gnus-summary-mail-forward - included headers

gnus-summary-mail-forward - included headers

[Jim Davidson]

Thomas Hungenberg <th@demonium.de> writes:

> I'm using PGnus 0.95 and I wonder if it is possible to set the
> header-lines which should be included when doing a
> "gnus-summary-mail-forward" of a news article in any variable.
> I don't like header-lines like "Lines:" to be included in the
> forwarded message.

I agree.  In the released version of Gnus, the function
gnus-summary-mail-forward takes a first argument, full-headers, which
indicates whether to forward all the headers.

In PGnus, this has been modified.  The argument has been renamed to not-used,
and is ignored.  All the headers are included in the forward message.  I much
preferred the former behavior.

-Jim

Re: gnus-summary-mail-forward - included headers

[Kai Großjohann]

Jim Davidson <jim.davidson@liberate.com> writes:

> In PGnus, this has been modified.  The argument has been renamed to
> not-used, and is ignored.  All the headers are included in the
> forward message.  I much preferred the former behavior.

Well, pgnus now does MIME forwards.  And that means that the remote
client is explicitly told that the attachment is an RFC822 message.
And presumably a mail reader is capable of displaying mail messages in
a meaningful way (including hiding some headers, if so desired).

We know that very few mailers are really *good* programs and grok MIME
the way MIME was meant to be grokked.  But one can dream...

kai
-- 
I like BOTH kinds of music.

Re: gnus-summary-mail-forward - included headers

[Toby Speight]

Kai> Kai Großjohann <URL:mailto:Kai.Grossjohann@CS.Uni-Dortmund.DE>

0> In article <vafemg18mi1.fsf@lucy.cs.uni-dortmund.de>, Kai wrote:

Kai> Jim Davidson <jim.davidson@liberate.com> writes:

>> In PGnus, this has been modified.  The argument has been renamed to
>> not-used, and is ignored.  All the headers are included in the
>> forward message.  I much preferred the former behavior.

Kai> Well, pgnus now does MIME forwards.  And that means that the
Kai> remote client is explicitly told that the attachment is an RFC822
Kai> message.  And presumably a mail reader is capable of displaying
Kai> mail messages in a meaningful way (including hiding some headers,
Kai> if so desired).

For me, that isn't the issue.  I'd prefer the default not to include
stuff like "Path", "X-Gnus-Mail-Source" and "Xref" (call me paranoid,
if you like) and I think that the likes of "X-From-Line", "X-From_",
"Envelope-to" and "Delivery-date" (added by the mail delivery agent)
and probably also "Received" are a waste of space, except if you are
debugging message delivery problems.

Trimming the headers by hand is a pain, and would be better if
automated, as it was before.


On a related note, what's the best way to narrow to the current part
after inserting a message/rfc822?  I have the following to submit
articles to alt.humor.best-of-usenet:

(defun gnus-summary-forward-to-abhou nil
  (interactive)
  (let ((message-mode-hook '(my-message-mode-hook)))
    (gnus-summary-post-forward)
    (message-goto-newsgroups)
    (insert "alt.humor.best-of-usenet")
    (and (search-forward mail-header-separator nil t)
         (let ((start (point-marker))
               (end (copy-marker (point-max) t)))
           (message-strip-pgp-sig start end)
           (message-strip-sigfile start end)
           (nuke-trailing-blanks-in-region start end)))
    (set (make-local-variable 'gnus-check-before-posting)
         '(control-chars multiple-headers))
    (set (make-local-variable 'message-syntax-checks)
         '((long-lines . disabled)))))

(defun message-strip-sigfile (start end)
  (interactive "*r")
  (goto-char end)
  ;; first remove trailing blank lines
  (skip-syntax-backward "-")
  (let ((sig-end (point))
        (sig-start (re-search-backward "\n-- *\n" start t)))
    (when (and sig-start
               (<= (count-lines sig-start sig-end) 6) ; 4-line sigblock + "\n-- \n" (2 lines)
               (y-or-n-p "Strip signature? "))
      (undo-boundary)
      (delete-region sig-start end))))


