ignore bcc when followup

ignore bcc when followup

[Uwe Brauer]

Hi

as longer as I think about it: At least optionally the BCC field should
be ignored in the reply even when calling followup.

I look into the code, looks difficult to modify. Any ideas?

Regards

Uwe Brauer 

SOLVED (was: ignore bcc when followup)

[Uwe Brauer]

>>> "Uwe" == Uwe Brauer <oub@mat.ucm.es> writes:

   > Hi
   > as longer as I think about it: At least optionally the BCC field should
   > be ignored in the reply even when calling followup.

   > I look into the code, looks difficult to modify. Any ideas?

   > Regards

   > Uwe Brauer 


The following code does this

(add-hook 'gnus-message-setup-hook 'my-check-bcc)
(defun my-check-bcc ()
  "Try to avoid painful situations when using followup, and there is a Bcc field."
  (interactive)
  (cond ((and gnus-article-reply
              (message-with-reply-buffer
                (re-search-forward
                 "^Bcc:\\|BCC:\\|bcc:"
                 nil t)))
         (condition-case nil
             (progn
               (message-remove-header "cc")
               (message-goto-body)
               (message "There is a BCC field, we have deleted it!!!"))
           (error
            (progn
                (message "Followup??")))))))

Re: ignore bcc when followup

[Adam Sjøgren]

Uwe writes:

> as longer as I think about it: At least optionally the BCC field should
> be ignored in the reply even when calling followup.

Uhm, do you get a Bcc-header when you follow up to an email with a
Bcc-header in it?

I don't.

(How do you even receive an email with a Bcc header in it? Usually they
aren't sent - what email program sent it?)

I just made an experiment: I sent myself an email, edited it after
receipt, adding a Bcc (and Gcc, just for fun) header, and hit F.

The headers in the follow up that did not include the Bcc (nor the Gcc)
from the email.

Original headers:

  From: Adam Sjøgren <asjoegren@gmail.com>
  Subject: Test
  To: Adam Sjøgren <asjo@koldfront.dk>
  Date: Wed, 22 Feb 2017 20:51:10 +0100 (1 minute, 24 seconds ago)
  Bcc: test@koldfront.dk
  Gcc: abba@koldfront.dk
  Attachment: [3. text/html]...

Follow up headers:

  To: Adam Sjøgren <asjoegren@gmail.com>
  Subject: Re: Test
  From: asjo@koldfront.dk (Adam Sjøgren)
  Gcc: nnml+archive:mail-2017
  Organization: koldfront - analysis & revolution, Copenhagen, Denmark
  OpenPGP: id=476630590A231909B0A0961A49D0746121BDE416; url=https://asjo.koldfront.dk/gpg.asc
  X-Now-Playing: Den vänstra stranden, Då som nu för alltid (Kent)
  X-Hashcash: 1:21:170222:asjoegren@gmail.com::BVrk/jbmN3tCC90h:000000000000000000000000000000000000000000JJU3
  --text follows this line--

How did you make Gnus include them?


  Best regards,

    Adam

-- 
 "People who post screen shots of text are jerks."            Adam Sjøgren
                                                         asjo@koldfront.dk

Re: ignore bcc when followup

[Uwe Brauer]

>>> "Adam" == Adam Sjøgren <asjo@koldfront.dk> writes:

   > Uwe writes:
   >> as longer as I think about it: At least optionally the BCC field should
   >> be ignored in the reply even when calling followup.

   > Uhm, do you get a Bcc-header when you follow up to an email with a
   > Bcc-header in it?

No. I am saying (see below my experiment.)

I receive an email which goes like

From: Joe
To: John
Bcc: Uwe

Now I, Uwe hit, by mistake, followup, instead of reply, then they mail
looks at it should

From: Uwe
To: Joe
CC: John

Which is bad, because John did not know that mail was sent to me. I
should have used reply instead of followup, but since a lot of people
complained that I hit reply instead of followup, I know hit followup
automatically.

So my small function will save me such embarrassing situations in the
future.

Did I explain myself?

Below the experiments I did.



All can I say I sent an email to myself, with a bcc to another email

Here is what I received
,----
| 
| From: Uwe Brauer <oub@mat.ucm.es>
| Subject: test bcc
| To: Uwe Brauer <oub@mat.ucm.es> 
| Date: Wed, 22 Feb 2017 21:26:58 +0000 (1 minute, 5 seconds ago)
| Bcc: oub.oub.oub@gmail.com
| Reply-To: Uwe Brauer <oub@mat.ucm.es>
| 
| adfadf
`----

And the other way around

,----
| From: Uwe Brauer <oub@mat.ucm.es>
| Subject: test 2
| To: Uwe Brauer <oub.oub.oub@gmail.com> 
| Date: Wed, 22 Feb 2017 21:27:30 +0000 (48 seconds ago)
| Bcc: oub@mat.ucm.es
| Reply-To: Uwe Brauer <oub@mat.ucm.es>
| 
| This is true
`----


   > I don't.

