mml problem for gnupgp

mml problem for gnupgp

[Uwe Brauer]

[-- Attachment #1: Type: text/plain, Size: 351 bytes --]




Hello


The following puzzles me: I have used successfully mml with gnupg for
a while. Recently I started again using it.
However when I try to sign or encrypt a message I receive after I am
asked for the passwd,  the following error message which I attach below. 

I test gpg on the command line and there seems no problem

Thanks 

Uwe Brauer 



[-- Attachment #2: mml-bug --]
[-- Type: application/octet-stream, Size: 736 bytes --]

Debugger entered--Lisp error: (search-failed "

")
  search-forward("\n\n")
  pgg-decode-armor-region(1 1)
  pgg-parse-armor("")
  mml2015-pgg-sign((part (sign . "pgpmime") (tag-location . 351) (contents . "This is a test \n")))
  mml2015-sign((part (sign . "pgpmime") (tag-location . 351) (contents . "This is a test \n")))
  mml-pgpmime-sign-buffer((part (sign . "pgpmime") (tag-location . 351) (contents . "This is a test \n")))
  mml-generate-mime-1((part (sign . "pgpmime") (tag-location . 351) (contents . "This is a test \n")))
  mml-generate-mime()
  message-encode-message-body()
  message-send-mail(nil)
  message-send-via-mail(nil)
  message-send(nil)
  message-send-and-exit(nil)
  call-interactively(message-send-and-exit)

[-- Attachment #3: Type: text/plain, Size: 161 bytes --]

_______________________________________________
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english

Re: mml problem for gnupgp

[Katsumi Yamaoka]

[-- Attachment #1.1: Type: text/plain, Size: 714 bytes --]

>>>>> In <87r6vn8u48.fsf@mat.ucm.es> Uwe Brauer wrote:

> The following puzzles me: I have used successfully mml with gnupg for
> a while. Recently I started again using it.
> However when I try to sign or encrypt a message I receive after I am
> asked for the passwd,  the following error message which I attach below.

> I test gpg on the command line and there seems no problem

> Thanks

> Uwe Brauer

> Debugger entered--Lisp error: (search-failed "

> ")
>   search-forward("\n\n")
>   pgg-decode-armor-region(1 1)

It seems that generating a detached sign failed.  (1 1) means
there is no content in the buffer.  I'm sorry I don't know what
causes it since I have no problem with the latest Gnus and GnuPG.

[-- Attachment #1.2: Type: application/pgp-signature, Size: 191 bytes --]

[-- Attachment #2: Type: text/plain, Size: 161 bytes --]

_______________________________________________
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english

Re: mml problem for gnupgp

[Uwe Brauer]

>>>>> " " == Katsumi Yamaoka <yamaoka@jpl.org> writes:

   >> I test gpg on the command line and there seems no problem

   >> Thanks

   >> Uwe Brauer

   >> Debugger entered--Lisp error: (search-failed "

   >> ")
   >> search-forward("\n\n")
   >> pgg-decode-armor-region(1 1)

    > It seems that generating a detached sign failed.  (1 1) means
    > there is no content in the buffer.  I'm sorry I don't know what
    > causes it since I have no problem with the latest Gnus and GnuPG.

Thanks for your answer, the odd thing is that mozilla/thunderbird with
the enigmail plugin work both fine. So it seems that gnupg works.

On the other hand I even can't neither encrypt nor decrypt messages,
after writing the passwd nothing happens, or when sending it I am not
even asked about the passwd. Very odd.

I also desinstalled my old gnus official XEmacs pgk in case there had
been any conflict. What really puzzled me is that all this worked a
while ago and now not any longer.

Thanks

Uwe Brauer 

Problem could be encode-coding-region (was: mml problem for gnupgp)

[Uwe Brauer]

[-- Attachment #1: Type: text/plain, Size: 372 bytes --]

Hello 

I just run pgg-sign-region and pgg-encrypt-región
for the first command I obtain (void-function encode-coding-region)
for the second it seems a similar error. I attach both backtraces.

I am on XEmacs 21.4.19 no mule. So I tried no mule and then the region
got deleted. So I presume something with my gnupg setting is wrong but
I am not sure what.