(defun message-strip-pgp-sig (start end)
  (interactive "*r")
  (goto-char start)
  (while (re-search-forward "^---*BEGIN PGP \\([A-Z ]*\\)--*\n+" end t)
        (cond ((string= "SIGNATURE" (match-string 1))
               (let ((sig-start (match-beginning 0))
                     (sig-end (or (search-forward "\n-----END PGP SIGNATURE-----" end t) end)))
                 (delete-region sig-start sig-end)))
              ((string= "SIGNED MESSAGE" (match-string 1))
               (let ((header-start (match-beginning 0))
                     (header-end (match-end 0))
                     (signed-region-end (or (search-forward "\n--" end t) end)))
                 (goto-char header-end)
                 (while (search-forward "\n- " signed-region-end t)
                   (replace-match "\n"))
                 (delete-region header-start header-end)))
              (t
               (delete-region (match-beginning 0) (match-end 0))))))

(defun nuke-trailing-blanks-in-buffer nil
  (interactive "*")
  (nuke-trailing-blanks-in-region (point-min) (point-max)))
(defun nuke-trailing-blanks-in-region (start end)
  (interactive "*r")
  (save-excursion
    (goto-char start)
    (while (re-search-forward "[ \t]+$" end t)
      (replace-match "" t t))))


but of course, message-strip-sigfile tears off the part's end-tag.

Re: gnus-summary-mail-forward - included headers

[Lars Magne Ingebrigtsen]

Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:

> Trimming the headers by hand is a pain, and would be better if
> automated, as it was before.

I've now added message-ignored-forward-headers.

> On a related note, what's the best way to narrow to the current part
> after inserting a message/rfc822?

I've now added a keystroke for doing this to the MML keymap, but I
haven't written the command.  :-)

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

message-ignored-forward-headers (was: Re: gnus-summary-mail-forward - included headers)

[Thomas Hungenberg]

On 25 Sep 1999 11:18:33 +0200, Lars Magne Ingebrigtsen <larsi@gnus.org> said:

> Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:
>> Trimming the headers by hand is a pain, and would be better if
>> automated, as it was before.

> I've now added message-ignored-forward-headers.

In which version this has been added?
I can't find it in 0.97 ?!

Aaah, I just took a look at message.el and noticed that the
variable is named "message-forward-ignored-headers" instead of
"message-ignored-forward-headers" there.


      - Thomas

-- 
Thomas Hungenberg -- Student of Applied Computer Science
Private homepage: http://home.pages.de/~th/ -- ICQ-UIN: #3294439
PGP encrypted mail welcome! -- PGP-Public-Key on my homepage!

Re: message-ignored-forward-headers (was: Re: gnus-summary-mail-forward - included headers)

[Jack Vinson]

>>>>> "TH" == Thomas Hungenberg <th@demonium.de> writes:

TH> Aaah, I just took a look at message.el and noticed that the
TH> variable is named "message-forward-ignored-headers" instead of
TH> "message-ignored-forward-headers" there.

Does this actually work for anyone?  I tried to forward a message and
discovered that the forwarded part contained all the headers I didn't want
anyway.

Stepping through the code, it looked like this was being applied to the
header of the message being created.  It should be applied to the forwarded
PART that is a message instead.

-- 
Jack Vinson
Bart: I am not a dentist.

Re: gnus-summary-mail-forward - included headers

[François Pinard]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> > On a related note, what's the best way to narrow to the current part
> > after inserting a message/rfc822?

> I've now added a keystroke for doing this to the MML keymap, but I
> haven't written the command.  :-)

By the way, in my little `fp-mml.el' exercise, I have a function for that:

(defun fp-mml-edit-current-part ()
  "Setup an indirect buffer to edit the MML part containing point."

I did not resend `fp-mml.el' to this group, as it did not change in a
while, but just ask if you want to peek at it.  Or maybe I should just
put it somewhere for people to `wget'?

-- 
François Pinard   http://www.iro.umontreal.ca/~pinard

Re: gnus-summary-mail-forward - included headers

[Kai Großjohann]

François Pinard <pinard@iro.umontreal.ca> writes:

> [...]  Or maybe I should just put it somewhere for people to `wget'?

Yes, please.  I always wanted to try it, but I was swamped...

kai
-- 
Life is hard and then you die.