   > (How do you even receive an email with a Bcc header in it? Usually they
   > aren't sent - what email program sent it?)

The email in equation was sent to me from someone using apple mail

   > I just made an experiment: I sent myself an email, edited it after
   > receipt, adding a Bcc (and Gcc, just for fun) header, and hit F.

   > The headers in the follow up that did not include the Bcc (nor the Gcc)
   > from the email.

   > Original headers:

   >   From: Adam Sjøgren <asjoegren@gmail.com>
   >   Subject: Test
   >   To: Adam Sjøgren <asjo@koldfront.dk>
   >   Date: Wed, 22 Feb 2017 20:51:10 +0100 (1 minute, 24 seconds ago)
   >   Bcc: test@koldfront.dk
   >   Gcc: abba@koldfront.dk
   >   Attachment: [3. text/html]...

   > Follow up headers:

   >   To: Adam Sjøgren <asjoegren@gmail.com>
   >   Subject: Re: Test
   >   From: asjo@koldfront.dk (Adam Sjøgren)
   >   Gcc: nnml+archive:mail-2017
   >   Organization: koldfront - analysis & revolution, Copenhagen, Denmark
   >   OpenPGP: id=476630590A231909B0A0961A49D0746121BDE416; url=https://asjo.koldfront.dk/gpg.asc
   >   X-Now-Playing: Den vänstra stranden, Då som nu för alltid (Kent)
   >   X-Hashcash: 1:21:170222:asjoegren@gmail.com::BVrk/jbmN3tCC90h:000000000000000000000000000000000000000000JJU3
   >   --text follows this line--

   > How did you make Gnus include them?


   >   Best regards,

   >     Adam

Re: ignore bcc when followup

[Adam Sjøgren]

Uwe writes:

> I receive an email which goes like
>
> From: Joe
> To: John
> Bcc: Uwe
>
> Now I, Uwe hit, by mistake, followup, instead of reply, then they mail
> looks at it should
>
> From: Uwe
> To: Joe
> CC: John
>
> Which is bad, because John did not know that mail was sent to me.

Ah, I misunderstood you. I thought you meant that the follow up
contained a copy of the Bcc.

It would make sense for Gnus to ask you, when doing a wide reply/follow
up (i.e. including the original recipient) to an email you were Bcc'ed
on, if that is really what you meant.

I have an advice on gnus-summary-reply that asks me if I really wanted
to reply, if I am in a newsgroup (where I usually want to follow up) -
similar to that.

> Did I explain myself?

Yup.

I actually did what you described just the other day.


  Best regards,

    Adam

-- 
 "Vi är inte riktigt kloka när man tänker efter."             Adam Sjøgren
                                                         asjo@koldfront.dk

Re: ignore bcc when followup

[Adam Sjøgren]

Uwe writes:

>    > (How do you even receive an email with a Bcc header in it? Usually they
>    > aren't sent - what email program sent it?)
>
> The email in equation was sent to me from someone using apple mail

That is odd behaviour of Apple Mail, I think.

Gnus, mutt and Gmail all remove the Bcc header from both the email sent
to the original recipient AND from the blind carbon copy.

I've never seen a Bcc-header on an incoming email. Guess I don't know
anybody Bcc'ing me from Apple Mail :-)

This does, however, make it harder to detect when to warn - if the
incoming email is to a mailing list, your address might very well not be
present in neither To nor Cc, but you don't want a warning.


  Best regards,

    Adam

-- 
 "A cat has nine lives, but a bullfrog croaks                 Adam Sjøgren
  every day."                                            asjo@koldfront.dk

Re: SOLVED

[David Engster]

Uwe Brauer writes:
>>>> "Uwe" == Uwe Brauer <oub@mat.ucm.es> writes:
>
>    > Hi
>    > as longer as I think about it: At least optionally the BCC field should
>    > be ignored in the reply even when calling followup.
>
>    > I look into the code, looks difficult to modify. Any ideas?
>
>    > Regards
>
>    > Uwe Brauer 
>
>
> The following code does this

Be warned that this will not always work. The handling of Bcc is a chaos
and depends on the involved MUAs and MTAs. Some MTAs completely strip
the Bcc field, some don't. Some MUAs send a separate mail for the people
in the Bcc, so that at least they cannot do a reply all.