Uwe Brauer 



[-- Attachment #2: error-sign --]
[-- Type: application/octet-stream, Size: 1833 bytes --]

Debugger entered--Lisp error: (void-function encode-coding-region)
  encode-coding-region(1 32 nil)
  pgg-sign-region(499 530)
  call-interactively(pgg-sign-region)
  command-execute(pgg-sign-region t)
  execute-extended-command(nil)
  call-interactively(execute-extended-command)
  recursive-edit()
  byte-code("..." [buffer-read-only noninteractive debugger-buffer middlestart debugger-args debugger-batch-max-lines pop-to-buffer debugger-mode debugger-setup-buffer count-lines 2 "...\n" message buffer-string kill-emacs debug backtrace-debug 3 t debugger-reenable "" nil recursive-edit standard-output] 3)
  debug(error (wrong-type-argument listp "oub"))
  signal(wrong-type-argument (listp "oub"))
  byte-code("..." [error pgg-epg-secret-key-id-list pgg-remove-passphrase-from-cache signal] 4)
  pgg-epg-encrypt-region(161 181 ("oub@mat.ucm.es") nil nil)
  funcall(pgg-epg-encrypt-region 161 181 ("oub@mat.ucm.es") nil nil)
  apply(funcall pgg-epg-encrypt-region (161 181 ("oub@mat.ucm.es") nil nil))
  pgg-invoke("encrypt-region" epg 161 181 ("oub@mat.ucm.es") nil nil)
  pgg-encrypt-region(161 181 ("oub@mat.ucm.es"))
  eval((pgg-encrypt-region (region-beginning) (region-end) (quote ("oub@mat.ucm.es"))))
  edit-and-eval-command("Redo: " (set-variable (quote debug-on-error) t) (command-history . 1))
  repeat-complex-command(1)
  call-interactively(repeat-complex-command)
  recursive-edit()
  byte-code("..." [buffer-read-only noninteractive debugger-buffer middlestart debugger-args debugger-batch-max-lines pop-to-buffer debugger-mode debugger-setup-buffer count-lines 2 "...\n" message buffer-string kill-emacs debug backtrace-debug 3 t debugger-reenable "" nil recursive-edit standard-output] 3)
  debug(error (undefined-keystroke-sequence [#<keypress-event control-C> #<keypress-event control-M> #<keypress-event t>]))

[-- Attachment #3: error-encrypt --]
[-- Type: application/octet-stream, Size: 1239 bytes --]

Debugger entered--Lisp error: (wrong-type-argument listp "oub")
  signal(wrong-type-argument (listp "oub"))
  byte-code("..." [error pgg-epg-secret-key-id-list pgg-remove-passphrase-from-cache signal] 4)
  pgg-epg-encrypt-region(161 181 ("oub@mat.ucm.es") nil nil)
  funcall(pgg-epg-encrypt-region 161 181 ("oub@mat.ucm.es") nil nil)
  apply(funcall pgg-epg-encrypt-region (161 181 ("oub@mat.ucm.es") nil nil))
  pgg-invoke("encrypt-region" epg 161 181 ("oub@mat.ucm.es") nil nil)
  pgg-encrypt-region(161 181 ("oub@mat.ucm.es"))
  eval((pgg-encrypt-region (region-beginning) (region-end) (quote ("oub@mat.ucm.es"))))
  edit-and-eval-command("Redo: " (set-variable (quote debug-on-error) t) (command-history . 1))
  repeat-complex-command(1)
  call-interactively(repeat-complex-command)
  recursive-edit()
  byte-code("..." [buffer-read-only noninteractive debugger-buffer middlestart debugger-args debugger-batch-max-lines pop-to-buffer debugger-mode debugger-setup-buffer count-lines 2 "...\n" message buffer-string kill-emacs debug backtrace-debug 3 t debugger-reenable "" nil recursive-edit standard-output] 3)
  debug(error (undefined-keystroke-sequence [#<keypress-event control-C> #<keypress-event control-M> #<keypress-event t>]))

[-- Attachment #4: Type: text/plain, Size: 161 bytes --]

_______________________________________________
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english

vm-pgg works (was: mml problem for gnupgp)

[Uwe Brauer]

Hello

I just tried out the new vm-pgg which brings gnupg support to vm and
it works, so I would think that my gnupg installation is correct (what
I guessed since enigmail works). 

