gpg decryption failure
71 views
Skip to first unread message
Werner LEMBERG
Mar 1, 2017, 1:23:43 AM3/1/17
to mew...@googlegroups.com, ka...@iijlab.net
[mew from git 2016-Nov-24]
[emacs from git 2016-Dez-03]
[gpg 2.0.24]
Kazu et al.,
I get
<body> PGP NOT decrypted for some reasons
after pressing C-c C-z.
If I activate mew-debug, I see the following message:
<PGP filter1>
gpg: NOTE: old default options file `/home/wl/.gnupg/options' ignored
gpg: malformed CRC
gpg: quoted printable character in armor - probably a buggy MTA has been used
Attached is the e-mail in question (compressed). It seems that gpg is
applied to early, before properly decoding quoted printable characters
– indeed, the last two lines of the encrypted message are
kJX7e04u5R+bF80+4da4YNezltjzkJFAK8xyRsp67ry2Yg3blJ6vydSSrNcJyg==
=z+U0
Note that I *can* decrypt the message if I let mew display the body,
write the region to a buffer, and calling gpg manually.
In other words, the bug is not a show-stopper, but very annoying :-)
Werner
Werner LEMBERG
Mar 1, 2017, 1:43:55 AM3/1/17
to mew...@googlegroups.com, ka...@iijlab.net
> It seems that gpg is applied to early, before properly decoding
> quoted printable characters – indeed, the last two lines of the
> encrypted message are
>
> kJX7e04u5R+bF80+4da4YNezltjzkJFAK8xyRsp67ry2Yg3blJ6vydSSrNcJyg==
> =z+U0
After decoding, of course. Before it looks like
> quoted printable characters – indeed, the last two lines of the
> encrypted message are
>
> kJX7e04u5R+bF80+4da4YNezltjzkJFAK8xyRsp67ry2Yg3blJ6vydSSrNcJyg==
> =z+U0
kJX7e04u5R+bF80+4da4YNezltjzkJFAK8xyRsp67ry2Yg3blJ6vydSSrNcJyg=3D=3D
=3Dz+U0
Werner
Kazu Yamamoto
Mar 3, 2017, 12:36:42 AM3/3/17
to mew...@googlegroups.com
Hi Werner,
> Attached is the e-mail in question (compressed). It seems that gpg is
> applied to early, before properly decoding quoted printable characters
> – indeed, the last two lines of the encrypted message are
Yes. The current C-cC-z is designed to target non-MIME (PGP) messages.
Even PGP messages are contained in MIME, content-transfer-encoding is
not necessary thanks to ASCII Armor. But the message is encoded with
quoted-printable.
What we can do is to implement C-uC-cC-z which passes the content of
the Message buffer to GPG.
--Kazu
> Attached is the e-mail in question (compressed). It seems that gpg is
> applied to early, before properly decoding quoted printable characters
> – indeed, the last two lines of the encrypted message are
Even PGP messages are contained in MIME, content-transfer-encoding is
not necessary thanks to ASCII Armor. But the message is encoded with
quoted-printable.
What we can do is to implement C-uC-cC-z which passes the content of
the Message buffer to GPG.
--Kazu
Werner LEMBERG
Mar 3, 2017, 2:13:50 AM3/3/17
to ka...@iij.ad.jp, mew...@googlegroups.com
Hello Kazu!
>> Attached is the e-mail in question (compressed). It seems that gpg
>> is applied to early, before properly decoding quoted printable
>> characters – indeed, the last two lines of the encrypted message
>> are
>
> Yes. The current C-cC-z is designed to target non-MIME (PGP)
> messages. Even PGP messages are contained in MIME,
> content-transfer-encoding is not necessary thanks to ASCII Armor.
> But the message is encoded with quoted-printable.
> What we can do is to implement C-uC-cC-z which passes the content of
> the Message buffer to GPG.
C-uC-cC-z and the current C-cC-z.
Thanks for taking care of the issue
Werner
Reply all
Reply to author
Forward
0 new messages