The best solution is to simply tell everyone to not use Bcc, but to do
separate mails instead, even if it is cumbersome.

What would be a good feature for Gnus is to better highlight if you
received this mail via To, CC or none of the above (like mailing lists
or Bcc). What actually happens quite often at my day job is that I merge
review requests that were only sent to me via CC...

-David

Re: Bcc

[Adam Sjøgren]

David writes:

> Some MUAs send a separate mail for the people in the Bcc

Which MUAs do not do this?

Isn't that the definition of what Bcc does?


  Best regards,

    Adam

-- 
 "The success of open source code is perhaps the only         Adam Sjøgren
  thing in the computer field that hasn't surprised me   asjo@koldfront.dk
  during the past several decades."

Re: ignore bcc when followup

[Uwe Brauer]

>>> "Adam" == Adam Sjøgren <asjo@koldfront.dk> writes:

    > Uwe writes:

    >> > (How do you even receive an email with a Bcc header in it?
    >> > Usually they aren't sent - what email program sent it?)
    >> 
    >> The email in equation was sent to me from someone using apple mail

    > That is odd behaviour of Apple Mail, I think.

    > Gnus, mutt and Gmail all remove the Bcc header from both the email sent
    > to the original recipient AND from the blind carbon copy.

No they don't at least gnus does not, in my previous email the example I
gave were generated by gnus, that is gnus display the BCC field, in a
message which was sent my gnus, I wonder why he does not do it for you.

Uwe 

Re: ignore bcc when followup

[Adam Sjøgren]

Uwe writes:

>>>> "Adam" == Adam Sjøgren <asjo@koldfront.dk> writes:

>     > Gnus, mutt and Gmail all remove the Bcc header from both the email sent
>     > to the original recipient AND from the blind carbon copy.

> No they don't at least gnus does not, in my previous email the example I
> gave were generated by gnus, that is gnus display the BCC field, in a
> message which was sent my gnus, I wonder why he does not do it for you.

Gnus does not send the Bcc header neither to the recipient in To: nor to
the recipient in Bcc:.

Example:

I am sending this email from Gnus:

  To: asjoegren@gmail.com
  Bcc: asjo@mcom.com
  Subject: Test of Bcc
  From: asjo@koldfront.dk (Adam Sjøgren)
  Gcc: nnml+archive:mail-2017
  Organization: koldfront - analysis & revolution, Copenhagen, Denmark
  OpenPGP: id=476630590A231909B0A0961A49D0746121BDE416; url=https://asjo.koldfront.dk/gpg.asc
  --text follows this line--
  Test of Bcc

  -- 
   "We are all in the gutter, but some of us are looking        Adam Sjøgren
    at the stars."                                         asjo@koldfront.dk

This is the conversation between Gnus and the mail server:

  220 tullinup.koldfront.dk ESMTP Postfix (Debian/GNU).
  EHLO tullinup.
  250-tullinup.koldfront.dk.
  250-PIPELINING.
  250-SIZE 209715200.
  250-VRFY.
  250-ETRN.
  250-STARTTLS.
  250-ENHANCEDSTATUSCODES.
  250-8BITMIME.
  250-DSN.
  250 SMTPUTF8.
  MAIL FROM:<asjo@koldfront.dk> SIZE=1032.
  250 2.1.0 Ok.
  RCPT TO:<asjo@mcom.com>.
  250 2.1.5 Ok.
  RCPT TO:<asjoegren@gmail.com>.
  250 2.1.5 Ok.
  DATA.
  354 End data with <CR><LF>.<CR><LF>.
  From: asjo@koldfront.dk (Adam =?utf-8?Q?Sj=C3=B8gren?=)
  To: asjoegren@gmail.com.
  Subject: Test of Bcc.
  Organization: koldfront - analysis & revolution, Copenhagen, Denmark.
  User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.0.50 (gnu/linux).
  X-Hashcash: 1:21:170223:asjoegren@gmail.com::qtSNPl6AEL11AAqE:000000000000000000000000000000000000000000mh2T.
  OpenPGP: id=476630590A231909B0A0961A49D0746121BDE416; url=https://asjo.koldfront.dk/gpg.asc.
  X-Face: )qY&CseJ?.:=8F#^~GcSA?F=9eu'{KAFfL1C3/A&:nE?PW\i65"ba0NS)97,Q(^@xk}n4Ou.
   rPuR#V8I(J_@~H($[ym:`K_+]*kjvW>xH5jbgLBVFGXY:(#4P>zVBklLbdL&XxL\M)%T}3S/IS9lMJ.
   ^St'=VZBR<gm`!Dj`dIpp?+$"$l_'JKDN\w-jB;fo0Qy}Tbw.
  Date: Thu, 23 Feb 2017 13:44:15 +0100.
  Message-ID: <87y3wx13hc.fsf@tullinup.koldfront.dk>.
  MIME-Version: 1.0.
  Content-Type: text/plain; charset=utf-8.
  Content-Transfer-Encoding: quoted-printable.
  .
  Test of Bcc.
  .
  --=20.
   "We are all in the gutter, but some of us are looking        Adam Sj=C3=B8=.
  gren.
    at the stars."                                         asjo@koldfront.dk.
  ..