So the culprit but be somehow pgg, which would be odd, since 
vm-pgg calls it. So may be it is  mml-sec ?

Uwe Brauer 

Re: Problem could be encode-coding-region

[Katsumi Yamaoka]

>>>>> In <87ejrm8c2j.fsf_-_@mat.ucm.es> Uwe Brauer wrote:

> I just run pgg-sign-region and pgg-encrypt-región
> for the first command I obtain (void-function encode-coding-region)
> for the second it seems a similar error. I attach both backtraces.

> I am on XEmacs 21.4.19 no mule. So I tried no mule and then the region
> got deleted. So I presume something with my gnupg setting is wrong but
> I am not sure what.

I tested signing and encrypting a messege using `C-c C-m s p'
and `C-c C-m c p' in XEmacs 21.4.19 with no Mule.  No Problem.
Yes, using `pgg-sign-region' and `pgg-encrypt-region' causes
such an error.  But do you really need to use them?  If so,
please contact the PGG author.  I don't modify PGG files because
the author will probably not want me to do it.

OTOH, I found similar things by performing

egrep '\((de|en)code-coding-' *.el

in files other than PGG.  So, I will fix only them later.

Regards,

Re: Problem could be encode-coding-region

[Uwe Brauer]

>>>>> "Katsumi" == Katsumi Yamaoka <yamaoka@jpl.org> writes:

   Katsumi> I tested signing and encrypting a messege using `C-c C-m s p'
   Katsumi> and `C-c C-m c p' in XEmacs 21.4.19 with no Mule.  No Problem.
   Katsumi> Yes, using `pgg-sign-region' and `pgg-encrypt-region' causes
   Katsumi> such an error.  But do you really need to use them?  If so,
   Katsumi> please contact the PGG author.  I don't modify PGG files because
   Katsumi> the author will probably not want me to do it.

I thought  I have to use them, what other signing and encryption
function exists for gnupgp??? My setting is

(setq pgg-cache-passphrase t
       pgg-default-scheme 'gpg
       pgg-default-user-id "oub"
       pgg-passphrase-cache-expiry 1800
       pgg-scheme 'gpg)

Uwe Brauer 

Re: Problem could be encode-coding-region

[Katsumi Yamaoka]

>>>>> In <87u00hnsk9.fsf@mat.ucm.es> Uwe Brauer wrote:

Katsumi> Yes, using `pgg-sign-region' and `pgg-encrypt-region' causes
Katsumi> such an error.  But do you really need to use them?  If so,

> I thought  I have to use them, what other signing and encryption
> function exists for gnupgp??? My setting is

> (setq pgg-cache-passphrase t
>        pgg-default-scheme 'gpg
>        pgg-default-user-id "oub"
>        pgg-passphrase-cache-expiry 1800
>        pgg-scheme 'gpg)

Another way is to use the APEL XEmacs package[1] which provides
some dummy Mule functions except `buffer-file-coding-system'.  I
verified I can use `pgg-sign-region' and `pgg-encrypt-region' in
XEmacs 21.4.19 (with no mule no file-coding) by putting the
following snippet in the XEmacs init file.

--8<---------------cut here---------------start------------->8---
(require 'poem)
(defvar buffer-file-coding-system nil)
--8<---------------cut here---------------end--------------->8---

[1] ftp://ftp.xemacs.org/xemacs/packages/apel-1.32-pkg.tar.gz

Re: Problem could be encode-coding-region

[Uwe Brauer]

>>>>> In <87u00hnsk9.fsf@mat.ucm.es> Uwe Brauer wrote:
   Katsumi> Yes, using `pgg-sign-region' and `pgg-encrypt-region' causes
   Katsumi> such an error.  But do you really need to use them?  If so,

   >> I thought  I have to use them, what other signing and encryption
   >> function exists for gnupgp??? My setting is

   >> (setq pgg-cache-passphrase t
   >> pgg-default-scheme 'gpg
   >> pgg-default-user-id "oub"
   >> pgg-passphrase-cache-expiry 1800
   >> pgg-scheme 'gpg)

    > Another way is to use the APEL XEmacs package[1] which provides
    > some dummy Mule functions except `buffer-file-coding-system'.  I
    > verified I can use `pgg-sign-region' and `pgg-encrypt-region' in
    > XEmacs 21.4.19 (with no mule no file-coding) by putting the
    > following snippet in the XEmacs init file.

