Steve Schmerler | 26 Jul 15:43 2016

crypt-hook: truncated list of gpg recipients passed to pgpewrap

Hi all

The following is related to [1], i.e. use multiple crypt-hook commands.
I'm using something like this[*]

    crypt-hook mailing.list <at> EX9B197029581FAC
    crypt-hook mailing.list <at> 4C95T7E7F627637D
    crypt-hook mailing.list <at> 74464897317CDA88
    crypt-hook mailing.list <at> FECEC90758BB1D95
    crypt-hook mailing.list <at> 866BAB720X7E9L98

Running this with mutt -d2, I see in the debug log that the list of
recipients is passed to pgpewrap, but truncated as follows (lines

[2016-07-26 14:49:12] mutt_pgp_command: /usr/lib/mutt/pgpewrap gpg2 
    --batch --quiet --no-verbose --textmode --output - --encrypt
    --encrypt-to XXXXXXXX --sign -u YYYYYYYYY --armor --always-trust 
    -- -r 0xEX9B197029581FAC 0x4C95T7E7F627637D 0x74464897317CDA
    -- /tmp/mutt-foo-1000-aaa-bbbbbbbbbbbbbb

i.e. mutt 

* stops at the third[**] key 74464897317CDA88 (length 16) and
* passes a truncated version 74464897317CDA (length 14) to pgpewrap

which then gives me a "gpg: skipped: Invalid user ID". The related pgp
setting is

    set pgp_encrypt_sign_command="/usr/lib/mutt/pgpewrap gpg2 
(Continue reading)

Gregor Zattler | 25 Jul 23:44 2016

how to escape complex pattern for folder-hook?

Dear mutt users,

I get status emails in some mail folder which I occasionally
delete with delete-pattern (bound to "D") and this pattern: 

~s"Cron <grfz <at> len> nice ionice -c 3 ~/bin/my-notmuch-new-and-tagging"   ~b^"No new mail." !~b"^Processed"

I would like to do this via a folder-hook like this:

folder-hook +root 'push <delete-pattern>~s"Cron <grfz <at> len> nice ionice -c 3
~/bin/my-notmuch-new-and-tagging"   ~b^"No new mail." !~b"^Processed"'

but I don't know how to quote or escape the second argument.  I
always get error messages "too many arguments".

Can anyone please explain what I'm doung wrong?

Ciao, Gregor

 -... --- .-. . -.. ..--.. ...-.-

John Long | 18 Jul 08:05 2016

[SPAM?] [OT] How to move mail from Linux to Android?

So I find myself wanting to move several gig of emails from Linux to
Android. The bulk of it is in MH format although I have a few mutt instances
using maildir.

I figure there is probably a ton of email savvy on this list, can anybody
give me an idea how I can accomplish this?

I haven't picked an email client on Android yet. I have tried a few and am
not really satisfied with anything so far so I'm open to something if it
would help with the solution.

Any suggestions guys? Thanks.



ASCII ribbon campaign ( ) Powered by Lemote Fuloong
 against HTML e-mail   X  Loongson MIPS and OpenBSD
   and proprietary    / \
     attachments     /   \  Code Blue or Go Home!
 Encrypted email preferred  PGP Key 2048R/DA65BC04 

Jeremy Volkening | 15 Jul 22:00 2016

new mail indicator in browser view


I've been giving mutt a spin for the past few days. I like it quite a 
bit except for one issue I have been unable to solve: how to get the 
mailbox browser to properly display new message status.

Platform: Debian jessie running NeoMutt (latest from git but have tried 
mutt from official repo with same result)

Use case: I have two IMAP accounts I'm accessing. I have tried using 
mutt's built-in IMAP functionality as well as using mbsync to sync to 
local maildir and mutt to access that. Both ways work fine except for 
the issue described below.

Expected behavior: When I switch to mailbox browser view (<c><?><Tab>) 
or start in browser view ("mutt -y") I expect to see a new message 
indicator (for mbsync/maildir setup) or new message count (for IMAP 
setup) for each folder defined using "mailboxes". The status should be 
updated whenever mutt checks for new mail.

Observed behavior: Most of the time the indicators are empty or zero 
regardless of new mail status. Using IMAP directly, the new message 
count will be correct for a given mailbox after viewing it and returning 
to the browser view, but will be "forgotten" after two or more other 
mailboxes are visited. Using the mbsync/maildir setup, the only time the 
"N" status indicator is set is when there is new mail present when 
starting as "mutt -y" - it never appears in response to new mail 
received after mutt is started, regardless of what I try. I should note 
that new mail behavior from within index view works fine - the message 
appears (along with a note in the status bar) shortly after the mail 
(Continue reading)

Steve Schmerler | 14 Jul 15:40 2016

gpg groups or encrypt to multiple recipients (mailing list)


I'd like to ask what approach people have to send encrypted mail to
multiple mailing list recipients in mailing list via a mailing list
address (say mailing.list <at>

Sadly, my company's policy is to not have a mailing list key and let the
mail server re-encrypt the mail. I know it sucks, but I need to send
encrypted mail to the list anyway. I'm aware of [1,2] and used the
approach in [2]:

    # .gnupg/gpg.conf
    group 0x12345678 = key1 key2 .. keyN

    # muttrc 
    crypt-hook mailing.list <at> 0x12345678

This is OK for me, apart from making up a hex key -- this is really
awkward (I literally used 0x12345678).

