Problem with "Send bounced message"

Problem with "Send bounced message"

[Karl Eichwalder]

[-- Attachment #1: Type: text/plain, Size: 467 bytes --]

Running "Send bounced message" (S D b) on the attached message does not
work properly.  Gnus skips this part:

    --7CZp05NP8/gJM8Cl
    Content-Type: text/plain; charset=iso-8859-1
    Content-Disposition: inline
    Content-Transfer-Encoding: quoted-printable
    
    Martin, could you please help me to determine why this submission gets
    rejected while any other from Luc gets accepted ?
    
    Thanks, Mt.

And sends the included forwarded message only.


[-- Attachment #2: art.txt.gz --]
[-- Type: application/x-gzip, Size: 2450 bytes --]

[-- Attachment #3: Type: text/plain, Size: 214 bytes --]



-- 
                                                         |      ,__o
                                                         |    _-\_<,
http://www.gnu.franken.de/ke/                            |   (*)/'(*)

Re: Problem with "Send bounced message"

[Katsumi Yamaoka]

[-- Attachment #1: Type: text/plain, Size: 593 bytes --]

Hi,

>>>>> In <shu163xfsx.fsf@tux.gnu.franken.de>
>>>>>	Karl Eichwalder <ke@gnu.franken.de> wrote:

> Running "Send bounced message" (S D b) on the attached message does not
> work properly.  Gnus skips this part:

[...]

> And sends the included forwarded message only.

It is because there is the Return-Path header in the forwarded part
of the original message.  You may get the correct message by removing
it using the `e' command (M-x gnus-summary-edit-article) in advance.

I also made a patch (attached) to message.el.  Although I cannot have
belief that it is effective in all cases.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: Type: text/x-patch, Size: 1136 bytes --]

--- message.el~	2003-10-19 22:09:05 +0000
+++ message.el	2003-10-21 07:20:34 +0000
@@ -6203,16 +6203,20 @@
       (mm-insert-part handles)
       (undo-boundary)
       (goto-char (point-min))
-      (search-forward "\n\n" nil t)
-      (if (or (and (re-search-forward message-unsent-separator nil t)
-		   (forward-line 1))
-	      (re-search-forward "^Return-Path:.*\n" nil t))
+      (re-search-forward "\n\n+" nil t)
+      (setq boundary (point))
+      (search-forward "\n\n" nil 'move)
+      (if (or (re-search-backward message-unsent-separator boundary t)
+	      (re-search-backward "^Return-Path:.*\n" boundary t))
 	  ;; We remove everything before the bounced mail.
-	  (delete-region
-	   (point-min)
-	   (if (re-search-forward "^[^ \n\t]+:" nil t)
-	       (match-beginning 0)
-	     (point)))
+	  (progn
+	    (forward-line 1)
+	    (delete-region
+	     (point-min)
+	     (if (re-search-forward "^[^ \n\t]+:" nil t)
+		 (match-beginning 0)
+	       (point))))
+	(goto-char boundary)
 	(when (re-search-backward "^.?From .*\n" nil t)
 	  (delete-region (match-beginning 0) (match-end 0)))))
     (mm-enable-multibyte)

Re: Problem with "Send bounced message"

[Katsumi Yamaoka]

>>>>> In <b9yvfqj839d.fsf@jpl.org> Katsumi Yamaoka wrote:

> I also made a patch (attached) to message.el.  Although I cannot
> have belief that it is effective in all cases.

> --- message.el~	2003-10-19 22:09:05 +0000
> +++ message.el	2003-10-21 07:20:34 +0000

Although there was a bug, it is improvement and I confirmed that
it doesn't change the original behavior of the message-bounce
function.  So, I've committed the revised code.

Maybe it should decode non-ascii text in the *bounce* buffer.
However, I haven't done so since it may break the MIME structure
or the digital signature.
-- 
Katsumi Yamaoka <yamaoka@jpl.org>