    > (require 'poem)
    > (defvar buffer-file-coding-system nil)

You are correct, indeed now pgg-sign-region and pgg-encrypt-region
work! Thank you very much. However and there is a big however when I
use (mml-secure-message-sign-pgpmime) which uses that function

Then the signing seems to be ignored, that is the tag 
<#secure method=pgpmime mode=sign>

 is inserted as it should, asked for the passwd but then nothing
happens.
Shall I debug (mml-secure-message-sign-pgpmime)?

I am very puzzled now. As I said before vm-pgg works. So the culprit
seems to be mml-secure-message-sign-pgpmine, what other function
could I use ????

Uwe Brauer 

Re: Problem could be encode-coding-region

[Katsumi Yamaoka]

[-- Attachment #1.1: Type: text/plain, Size: 1361 bytes --]

>>>>> In <87odqoewsr.fsf@mat.ucm.es> Uwe Brauer wrote:

> However and there is a big however when I use
> (mml-secure-message-sign-pgpmime) which uses that function

> Then the signing seems to be ignored, that is the tag
> <#secure method=pgpmime mode=sign>

>  is inserted as it should, asked for the passwd but then nothing
> happens.

Hmm, I have no problem in using PGG and XEmacs with no mule no
file-coding.  Actually I signed this message by it.

> Shall I debug (mml-secure-message-sign-pgpmime)?

No, you should trace `mml2015-pgg-sign' and `pgg-gpg-sign-region'.
If you'd like to edebug (info "(gnus)Troubleshooting"), perform
`M-x edebug-defun RET' on those functions, and type `C-c C-m s p'
and then `C-u C-c C-m P' in the message buffer.

BTW, what do you get by evaluating the following form?

--8<---------------cut here---------------start------------->8---
(progn
  (require 'pgg)
  (require 'pgg-gpg)
  (setq pgg-default-user-id "oub")
  (with-temp-buffer
    (insert "testing\n")
    (pgg-gpg-sign-region (point-min) (point-max))
    (pop-to-buffer pgg-output-buffer)
    (unless (get-buffer-window pgg-errors-buffer)
      (split-window-vertically)
      (switch-to-buffer pgg-errors-buffer))))
--8<---------------cut here---------------end--------------->8---

(Copy it to the *scratch* buffer and type `C-j' at the end of
the last line.)

[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]

[-- Attachment #2: Type: text/plain, Size: 161 bytes --]

_______________________________________________
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english

Re: Problem could be encode-coding-region

[Katsumi Yamaoka]

>>>>> In <b4mbqmowbsa.fsf@jpl.org> Katsumi Yamaoka wrote:

> BTW, what do you get by evaluating the following form?

[...]

> (Copy it to the *scratch* buffer and type `C-j' at the end of
> the last line.)

Sorry, it should be:

--8<---------------cut here---------------start------------->8---
(progn
  (require 'pgg)
  (require 'pgg-gpg)
  (setq pgg-default-user-id "oub")
  (with-temp-buffer
    (insert "testing\n")
    (let ((pgg-text-mode t))
      (pgg-gpg-sign-region (point-min) (point-max)))
    (pop-to-buffer pgg-output-buffer)
    (unless (get-buffer-window pgg-errors-buffer)
      (split-window-vertically)
      (switch-to-buffer pgg-errors-buffer))))
--8<---------------cut here---------------end--------------->8---

Re: Problem could be encode-coding-region

[Rud1ger Sch1erz]

Katsumi Yamaoka <yamaoka@jpl.org> writes:

> Hmm, I have no problem in using PGG and XEmacs with no mule no
> file-coding.  Actually I signed this message by it.

Hm, I tried your public key 0x426F642F, but verify failed. 

I cannot import your key 0x0E2D88B9, which seems to be a more recent
one. My gpg complains about "no user-id". May be it's not signed by
yourself?

But I don't know. I'm new to this pgp stuff.