Are there other workarounds in use? The solution above can't be the only

Apart from the gpg group thing, I would have been fine with the option
of interactively selecting the keys to encrypt to in the PGP menu,
independently from the actual recipient's address. Is this possible?

(Continue reading)

Matthias Apitz | 13 Jul 09:24 2016

[SPAM?] validation of e-mail addrs


Is there some function in mutt's source which validates e-mail addrs?


Matthias Apitz, ✉ guru <at>, ⌂  ☎ +49-176-38902045
"Wer übersieht, dass wir uns den anderen weggenommen haben und sie uns wiederhaben wollen,
kann von den Kämpfen der letzten Tage keinen verstehen. Und kann natürlich auch keinen
dieser Kämpfe bestehen." Hermann Kant in jW 1.10.1989

Will Yardley | 12 Jul 19:06 2016

[SPAM?] updated trash folder patch?

Does anyone have versions of Cedric Duval's trash folder patch (or
similar) updated for >= 1.6.1 yet?


Xu Wang | 12 Jul 07:47 2016

list-reply enabled for a non-mailing list

Hello to all,

I have some messages that are not from a mailing list (I am sure that
mutt also knows it is not list because there is no "L" flag). But
<list-reply> is still enabled. How is this possible?

Should there be any case where there is no L flag but <list-reply> is
enabled? How do I disable it so I don't make mistake?

Kind regards,


Andre Majorel | 8 Jul 12:14 2016

<limit> / l ~b has become extremely slow

After upgrading from Mutt 1.5.21 to 1.5.23, trying to <limit> on
the contents of the bodies of emails ("l ~b some-string") has
become shockingly slow. It goes through about 10 messages per
second, taking minutes to filter a medium-size mailbox.

Granted, this is on a ten-year-old Athlon X2 but, as I recall,
Mutt 1.5.21 used to do the same operation much faster on the
same hardware. And I don't mean 20% faster, more like an order
of magnitude faster.

Filtering on headers ("l ~h some-string") is still as fast as it
used to be.

Anyone knows what happened and what can be done about it ?
Thanks in advance.

Top ten of the output of ltrace -c :

% time     seconds  usecs/call     calls      function
------ ----------- ----------- --------- --------------------
 15.52  213.163929         152   1396375 strchr
 12.39  170.096294         152   1115322 strcmp
 11.40  156.587240         152   1026538 __ctype_b_loc
 10.35  142.148496         152    930708 strlen
 10.14  139.233797         154    900301 fgets
  9.74  133.693317         152    877458 strspn
  6.33   86.957602     2350205        37 wgetch
  5.14   70.518078         153    460849 fgetc
  4.04   55.524286         546    101517 __realpath_chk
  1.57   21.558909         152    141393 memcpy
(Continue reading)

Niklas Reppel | 8 Jul 09:44 2016

Does mutt cache passwords or sessions or anything ?


I'm using mutt with built-in imaps/smtp and i've recently experienced 
something strange.

I've changed my mail password using my mail provider's web page.

With other clients (the provider's webmail client and Thunderbird, and 
Dekko on Ubuntu Touch) the behaviour is as
expected (ONLY the new password works).

But with mutt, the old password AND the new password work. With the new 
password (!), the login takes way longer.

Even stranger, i've changed the password multiple times, and the 
"intermediate" passwords got invalidated as expected,
but the oldest one still works ...

Could there be some kind of password/session/whatever cache ? Or should 
i worry ?


Ken Moffat | 7 Jul 23:19 2016

mbox problem on new system


I'm an inveterate mbox user (that goes back to the days when ext2/3
were reputed to not like vast numbers of files in a directory).

I'm now building a new machine to replace the current antiquated
box I use as my home server.  The current server (x86_64 linux) was
last rebuilt in September last year with coreutils-8.24, gcc-5.2 and
for a few weeks it has been running mutt-1.6.1.  The new base system
is from a few months ago, with coreutils-8.25 and gcc-5.3 (the base
is linuxfromscratch - old is 7.8, new is 7.9, and everything after
is from development BLFS - 'beyond linuxfromscratch').

In ~/.muttrc I have

set folder="~/Mail"

and ~/Mail is a symlink to ~/mailboxes/Jul/ for this month's mails
(and also symlinks to some other mboxes).

As part of bringing up the new machine I have copied /home from the
old one (will need to repeat that when I'm ready for the real
changeover), to sort out the many problems which I expect to
encounter.  But I did not expect that mutt would now cause me

When I open mutt on the old machine, I get a list of all the
mailboxes for the current month.  But on the new machine I get

/home/ken/Mail is not a mailbox.
(Continue reading)