Re: gnus-summary-mail-forward - included headers

[Hrvoje Niksic]

François Pinard <pinard@iro.umontreal.ca> writes:

> I did not resend `fp-mml.el' to this group, as it did not change in
> a while, but just ask if you want to peek at it.  Or maybe I should
> just put it somewhere for people to `wget'?

Please do.  I'd like to try it under XEmacs.

I added the support for indirect buffers to XEmacs, but they've never
been tested extensively because I know of no application that actually
uses them.  I'd love to see one.

Re: message-ignored-forward-headers (was: Re: gnus-summary-mail-forward - included headers)

[Jim Davidson]

Jack Vinson <vinson@unagi.cis.upenn.edu> writes:

> >>>>> "TH" == Thomas Hungenberg <th@demonium.de> writes:
> 
> TH> Aaah, I just took a look at message.el and noticed that the
> TH> variable is named "message-forward-ignored-headers" instead of
> TH> "message-ignored-forward-headers" there.
> 
> Does this actually work for anyone?  I tried to forward a message and
> discovered that the forwarded part contained all the headers I didn't want
> anyway.
> 
> Stepping through the code, it looked like this was being applied to the
> header of the message being created.  It should be applied to the forwarded
> PART that is a message instead.
> 
> -- 
> Jack Vinson
> Bart: I am not a dentist.

I'm experiencing the same behavior.  It appears that the function 
        message-forward

includes a spurious call to 
        (message-narrow-to-head)

-- 
Jim Davidson
jdavidso at csi.com

Re: message-ignored-forward-headers (was: Re: gnus-summary-mail-forward - included headers)

[Jack Vinson]

>>>>> "JD" == Jim Davidson <jdavidso@nospam.com> writes:

JD> Jack Vinson <vinson@unagi.cis.upenn.edu> writes:
>> 
>> Does this actually work for anyone?  I tried to forward a message and
>> discovered that the forwarded part contained all the headers I didn't want
>> anyway.
>> 

JD> I'm experiencing the same behavior.  It appears that the function
JD> message-forward includes a spurious call to
JD>         (message-narrow-to-head)

Yep.  Comment out this call at the end of message-forward and all is well.
The *real* problem is that message-narrow-to-head calls (widen) as its
first activity.  As a result, the message is narrowed to its TRUE head
instead of the head of the forwarded message.

Lars, which is more important to fix?  It looks like all the
message-narrow-to.*-head.* functions call widen first.

-- 
Jack Vinson
Bart: No one is interested in my underpants. 

Re: message-ignored-forward-headers (was: Re: gnus-summary-mail-forward - included headers)

[Lars Magne Ingebrigtsen]

Jack Vinson <vinson@unagi.cis.upenn.edu> writes:

> Yep.  Comment out this call at the end of message-forward and all is well.
> The *real* problem is that message-narrow-to-head calls (widen) as its
> first activity.  As a result, the message is narrowed to its TRUE head
> instead of the head of the forwarded message.

Yup.  Fix in Pterodactyl Gnus v0.98.1.

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

Re: gnus-summary-mail-forward - included headers

[Lars Magne Ingebrigtsen]

François Pinard <pinard@iro.umontreal.ca> writes:

> I did not resend `fp-mml.el' to this group, as it did not change in a
> while, but just ask if you want to peek at it.  Or maybe I should just
> put it somewhere for people to `wget'?

Please do post it again -- I think I must have missed it the first
time around.

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

[patch] message-forward-as-mime (was: Re: message-ignored-forward-headers (was: Re: gnus-summary-mail-forward - included headers))

[Matt Pharr]

[-- Attachment #1: Type: text/plain, Size: 644 bytes --]


While we're on the topic of forwarding messages, what was the final verdict 
on providing a version of message-forward that doesn't do make the message
a MIME part?  I know that this has come up before but the old non-MIME way
was never reimplemented.

If there's no objection, the following patch adds a message-forward-as-mime
variable which is true by default, preserving the current behavior.  When
nil, however, forwarded messages are just inserted as inline text like the
old days.  I almost always don't want to have forwards be MIME parts, since
most of the people I correspond with still don't have very MIME-aware mail
readers. :-(


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: message-forward-as-mime patch --]
[-- Type: text/x-patch, Size: 1662 bytes --]

*** lisp/message.el~	Mon Nov  8 14:30:43 1999
--- lisp/message.el	Mon Nov  8 14:28:52 1999
***************
*** 295,300 ****
--- 295,305 ----
   :type '(radio (function-item message-forward-subject-author-subject)
  	       (function-item message-forward-subject-fwd)))
  
+ (defcustom message-forward-as-mime t
+   "*If non-nil, forward messages as an inline/rfc822 MIME section.  Otherwise, directly inline the old message in the forwarded message."
+   :group 'message-forwarding
+   :type 'boolean)
+ 
  (defcustom message-wash-forwarded-subjects nil
    "*If non-nil, try to remove as much old cruft as possible from the subject of messages before generating the new subject of a forward."
    :group 'message-forwarding
***************
*** 3821,3832 ****
      ;; Put point where we want it before inserting the forwarded
      ;; message.
      (message-goto-body)
!     (insert "\n\n<#part type=message/rfc822 disposition=inline>\n")
      (let ((b (point))
  	  e)
        (mml-insert-buffer cur)
        (setq e (point))
!       (insert "<#/part>\n")
        (when message-forward-ignored-headers
  	(save-restriction
  	  (narrow-to-region b e)
--- 3826,3840 ----
      ;; Put point where we want it before inserting the forwarded
      ;; message.
      (message-goto-body)
!     (if message-forward-as-mime
! 	 (insert "\n\n<#part type=message/rfc822 disposition=inline>\n")
!       (insert "\n\n"))
      (let ((b (point))
  	  e)
        (mml-insert-buffer cur)
        (setq e (point))
!       (and message-forward-as-mime
! 	   (insert "<#/part>\n"))
        (when message-forward-ignored-headers
  	(save-restriction
  	  (narrow-to-region b e)

[-- Attachment #3: Type: text/plain, Size: 122 bytes --]


-matt
-- 
Matt Pharr                                   mmp@graphics.stanford.edu
<URL:http://graphics.stanford.edu/~mmp>

Re: gnus-summary-mail-forward - included headers

[Dan Christensen]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:
> 
> > Trimming the headers by hand is a pain, and would be better if
> > automated, as it was before.
> 
> I've now added message-ignored-forward-headers.

I don't see this in the source or in the documentation, although
I do see a Changelog entry.  What's up?

Dan

-- 
Dan Christensen
jdc@math.jhu.edu

Re: gnus-summary-mail-forward - included headers

[Matt Pharr]

Dan Christensen <jdc@jhu.edu> writes:
> Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> > Toby Speight <Toby.Speight@streapadair.freeserve.co.uk> writes:
> > 
> > > Trimming the headers by hand is a pain, and would be better if
> > > automated, as it was before.
> > 
> > I've now added message-ignored-forward-headers.
> 
> I don't see this in the source or in the documentation, although
> I do see a Changelog entry.  What's up?

Actually, it's "message-forward-ignored-headers"

-matt
-- 
Matt Pharr                                   mmp@graphics.stanford.edu
<URL:http://graphics.stanford.edu/~mmp>

Re: [patch] message-forward-as-mime (was: Re: message-ignored-forward-headers (was: Re: gnus-summary-mail-forward - included headers))

[Toni Drabik]

Matt Pharr <mmp@graphics.stanford.edu> writes:

> While we're on the topic of forwarding messages, what was the final
> verdict on providing a version of message-forward that doesn't do make the
> message a MIME part?

Lars rejected it. AFAIK, there were some attempts to make him change his
mind (hi, Hrvoje) -- unfortunately, their outcome wasn't succesful.


> If there's no objection, the following patch adds a
> message-forward-as-mime variable which is true by default, preserving the
> current behavior. When nil, however, forwarded messages are just inserted
> as inline text like the old days.

I don't know if it'll end up in future Gnusae, FWIW I'll apply it to my
local copy of pgnus. Thanks!

-- 
Toni Drabik <tdrabik@public.srce.hr>
Warning: This article may be fatal if swallowed.