Cheers,
Rudiger

Re: Problem could be encode-coding-region

[Katsumi Yamaoka]

>>>>> In <ud574kku1.fsf@1a-software.de> Rud1ger Sch1erz wrote:

> Hm, I tried your public key 0x426F642F, but verify failed.

That's my old key.  I cannot remove it in the key servers since
the revocation certificate was in the crashed hard disk drive.

> I cannot import your key 0x0E2D88B9, which seems to be a more recent
> one. My gpg complains about "no user-id". May be it's not signed by
> yourself?

> But I don't know. I'm new to this pgp stuff.

I'm not well acquainted with PGP/GnuPG too but you will be able
to fetch the key automatically if you set the `keyserver' entry
in the ~/.gnupg/gpg.conf file properly.  Otherwise, you may
import my key <URL:http://www.jpl.org/pubkey.txt> manually.

Regards,

Re: Problem could be encode-coding-region

[Uwe Brauer]

>>>>> "Katsumi" == Katsumi Yamaoka <yamaoka@jpl.org> writes:

>>>>> In <87odqoewsr.fsf@mat.ucm.es> Uwe Brauer wrote:
   >> However and there is a big however when I use
   >> (mml-secure-message-sign-pgpmime) which uses that function

   >> Then the signing seems to be ignored, that is the tag
   >> <#secure method=pgpmime mode=sign>

   >> is inserted as it should, asked for the passwd but then nothing
   >> happens.

   Katsumi> Hmm, I have no problem in using PGG and XEmacs with no mule no
   Katsumi> file-coding.  Actually I signed this message by it.

   >> Shall I debug (mml-secure-message-sign-pgpmime)?

   Katsumi> No, you should trace `mml2015-pgg-sign' and `pgg-gpg-sign-region'.
   Katsumi> If you'd like to edebug (info "(gnus)Troubleshooting"), perform
   Katsumi> `M-x edebug-defun RET' on those functions, and type `C-c C-m s p'
   Katsumi> and then `C-u C-c C-m P' in the message buffer.


 Ok I try that, however pgg-gpg-sign-region is not interactive, while
 pgg-sign-region is and the latter works fine for me.
   Katsumi> BTW, what do you get by evaluating the following form?

   Katsumi> (progn
   Katsumi>   (require 'pgg)
   Katsumi>   (require 'pgg-gpg)
   Katsumi>   (setq pgg-default-user-id "oub")
   Katsumi>   (with-temp-buffer
   Katsumi>     (insert "testing\n")
   Katsumi>     (pgg-gpg-sign-region (point-min) (point-max))
   Katsumi>     (pop-to-buffer pgg-output-buffer)
   Katsumi>     (unless (get-buffer-window pgg-errors-buffer)
   Katsumi>       (split-window-vertically)
   Katsumi>       (switch-to-buffer pgg-errors-buffer))))

   Katsumi> (Copy it to the *scratch* buffer and type `C-j' at the end of
   Katsumi> the last line.)

Everything is fine. So I think pgg is ok, what is wrong seems to me
the  mml code. Unai san mentioned to skip the mml code part. Does
there exist a patch? Btw decrypting works fine too.

Thanks

Uwe Brauer

Re: Problem could be encode-coding-region

[Uwe Brauer]

   Katsumi> Hmm, I have no problem in using PGG and XEmacs with no mule no
   Katsumi> file-coding.  Actually I signed this message by it.

   >> Shall I debug (mml-secure-message-sign-pgpmime)?

   Katsumi> No, you should trace `mml2015-pgg-sign' and `pgg-gpg-sign-region'.
   Katsumi> If you'd like to edebug (info "(gnus)Troubleshooting"), perform
   Katsumi> `M-x edebug-defun RET' on those functions, and type `C-c C-m s p'
   Katsumi> and then `C-u C-c C-m P' in the message buffer.

I did this, and the XEmacs gets stuck in pgg-gpg-sign-region


    -    (pgg-gpg-wait-for-status process '("BEGIN_SIGNING"
         "GOOD_PASSPHRASE")))
         is fine but then 

    -  	 (process (pgg-gpg-start-process args))) it gets stuck

`pgg-gpg-use-agent' is nil, I will try now with t.
 
Uwe