered, possibly
fee-paying assessment - no thanks.

I have asked several times for someone to show me "open door" OAuth2
access use cases (providers), where it really is just some
authentication scheme similar to Kerberos, but nobody showed me one yet.
So apparently OAuth2 is only relevant for the Big Data business.

And that's why currently OAuth2 for a single free software author is
about as unattractive as a tooth ache.

Having said that, if you are willing to put up with a doubly unsupported
setup (a. fetchmail 7 being alpha, and b. OAUTH2 being experimental, and
likely staying so eternally unless the political landscape changes), get
fetchmail 7 and README.OAUTH2, or - as you have set out - venture into
proxies or authentication token obtaining libraries (I find Simon
Robinson's documentation on email-oauth2-proxy has many external links,
so glad multiple questions made me interested in that).

That is cutting edge territory, nowhere near production quality.
"Cutting" edge as in bleeding. Only recommended for toying around,
development, experimentation, learning and sharing.

Many of the OAuth2 modules are also written in Python, which may make
you personally uncomfortable, as you have disclosed in the other message
on Simon Robinson's OAuth2 proxy.

I suppose I will integrate OAuth2 no sooner than someone shows me a big
provider with more than three fetchmail users which will just use OAuth2
as authentication scheme, without requiring client registration and
other nonsense.

Regards,
Matthias

Re: [Fetchmail-users] making email-oauth2-proxy work

[Matthias A. <mat...@gm...>]

Am 19.10.22 um 12:35 schrieb rbell--- via Fetchmail-users:
> 	When I run it I get:
>
>    File "/home/russell/bin/emailproxy.py", line 146
> SyntaxError: Non-ASCII character '\xe2' in file /home/russell/bin/emailproxy.py on line 147, but no encoding declared; seehttp://python.org/dev/peps/pep-0263/  for details
>
> 	I add:
>
> 	# -*- coding: utf-8 -*-
>
> as the second line, which cures this complaint.
>
> 	Then I get:
>
>    File "/home/russell/bin/emailproxy.py", line 16, in <module>
>      import configparser
> ImportError: No module named configparser
>
>
> but there IS a module named configparser.  I searched on the Internet,
> didn't find a cure.  I suspect it's a simple question - when it comes
> to Python, I'm a simpleton.
>
Russell,

there may be confusion around Python versions 2 and 3. A configparser
module for Python 2 will not satisfy a Python 3 email-oauth2-proxy.

Please ensure you are using a supported Python 3 version (personally I'd
say at least 3.8, 3.10 is current, and 3.11 had release candidate status
when I checked a few days ago) -- and follow its authors instructions.

Also make sure that your python executable is really Python 3, else
install the latter and try typing "python3" instead of "python".

Your operating system may instead have a way for you to make python v3.x
a default version when running "python", but you give too little
information on your OS and Python versions or other details of the
setup, and my crystal ball is broken and not yet back from repair...

https://github.com/simonrob/email-oauth2-proxy#getting-started

https://github.com/simonrob/email-oauth2-proxy#dependencies-and-setup

Chances are you did NOT install the required packages the same way as
explained there:

> [...] from a terminal, install the script's requirements: |python -m
> pip install -r requirements.txt|, and start the proxy: |python
> emailproxy.py| – a menu bar/taskbar icon should appear. If instead of
> the icon you see an error in the terminal, it is likely that your
> system is missing dependencies for the |pywebview| or |pystray|
> packages. See the dependencies and setup
> <https://github.com/simonrob/email-oauth2-proxy#dependencies-and-setup>
> section [...]
HTH
Matthias

[Fetchmail-users] making email-oauth2-proxy work

[<rb...@al...>]

	When I run it I get:

  File "/home/russell/bin/emailproxy.py", line 146
SyntaxError: Non-ASCII character '\xe2' in file /home/russell/bin/emailproxy.py on line 147, but no encoding declared; see http://python.org/dev/peps/pep-0263/ for details

	I add:

	# -*- coding: utf-8 -*-

as the second line, which cures this complaint.

	Then I get:

  File "/home/russell/bin/emailproxy.py", line 16, in <module>
    import configparser
ImportError: No module named configparser


but there IS a module named configparser.  I searched on the Internet,
didn't find a cure.  I suspect it's a simple question - when it comes
to Python, I'm a simpleton.


russell bell

Re: [Fetchmail-users] Microsoft deprecated

[Mihai L. <mt...@gm...>]

On Tuesday, October 18, 2022 at 22:03:27 -0400, Jerry Geis wrote:

> Would you mind sharing your configs - thanks

Mutt configuration for Microsoft SMTP:

     set ssl_force_tls = no
     set smtp_url = smtp://<microsoft_email>@localhost:1587
     set smtp_pass = <local_pwd>

Mutt configuration for Gmail SMTP:

     set ssl_force_tls = no
     set smtp_url = smtp://<gmail_email>@localhost:2587
     set smtp_pass = <local_pwd>

Relevant parts of fetchmail configuration for Microsoft IMAP:

     poll localhost
     protocol IMAP
     port 1993
     user '<microsoft_email>' there is '<local_user>' here
     password <local_pwd>
     sslproto ''

Relevant parts of fetchmail configuration for Gmail IMAP:

     poll localhost
     protocol IMAP
     port 2993
     user '<gmail_email>' there is '<local_user>' here
     password <local_pwd>
     sslproto ''


Relevant parts of https://github.com/simonrob/email-oauth2-proxy 
configuration (probably I'm not using all):

     [IMAP-1993]
     local_address = localhost
     server_address = outlook.office365.com
     server_port = 993

     [POP-1995]
     server_address = outlook.office365.com
     server_port = 995

     [SMTP-1587]
     server_address = smtp.office365.com
     server_port = 587
     starttls = True

     [IMAP-2993]
     server_address = imap.gmail.com
     server_port = 993

     [POP-2995]
     server_address = pop.gmail.com
     server_port = 995

     [SMTP-2465]
     server_address = smtp.gmail.com
     server_port = 465

     [SMTP-2587]
     server_address = smtp.gmail.com
     server_port = 587
     starttls = True

I do not remember which values below I have set myself. 
email-oauth2-proxy also changes the same file 
in-place.

  <...> means I concealed the value. I think I got 
client_id from the Thunderbird sources (you can also use the ID of a 
project that you create and authorize yourself with Google and 
Microsoft, or already registered projects like Evolution, …).

     [<microsoft_email]
     permission_url = https://login.microsoftonline.com/common/oauth2/v2.0/authorize
     token_url = https://login.microsoftonline.com/common/oauth2/v2.0/token
     oauth2_scope = https://outlook.office365.com/IMAP.AccessAsUser.All https://outlook.office365.com/POP.AccessAsUser.All https://outlook.office365.com/SMTP.Send offline_access
     redirect_uri = http://localhost:8080
     client_id = <...>
     client_secret = <...>

     [<gmail_email>]
     permission_url = https://accounts.google.com/o/oauth2/auth
     token_url = https://www.googleapis.com/oauth2/v3/token
     oauth2_scope = https://mail.google.com/
     redirect_uri = http://localhost:8080
     client_id = <...>
     client_secret = <...>

HTH,
Mihai

Re: [Fetchmail-users] Microsoft deprecated

[Jerry G. <jer...@gm...>]

On Tue, Oct 18, 2022 at 7:18 PM Mihai Lazarescu <mt...@gm...> wrote:

> On Monday, October 17, 2022 at 15:23:54 -0400, Jerry Geis wrote:
>
> > I came across this https://github.com/simonrob/email-oauth2-proxy
> >
> > But I have not got it working yet.
>
> Works for me flawless since mid-August or so on Fedora for both
> Google and Microsoft accounts, for downloading (IMAP, fetchmail)
> and sending (SMTP, mutt).
>
> The only change in the fetchmail and mutt configurations was to
> direct the connection to localhost:<port>, with a local password
> (not the one on servers), without encryption. Of course, after
> configuring email-oauth2-proxy. :-)
>
> HTH,
> Mihai
>
>
> _______________________________________________
> Fetchmail-users mailing list
> Fet...@li...
> https://lists.sourceforge.net/lists/listinfo/fetchmail-users


Would you mind sharing your configs - thanks

Jerry

Re: [Fetchmail-users] Microsoft deprecated

[Mihai L. <mt...@gm...>]

On Monday, October 17, 2022 at 15:23:54 -0400, Jerry Geis wrote:

> I came across this https://github.com/simonrob/email-oauth2-proxy
>
> But I have not got it working yet.

Works for me flawless since mid-August or so on Fedora for both 
Google and Microsoft accounts, for downloading (IMAP, fetchmail) 
and sending (SMTP, mutt).

The only change in the fetchmail and mutt configurations was to 
direct the connection to localhost:<port>, with a local password 
(not the one on servers), without encryption. Of course, after 
configuring email-oauth2-proxy. :-)

HTH,
Mihai

Re: [Fetchmail-users] Microsoft deprecated

[Jerry G. <jer...@gm...>]

I came across this https://github.com/simonrob/email-oauth2-proxy

But I have not got it working yet.

jerry