Vacation Fcc

[Dovecot mailing list]

Hi list,

I use Dovecot's Pigeonhole Sieve implementation, and there is one little
feature I'm missing:  When a vacation message is sent, I'd like to keep a
copy of that message in my "sent" mailbox.

There is an RFC draft (draft-ietf-extra-sieve-fcc-09) for this, which adds
a ":fcc" argument to vacation, but apparently it's not supported by
Pigeonhole.

I have two questions:

Will ":fcc" be supported in the future?

In the meantime (or if supporting ":fcc" is not planned), is it possible
to achieve the same result with the existing Sieve extensions?  (I can
imagine a way to do this at the MTA level by checking the presence of the
"Auto-Submitted:" header, but did not test it.)

Thanks,

Gregory

[Dovecot mailing list]

On 11/10/2019 10:24, Gregory Heytings via dovecot wrote:

>
> Hi list,
>
> I use Dovecot's Pigeonhole Sieve implementation, and there is one
> little feature I'm missing:  When a vacation message is sent, I'd like
> to keep a copy of that message in my "sent" mailbox.
>
> There is an RFC draft (draft-ietf-extra-sieve-fcc-09) for this, which
> adds a ":fcc" argument to vacation, but apparently it's not supported
> by Pigeonhole.

Actually, it is RFC already: https://tools.ietf.org/html/rfc8580 .

> I have two questions:
>
> Will ":fcc" be supported in the future?

I'd say yes, but it is not simple to implement, nor is it high up in our
list of priorities.

> In the meantime (or if supporting ":fcc" is not planned), is it
> possible to achieve the same result with the existing Sieve
> extensions?  (I can imagine a way to do this at the MTA level by
> checking the presence of the "Auto-Submitted:" header, but did not
> test it.)

I can't think of any means to do that within the Sieve language.

Regards,

Stephan.

[Dovecot mailing list]

Hi Stephan,

Thanks for your answer!

>
>> I use Dovecot's Pigeonhole Sieve implementation, and there is one
>> little feature I'm missing:  When a vacation message is sent, I'd like
>> to keep a copy of that message in my "sent" mailbox.
>>
>> There is an RFC draft (draft-ietf-extra-sieve-fcc-09) for this, which
>> adds a ":fcc" argument to vacation, but apparently it's not supported
>> by Pigeonhole.
>
> Actually, it is RFC already: https://tools.ietf.org/html/rfc8580 .
>

Okay, thanks for the pointer!

>
>> I have two questions:
>>
>> Will ":fcc" be supported in the future?
>
> I'd say yes, but it is not simple to implement, nor is it high up in our
> list of priorities.
>

Okay, so I'd better look for another solution in the meantime...

>
>> In the meantime (or if supporting ":fcc" is not planned), is it
>> possible to achieve the same result with the existing Sieve
>> extensions?  (I can imagine a way to do this at the MTA level by
>> checking the presence of the "Auto-Submitted:" header, but did not test
>> it.)
>
> I can't think of any means to do that within the Sieve language.
>

I'll try to find a way to do it at the MTA level, but it seems more
complicated than what I initially thought.

Best,

Gregory

[Dovecot mailing list]

>
>>
>>> In the meantime (or if supporting ":fcc" is not planned), is it
>>> possible to achieve the same result with the existing Sieve
>>> extensions?  (I can imagine a way to do this at the MTA level by
>>> checking the presence of the "Auto-Submitted:" header, but did not
>>> test it.)
>>
>> I can't think of any means to do that within the Sieve language.
>>
>
> I'll try to find a way to do it at the MTA level, but it seems more
> complicated than what I initially thought.
>

For the record, it seems that this can't be done with Postfix, at least
not in a simple way.  One option would be to keep a copy of all emails
with 'always_bcc', and to forward autoreplied emails to the individual
user mailboxes.  Another option would be to write a specific milter, but
this is beyond my abilities.

>
>>> Will ":fcc" be supported in the future?
>>
>> I'd say yes, but it is not simple to implement, nor is it high up in
>> our list of priorities.
>

Perhaps ":fcc" is not simple to implement, but I personally would already
be happy with a ":bcc" option, which would send a blind-carbon copy of the
message to the given email address.  Is this not easier to implement?

Gregory