  250 2.0.0 Ok: queued as 8542210053471.

As you can see, Gnus does NOT send the Bcc-header, it sends the email
without the Bcc-header to the two recipients.

It does, however, store the Bcc header in the *local copy* of the email,
so *you* can see that you did in fact Bcc.


  Best regards,

    Adam

-- 
 "It's not denial. I'm just very selective about the          Adam Sjøgren
  reality I accept."                                     asjo@koldfront.dk

Re: ignore bcc when followup

[Uwe Brauer]

[-- Attachment #1: Type: text/plain, Size: 692 bytes --]

>>> "Adam" == Adam Sjøgren <asjo@koldfront.dk> writes:

   > Uwe writes:
   >>>>> "Adam" == Adam Sjøgren <asjo@koldfront.dk> writes:

   >> > Gnus, mutt and Gmail all remove the Bcc header from both the email sent
   >> > to the original recipient AND from the blind carbon copy.

   >> No they don't at least gnus does not, in my previous email the example I
   >> gave were generated by gnus, that is gnus display the BCC field, in a
   >> message which was sent my gnus, I wonder why he does not do it for you.

   > Gnus does not send the Bcc header neither to the recipient in To: nor to
   > the recipient in Bcc:.

For me it does, see the screenshots hope they will not be blocked



[-- Attachment #2: gnus-bcc.png --]
[-- Type: image/png, Size: 84494 bytes --]

[-- Attachment #3: seam-bcc1.png --]
[-- Type: image/png, Size: 152126 bytes --]

Re: ignore bcc when followup

[Adam Sjøgren]

Uwe writes:

>>>> "Adam" == Adam Sjøgren <asjo@koldfront.dk> writes:

>    > Gnus does not send the Bcc header neither to the recipient in To: nor to
>    > the recipient in Bcc:.
>
> For me it does, see the screenshots hope they will not be blocked

And for me, as you saw, it doesn't. Weird.

What method do you use to send email in Gnus?


  Best regards,

    Adam

-- 
 "I am not tropical. I am not a damn toucan."                 Adam Sjøgren
                                                         asjo@koldfront.dk

Re: ignore bcc when followup

[Uwe Brauer]

>>> "Adam" == Adam Sjøgren <asjo@koldfront.dk> writes:

   > Uwe writes:
   >>>>> "Adam" == Adam Sjøgren <asjo@koldfront.dk> writes:

   >> > Gnus does not send the Bcc header neither to the recipient in To: nor to
   >> > the recipient in Bcc:.
   >> 
   >> For me it does, see the screenshots hope they will not be blocked

   > And for me, as you saw, it doesn't. Weird.

Indeed. Maybe it is the setting? Either yours or mine?


   > What method do you use to send email in Gnus?

Smtpmail via tls to the server of my university, which is a gmail server
(since my university could not afford its own server anymore)

Re: Bcc

[David Engster]

Adam Sjøgren writes:
> David writes:
>
>> Some MUAs send a separate mail for the people in the Bcc
>
> Which MUAs do not do this?
>
> Isn't that the definition of what Bcc does?

To my knowledge, there's no definition how BCC is actually
implemented. RFC822 only states that the BCC header must not reach the
primary/secondary recipients (and it explicitly does not say whether the
BCC header should be seen by the tertiary recipients, as I've written to
Uwe).

I'm gone from system administration quite a while, so things might have
changed, but I remember that there were long discussions who is actually
responsible for *implementing* the BCC behavior: the MUA or the
MTA. Many MTAs take it upon themselves to handle BCC: they strip the BCC
header and then forwarded the mails to all recipients. I remember that
Exim did *not* do that by default, which a quick search shows still
surprises people:

https://www.bentasker.co.uk/documentation/linux/301-avoiding-bcc-leaks-with-exim

Now that might not be a proper MUA, but I'm pretty sure I've seen MUAs
back in the day that delegated BCC handling to the MTA. They probably
changed that when Exim became popular.

-David