resending & empty subjects

resending & empty subjects

[Vladimir Volovich]

Hi,

when resending a message with empty subject, gnus asks me to enter the
subject, which seems a bit wrong.

Also, while we are at this topic: when resending a message, gnus does
not remove the X-Gnus-Newsgroup header, X-From-Line too, which also
seems wrong (first, there were no such headers in the original
message, so they should not be present in the resent message, and
second, it may easily confuse gnus on the recipient's side).

	Best regards, -- Vladimir.

Re: resending & empty subjects

[Lars Magne Ingebrigtsen]

Vladimir Volovich <vvv@vvv.vsu.ru> writes:

> when resending a message with empty subject, gnus asks me to enter the
> subject, which seems a bit wrong.

Try to `(setq debug-on-quit t)' and then `C-g' when it asks.  Mail me
the resulting backtrace.

> Also, while we are at this topic: when resending a message, gnus does
> not remove the X-Gnus-Newsgroup header,

Yup.  Fix in Pterodactyl Gnus v0.43.

> X-From-Line too,

I think that one should stay to preserve the chain of events.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: resending & empty subjects

[Vladimir Volovich]

"LMI" == Lars Magne Ingebrigtsen writes:

 >> when resending a message with empty subject, gnus asks me to enter
 >> the subject, which seems a bit wrong.
 LMI> Try to `(setq debug-on-quit t)' and then `C-g' when it asks.
 LMI> Mail me the resulting backtrace.

This is from pgnus-0.36:

Signaling: (quit)
  read-from-minibuffer("Empty header for Subject; enter value: ")
  message-generate-headers((From Subject Date (optional . In-Reply-To) Message-ID Lines (optional . User-Agent)))
  message-send-mail()
  message-resend("root")
  gnus-summary-resend-message("root" nil)
* call-interactively(gnus-summary-resend-message)

 >> X-From-Line too,
 LMI> I think that one should stay to preserve the chain of events.

This seems Ok for me, too, unless if the RFC forbids to add inexistent
headers (i've not checked that).

	Best regards, -- Vladimir.

Re: resending & empty subjects

[Edward J. Sabol]

Excerpts from mail: (13-Nov-98) Re: resending & empty subjects by Lars Magne Ingebrigtsen
>> X-From-Line too,
> I think that one should stay to preserve the chain of events.

It's no big deal, but I think that might be in bad form. I would prefer it if
"X-From-Line:" were removed. To quote RFC822:

          Some systems permit mail recipients to  forward  a  message,
     retaining  the original headers, by adding some new fields.  This
     standard supports such a service, through the "Resent-" prefix to
     field names.
[...]
    Note:  In general, the "Resent-" fields should be treated as con-
            taining  a  set  of information that is independent of the
            set of original fields.  Information for  one  set  should
            not  automatically be taken from the other.

My interpretation of these passages is that the MUA should not alter the
original headers in any fashion when resending and that it should only add a
set of standard fields all of which start with the "Resent-" prefix. The
"X-From-Line:" should only be in the resent e-mail if it was in the original
e-mail as well. If Gnus is adding "X-From-Line:" as part of some internal
meta-information tracking mechanism, then it should remove it before
resending. IMHO, of course.

Re: resending & empty subjects

[Lars Magne Ingebrigtsen]

Vladimir Volovich <vvv@vvv.vsu.ru> writes:

> This is from pgnus-0.36:
> 
> Signaling: (quit)
>   read-from-minibuffer("Empty header for Subject; enter value: ")
>   message-generate-headers((From Subject Date (optional . In-Reply-To) Message-ID Lines (optional . User-Agent)))
>   message-send-mail()

Yup.  Fix in Pterodactyl Gnus v0.43.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: resending & empty subjects

[Lars Magne Ingebrigtsen]

"Edward J. Sabol" <sabol@alderaan.gsfc.nasa.gov> writes:

> My interpretation of these passages is that the MUA should not alter the
> original headers in any fashion when resending and that it should only add a
> set of standard fields all of which start with the "Resent-" prefix.

RFC822 doesn't say anything about whether original headers should be
retained or not when resending.  It only discusses adding more
information using the Resent-* mechanism.

> The "X-From-Line:" should only be in the resent e-mail if it was in
> the original e-mail as well.

The X-From-Line is the sendmail envelope "From " header.  It makes as
much sense to retain this header as the Received headers, which
weren't in the original message either, but were also added by the
transport mechanism.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen