Re: auto reply to html-mails
Hello, thank you for your answers - special thanks to David for the detailed considerations! Will Yardley wrote on 20.01.14: On Mon, Jan 20, 2014 at 03:29:14PM -0600, David Champion wrote: * On 20 Jan 2014, Jan-Herbert Damm wrote: i would like to send an automatic answer to html-mails sent to me (because i'm tired of writing back that i prefer plain-text). I don't like HTML mail either (for most constructions of HTML mail). However, as tired as you are of sending it, I'm sure others are tired of receiving it. So don't send it. It wasn't obvious to me that I seem to be motivated by a sense of mission (if that's a good translation of the german sendungsbewusstsein) trying to prove that I am right. configure some kind of text based browser to convert the rest (w3m or whatever), and get on with your life. HTML email is, these days, a battle that's already been lost. Sounds like a wise (slightly buddhist) approach... I do use w3m which works fine. But it made me angry that people force me to do extra keystrokes. aloha'e jan
auto reply to html-mails
Hello all, i would like to send an automatic answer to html-mails sent to me (because i'm tired of writing back that i prefer plain-text). I am aware that this is hardly an issue of mutt, but rather procmail or scripting. But i am curious how this could be approached. A goolgle-search brought up one thread concerning auto reply for vacation but that seems not really what i'm looking for. My setup for receiving mail is: fetchmail -- procmail -- spamassasin -- mutt and for sending: mutt -- msmtp -- ... A hint on how to proceed will help me. thanks jan
Re: OT making ascii tables
Hello Bernard, I read your post only recently, a bit late. Bernard Massot wrote on 07.05.13: On Mon, May 06, 2013 at 02:53:35PM +0200, Jan-Herbert Damm wrote: from this list (I can't find it anymore) I got a brilliant recipe to make plaintext-tables like so: = .TS box tab(|); cb|cb|cb|cb. If you use c instead of cb, you don't get the annoying escape characters. yes! That works!, merci beaucoup. salut jan
Re: OT making ascii tables
Hello Christian, Rado, Christian Brabandt wrote on 06.05.13: What you are seeing are ANSI Term sequences, which are usually used to color text in the terminal. It might help to explicitly set the TERM variable to dumb or vt100 or possibly set the -c parameter. You might however need to pipe the input through col -b again. 'less' has a raw mode to let ANSI pass through to the terminal, Yes, that is less -r or less -R. But that doesn't hold for other applications, like Vim (which can at least hides those characters using the ansi_escape plugin). thank you for your help! I am experimenting with groffs -c -option which is not yet working as I expected. (nroff -c works, but not with utf8). Slowly getting there. greetings jan
OT making ascii tables
Hello All, from this list (I can't find it anymore) I got a brilliant recipe to make plaintext-tables like so: = .TS box tab(|); cb|cb|cb|cb. Year | Hurricane | Deaths | Location .T l|c|c|c. 1780|Great Hurricane of 1780|27,500+|Antilles 1998|Hurricane Mitch|18,974 - 21,000|Honduras 1900|Galveston Hurricane|8,000 - 12,000|Cuba, Texas 1974|Hurricane Fifi|8,000 - 10,000|Honduras, Belize 1930|Dominican Republic Hurricane|2,000 - 8,000|Antilles, D.R. 1963|Hurricane Flora|7,186 - 8,000|Haiti, Cuba .TE then pipe this text to: tbl | nroff -Tascii | uniq output.file = lately tried: tbl | groff -k -Tutf8 | uniq output.file which works fine and I get german umlauts. But while cat will show me the output correctly vim or even less will not. I don't know where to start looking into the nature of this problem. I am shure there is a simple explanation. help appreciated greatly, sorry for being OT jan
Re: OT making ascii tables
Thank you Erik, Erik Christiansen wrote on 06.05.13: When you try :set fenc ? in vim, does it show: fileencoding=utf-8 Yes, it does and the interaction of vim and mutt is fine here is an example: cat renders: OBST/GEMUESE ┌─┬───┬──┬┐ │Bestnr. │ Produkt │ Hersteller │ Menge │ │16266│ Bananen │golden│ 2KG│ │21804│ Zwiebeln │ xxx │ 2KG│ │16300│ Zitronen │ xxx │ 6‐8ST │ │18455│ Gurken │ xxx │ 2ST│ │?│ weißkohl │ 1ST ││ └─┴───┴──┴┘ - vim or less: OBST/GEMUESE ┌─┬───┬──┬┐ │^[[1mBestnr. ^[[22m│ ^[[1mProdukt ^[[22m│ ^[[1mHersteller ^[[22m│ ^[[1mMenge ^[[22m│ │16266│ Bananen │golden│ 2KG│ │21804│ Zwiebeln │ xxx │ 2KG│ │16300│ Zitronen │ xxx │ 6‐8ST │ │18455│ Gurken │ xxx │ 2ST│ │?│ weißkohl │ 1ST ││ └─┴───┴──┴┘ could it be that the interaction with the terminal (gnome-terminal in my case) somehow has to do with this? jan If not, try: :set fileencoding=utf-8 At least that works for me when I use vim as the editor within mutt, and use e.g. å,æ,ø. (See? We're back on topic, pretty much. :-) No initial colon, if added to your ~/.vimrc - but you knew that. Erik -- Didn't the prophet Urdure vanquish the Dragon of Sluth on the plain of Gidral after three days fighting? ... And wasn't it true that the sons of Exequial beat the hosts of Myrilom? - from The Book of Om, in Carpe Jugulum, by Terry Pratchett. --
[OT] fetchmail slowing with succesive polls
Hello all, I searched web and fetchmail-faq but not the fetchmail-lists. sorry for noise but I think there are many fetchmail users here and I would appreciate a slight hint. Fetchmail polls the first email-adress very fast and slows down immensely with every one that follows. I see fetchmail working when I use the G command bound to macro index G !fetchmail -m 'procmail -d %T'\r macro pager G !fetchmail -m 'procmail -d %T'\r nothing special in my .fetchmailrc afaik. has anyone seen this behaviour? greetings jan
Re: [OT] fetchmail slowing with succesive polls
Hello Patrick, thanks! Patrick Shanahan wrote on 19.06.11: Are you only starting fetchmail from the mutt macro or do you also have it running as a daemon? I am also running fetchmail from the crontab every 20 min as: /usr/bin/fetchmail -s running fetchmail fom the macro puts me into a terminal where the fetchmail program prints dots to mark its progress. When it is polling the second and especially the following mail-adresses, mails with only say 2500 bytes take approximately 15 seconds on a 6000 mbit broadband cable connection. whereas the mails from the first poll take much less than a second. The problem is not that urgent, since i don't notice it when cron is doing the job. Yet sometimes i want to poll manually and find myself waiting and wondering. cheers, jan
strange content-types of attachments
hello all, please skip this mail and excuse the noise if you hate slightly OT posts. I recently got attachments which were really pdfs but couldn't be detected as such because they had strange content-type names: application/x-coremedia-dynamic application/force-download I wrote .mailcap-entries so that they get displayed but i couldn't find out where these names come from. Coremedia seems to be some marketing company and force-download.com is an ugly site that I couldn't make sense of. How come some people trying to attach pdfs (presumably via a webmail client) get these strange content-type entries and what might be the benefit? (My first paranoic thought is allways mafia.) jan
Re: multiple bcc entry
Hello, thank you, Monte! How can I pull in all the addresses in my .aliases file into the bcc-header-line? Monte Stevens wrote on 09.11.10: b (to open the bcc line) abcTab t (hold it down until you tag all aliases) ;EnterEnter This works well, and I don't even need the ; to operate on all tagged entries. You could also parse your alias file and pass addresses via the -b option. I've tried this, too. Also works. cheers jan
Re: attached pdf broken
Hello Simon, Simon Ruderich wrote on 17.08.10: This is just a wild guess, but is the content-type set correctly? I've had this problem once with a self-compiled version of mutt because I run it from the mutt source and thus mutt couldn't find .mime.types causing a wrong content-type (in my case just text/plain). I fixed it by linking /etc/mime.types to ~/.mime.types. This seems to have solved my problem! It is somewhat confusing since the entries from /etc/mime.types seem to have been used before... Computers never cease to amaze me. thanks Simon! jan
Re: attached pdf broken
Hello, Chip Camden wrote on 16.08.10: I'm not seeing this problem here, but it sounds like an issue with attachment encoding/decoding. Just to ask a stupid question -- are you using mutt on both ends of the send/receive? Not at first. I got complaints from others with other mailers that they can't open my pdf attachment. Then I tested with the attachments I sent (it was within my house) and with sending it to myself. Results were the same. mutt -v on my system says: [...] mutt -v here: Mutt 1.5.20hg (2009-06-26) Copyright (C) 1996-2009 Michael R. Elkins und andere. [...] System: Linux 2.6.31-21-generic (i686) ncurses: ncurses 5.7.20090803 (compiled with 5.6) libidn: 1.15 (compiled with 1.8) hcache backend: GDBM version 1.8.3. 10/15/2002 (built Nov 5 2008 02:36:47) Einstellungen bei der Compilierung: -DOMAIN +DEBUG -HOMESPOOL +USE_SETGID +USE_DOTLOCK +DL_STANDALONE +USE_FCNTL -USE_FLOCK +USE_POP +USE_IMAP +USE_SMTP -USE_SSL_OPENSSL +USE_SSL_GNUTLS +USE_SASL +USE_GSS +HAVE_GETADDRINFO +HAVE_REGCOMP -USE_GNU_REGEX +HAVE_COLOR +HAVE_START_COLOR +HAVE_TYPEAHEAD +HAVE_BKGDSET +HAVE_CURS_SET +HAVE_META +HAVE_RESIZETERM +CRYPT_BACKEND_CLASSIC_PGP +CRYPT_BACKEND_CLASSIC_SMIME -CRYPT_BACKEND_GPGME -EXACT_ADDRESS -SUN_ATTACHMENT +ENABLE_NLS -LOCALES_HACK +HAVE_WC_FUNCS +HAVE_LANGINFO_CODESET +HAVE_LANGINFO_YESEXPR +HAVE_ICONV -ICONV_NONTRANS +HAVE_LIBIDN +HAVE_GETSID +USE_HCACHE -ISPELL SENDMAIL=/usr/sbin/sendmail MAILPATH=/var/mail PKGDATADIR=/opt/mutt/share/mutt SYSCONFDIR=/opt/mutt/etc EXECSHELL=/bin/sh -MIXMASTER greetings jan Mutt 1.4.2.3i (2007-05-26) Copyright (C) 1996-2002 Michael R. Elkins and others. Mutt comes with ABSOLUTELY NO WARRANTY; for details type `mutt -vv'. Mutt is free software, and you are welcome to redistribute it under certain conditions; type `mutt -vv' for details. System: FreeBSD 8.1-STABLE (amd64) [using slang 20202] Compile options: -DOMAIN -DEBUG -HOMESPOOL +USE_SETGID +USE_DOTLOCK +DL_STANDALONE -USE_FCNTL +USE_FLOCK +USE_POP +USE_IMAP +USE_GSS +USE_SSL -USE_SASL +HAVE_REGCOMP -USE_GNU_REGEX -COMPRESSED +HAVE_COLOR -HAVE_START_COLOR -HAVE_TYPEAHEAD -HAVE_BKGDSET -HAVE_CURS_SET -HAVE_META -HAVE_RESIZETERM +HAVE_PGP -BUFFY_SIZE -EXACT_ADDRESS -SUN_ATTACHMENT +ENABLE_NLS -LOCALES_HACK +HAVE_WC_FUNCS +HAVE_LANGINFO_CODESET +HAVE_LANGINFO_YESEXPR +HAVE_ICONV -ICONV_NONTRANS +HAVE_GETSID +HAVE_GETADDRINFO -ISPELL SENDMAIL=/usr/sbin/sendmail MAILPATH=/var/mail PKGDATADIR=/usr/local/share/mutt SYSCONFDIR=/usr/local/etc EXECSHELL=/bin/sh -MIXMASTER To contact the developers, please mail to mutt-...@mutt.org. To report a bug, please use the flea(1) utility. vvv.initials 1.3.28.nr.threadcomplete rr.compressed -- Sterling (Chip) Camden| sterl...@camdensoftware.com | 2048D/3A978E4F http://camdensoftware.com | http://chipstips.com| http://chipsquips.com --
attached pdf broken
Hello All, I am trying to attach a pdf to a mail. After sending it, it is broken. A cmp with the unsent pdf-file tells me they differ in line 1 byte 9. A diff just says they differ. Copying line 1 from the original to the attached one won't help much. Could this be something mutt does? Or would my smarthost change my attachments? any ideas welcome. thanks jan
mutt and newsletters
hello all, I want to send newsletters to groups of people. I can define groups within mutt. Recipient XY is in groups A and B. A newsletter goes to groups A and B. How can i avoid that recipient XY gets the letter twice? Can mutt eliminate the double entries? Going OT: I can imagine a skipt could eliminate double entries in the list of recipients. It might be a good exercise in python, but on the other hand i am shure there is software that handles such newsletter issues already. I have been googling for open-source newsletter software but am overwhelmed with the results. Can someone help? thanks jan
Re: mutt and newsletters
Patrick Shanahan wrote on 28.07.10: A newsletter goes to groups A and B. How can i avoid that recipient XY gets the letter twice? Can mutt eliminate the double entries? use a 3rd group for the newsletter that goes to both groups cat groupA groupB |sort |uniq groupC straightforward and simple. thanks! jan
Re: unknown variables
Hello, Christian Brabandt wrote on 31.05.10: If you are using a Debian distribution, you can use sudo apt-get build-dep mutt to install all required packages, needed for compiling mutt yourself. doh! I actually did that before, long ago. installing libsasl-dev also seems to have been working though. thank you, everybody. jan
Re: unknown variables
Hello, yet another minor detail. Jan-Herbert Damm wrote on 01.06.10: thank you, everybody. Mutt is working now and I can send messages via its built-in smtp support as long as I am in mutt. But I need it to send me logfiles off the crontab (off the commandline) regularly. === j...@enit:~$ mutt -s another test m...@web.de logfile Interactive SMTP authentication not supported Could not send the message. === At home I use msmtp and a test there worked. Is that kind of workaround the only way to go? greetings jan
Re: unknown variables
Thank you, Jostein, Jostein Berntsen wrote on 31.05.10: You must compile mutt again with the --enable-pop and --enable-smtp flags. i did and now mutt prompts me that i need sasl. judging from Josteins post in the thread right before this one i will now have to recompile again --with-sasl (compiling on a very old machine... sigh) jan
Re: unknown variables
Hello, judging from Josteins post in the thread right before this one i will now have to recompile again --with-sasl another problem: ./configure can't find something concerning sasl2 === checking for sasl_client_init in -lsasl2... no configure: error: could not find libsasl2 === but === j...@enit:~/mutt-1.5.20$ sudo apt-get install libsasl2 Reading package lists... Done Building dependency tree... Done libsasl2 is already the newest version. === if this is a common issue, I am thankful for a hint. if it is too off topic, sorry for the noise. jan
where in TFM?
Dear All, i have spent over an hour on a *really* simple issue and experimented wildly but i can't find just the right spot in the manual or elsewhere. please give me a hint on tfm: i have addresses o...@web.de and t...@web.de. i have set from=o...@web.de send-hook . 'set from=o...@web.de' set use_from=yes set envelope_from=yes when replying to mails sent to my adress t...@web.de i want the from-header to show t...@web.de not o...@web.de. jan ps. some of what i've tried # reply-hook '~t t...@web.de' 'set from=...@web.de' reply-hook '.*two' 'set from=t...@web.de' # --
Re: where in TFM?
Thank you Monte and Nathan reverse_name does what i wanted. p.s. You didn't say what exactly happened when you tried the reply-hook commands you listed, but I'm guessing this sentence from the Mutt manual explains why they didn't do as you hoped: Note: reply-hooks are matched before the send-hook, regardless of the order specified in the users's configuration file. So I think the from= setting given in the send-hook . 'set from=o...@web.de' line will always override the one you are attempting to set in the reply-hook command. exactly: i could never get it to change the from header at all. not even with the my_hdr examples given in the manual. spending the afternoon off the computer working in the garden i suddenly suspected what you described above. the price for mutts configurability in my case is that i shuffle through the many options to find the relevant things... thanks again jan
Re: mutt and gpg not in tune
hello, now i have spent some additional hours on this problem which might turn out to be some idiot fault of the person behind the keyboard. unfortunately this was the case. i was stuck up on pressing s instead of p when i wanted to use gpg. it is *very* embarrassing and i apologize sincerely for the noise. i did learn a few things on the way though. jan
Re: mutt and gpg not in tune
Hello Niels, Niels den Otter wrote on 17.03.10: any hints are much needed... Just to be sure. You are opening an e-mail that is signed/encrypted with PGP and not with S/MIME? I am able to decrypt stored gpg encrypted messages. And replying to one of these messages also works fine. (I have: set pgp_replyencrypt=yes and set crypt_replysignencrypted) but to manually sign and/or encrypt causes the problems - generally i have only ever used gpg and hadn't known more than the name of S/MIME. So i don't know where i could even tell mutt to use it. Is the mime-type for the signature correct or is the PGP signature inline? I think that is correct but i am not sure i understand where to configure this. (.mailcap (?) has nothing concerning this) thanks for helping jan
Re: mutt and gpg not in tune
Hello Chuck, thanks for taking the time. Chuck Smith wrote on 17.03.10: I was looking around for a fix for you and found something interesting. Look at the Mutt User Manual in the section on PGP: http://wiki.mutt.org/?MuttGuide/UseGPG Look carefully at the this command: set pgp_encrypt_sign_command=pgpewrap gpg --passphrase-fd 0 --batch --quiet --no-verbose --textmode --output - --encrypt --sign %?a?-u %a? --armor --always-trust --encrypt-to 0xC9C40C31 -- -r %r -- %f I think that tells gpg to also encrypt to the writers pub-key, so that he is able to read the sent mails. I have encrypt-to myID in .gnupg/options to take care of that (that explains my initial confusion of why i could read files encrypted to somebody else). cheers jan
Re: mutt and gpg not in tune
dear mutt-users, Probably my question was too imprecise. Jan-Herbert Damm wrote on 16.03.10: hello all, gpg and mutt used to work fine. now mutt won't find any public keys anymore. I have run the following tests: [...] now i have spent some additional hours on this problem which might turn out to be some idiot fault of the person behind the keyboard. I think i have narrowed it down a bit. For some reason mutt doesn't seem to call gpg at all - and it might have to do with this error message (which i get if i just want to sign but not encrypt a mail) /home/jan/.smime/keys/.index: No such file or directory (errno = 2) I don't remember ever having a directory .smime/. any hints are much needed... jan beginn gpg-test--- j...@herb:~$ cat test.txt this is the content of test.txt j...@herb:~$ gpg -se -r Markus test.txt [prompted to enter my passphase] pub 2048g/0F663F56 2009-07-17 Markus Dxxx x...@ari.uni-heidelberg.de [...] j...@herb:~$ gpg -d test.txt.gpg [prompted to enter my passphase] gpg: verschlüsselt mit 2048-Bit ELG-E Schlüssel, ID 0F663F56, erzeugt 2009-07-17 Markus Dxxx x...@ari.uni-heidelberg.de gpg: verschlüsselt mit 2048-Bit ELG-E Schlüssel, ID A9EEDA47, erzeugt 2008-11-30 Jan-Herbert Damm (email-verschlüsselung) jan-h-d...@web.de this is the content of test.txt [...] end gpg-tests Thus i reckon that gpg is working, though i am confused why i can decrypt the testfile when i encrypted it with somebody elses pub-key. In my muttrc I have: ---beginn muttrc--- set pgp_timeout=1800 set pgp_replyencrypt=yes set pgp_sign_as=FA5B6661 set crypt_replyencrypt=yes set crypt_replysignencrypted source /etc/Muttrc.d/gpg.rc end muttrc--- gpg.rc is is a standard ubuntu preinstalled file which has worked before. It contains this line that seems to not be working correctly: set pgp_encrypt_sign_command=/usr/lib/mutt/pgpewrap gpg %?p?--passphrase-fd 0? --batch --quiet --no-verbose --textmode --output - --encrypt --sign %?a?-u %a? --armor --always-trust -- -r %r -- %f I am at a loss of how to go about analyzing what's going wrong here. If someone could tell me what else to try i am very grateful. jan --
Re: mutt and gpg not in tune
Hello, Dale Raby wrote on 17.03.10: I'm not an expert, but shouldn't mutt call /home/jan/.gnupg? Indeed! Unfortunately I have no idea how mutt generates the commands for encryption. In muttrc i can't find any hint to s/mime. The variable pgp_sign_command is set to a sensible gpg command. It is obviously not triggered. But Why?? jan (fighting with frustration)
mutt and gpg not in tune
hello all, gpg and mutt used to work fine. now mutt won't find any public keys anymore. I have run the following tests: beginn gpg-test--- j...@herb:~$ cat test.txt this is the content of test.txt j...@herb:~$ gpg -se -r Markus test.txt [prompted to enter my passphase] pub 2048g/0F663F56 2009-07-17 Markus Dxxx x...@ari.uni-heidelberg.de [...] j...@herb:~$ gpg -d test.txt.gpg [prompted to enter my passphase] gpg: verschlüsselt mit 2048-Bit ELG-E Schlüssel, ID 0F663F56, erzeugt 2009-07-17 Markus Dxxx x...@ari.uni-heidelberg.de gpg: verschlüsselt mit 2048-Bit ELG-E Schlüssel, ID A9EEDA47, erzeugt 2008-11-30 Jan-Herbert Damm (email-verschlüsselung) jan-h-d...@web.de this is the content of test.txt [...] end gpg-tests Thus i reckon that gpg is working, though i am confused why i can decrypt the testfile when i encrypted it with somebody elses pub-key. In my muttrc I have: ---beginn muttrc--- set pgp_timeout=1800 set pgp_replyencrypt=yes set pgp_sign_as=FA5B6661 set crypt_replyencrypt=yes set crypt_replysignencrypted source /etc/Muttrc.d/gpg.rc end muttrc--- gpg.rc is is a standard ubuntu preinstalled file which has worked before. It contains this line that seems to not be working correctly: set pgp_encrypt_sign_command=/usr/lib/mutt/pgpewrap gpg %?p?--passphrase-fd 0? --batch --quiet --no-verbose --textmode --output - --encrypt --sign %?a?-u %a? --armor --always-trust -- -r %r -- %f I am at a loss of how to go about analyzing what's going wrong here. If someone could tell me what else to try i am very grateful. jan
how to regain missing messages?
Hello, I have been away from my computer (and my mutt) for a couple of weeks in febuary. On the seventh someone flooded my account by sending large attachments. I could only fix it a week later thus missing all the mails of the muttlist during this week. How can I download those messages (or have them resent to me), so they plug into my mutt? I had a look at gmane and the archive but I am missing the brilliance to come up with an idea of how to feed the missing messages to mutt on my computer. BTW. I'm using pop3 if that makes a difference here. Thanks for any advice. jan
Re: how to regain missing messages?
Hello, Thank both of you for the suggestions. Tim Gray wrote on 02.03.10: On Mar 02, 2010 at 11:57 AM +0100, Rado S wrote: Get an mbox type archive of those mails, repair the 'From_' separator line (compare with real mbox-folders). Then copy them over to whatever you want. I used a mailman2mbox utility to convert the messed up mbox archive I downloaded from a list one time. However, after looking at the mutt-users lists, I didn't see any way to download any kind of archive. I also didn't see any way here. But I found a thread from 2002 with similar content. There one solution is a perl-script that converts yahoo-groups to mbox (still leaving the 'From_'-line problem, which might be solved by a simple sed ...) But I think I won't embark on this kind of endeavor missing only very little of the precious mutt-users content... jan
Re: replies with umlauts garbled
Hello, Christoph Kukulies wrote on 16.12.09: When I compose an email (using vi under ubuntu), and send that email, the recipients tell me that the umlauts are fine. But when I reply - in the course of follwups - to that email (that has made one turn-around), the umlauts come out garbled. Newly typed umlauts seem to be ok. Excerpt of my .muttrc: (it's a different mail account, not this one). Seemingly by coincidence there is a similar issue that arose in vim_use. There it is probably an issue of googlegroups but i thought a mention could do no harm: in the vim_use mailinglist it is this recent thread: Garbled mail on list. Was: Text replacing question Sorry for not being able to write anything substantial about the matter. cheers, jan -- Lerne, als ob du ewig lebst, liebe, als ob du morgen stirbst und tanze, als ob niemand zuschaut. -- abgeschaut bei Kyle Wheeler, Muttlist.
Re: Wanted, tool for copying/deleting old messages through a
Hello Chris, Chris G wrote on 25.09.09: Are there any obvious ways either within mutt or using some sort of external tool which would allow archiving and/or deletion of messages older than a certain date? since you are a regular poster this might be old news but are you aware of the thread named Archiving we had in Febuary? Kyle presented a hook that could possibly help you. folder-hook =mutt-users 'push tag-pattern~r3menter\ tag-prefix-condsave-message=Archive.mutt-usersenter\ untag-pattern~Aenter' Kyle explains how it works in that thread in great detail. jan
Re: Suspending Mutt-1.20. with CTRL-Z
Jan-Herbert Damm wrote on 11.07.09: I found a behaviour in my Mutt-1.20. [that should be mutt-1.5.20] CTRL-Z suspends mutt but fg doesn't seem to work. Mutt displays but pressing keys obviously still refers to the terminal below. Mutt-18.1 [sorry, mutt-1.5.18] custom ubuntu-binary doesn't show this problem. I found the problem. It is somewhat embarassing. I had copied the binary from its compiling directory under /opt to /usr/local/bin/mutt1520. Obviously that was a stupid idea, since it caused strange unforeseeable behaviour. I have only a vage idea why copying binaries is not good, but I guess I have to read up on system administration to really know why... sorry for the noise. jan
Re: signature-scripts
Hello, Thank you everybody for the suggestions! jan
signature-scripts
Hello all, Forgive me if this is slightly OT (replies off list welcome): I know that there are several people on this list (Kyle for one) that use scripts to add signatures dynamically. Learning programming makes me think signature-scripting could be a nice homework. I have googled for signature-scripts but with unsatisfying results. One idea i had: execute fortune and writing the output into a .signature file, which then gets used for mails. Are there other/smarter approaches? What other/better signature-resources could be used? Thanks. jan
Re: split display?
Tim Gray wrote on 20.07.09: On Mon 20, Jul'09 at 4:54 PM -0600, lee wrote: Maybe some sort of print statement was missing (cut off). Yeah I probably left it off or something. It should look something like this: for i in dirs: print '+l/%s' % i, or for i in dirs: print '+%s' % i, to get an output in accordance with the examples before: +dir1 +dir2 ... +dirN jan
Re: Conditionally removing signature
Hi, Concerning Roccos gpg.py script Cameron Simpson wrote on 10.07.09: Use egrep. [...] Thanks for pointing it out, Cameron. I got the regex to work but still the script outputs the group command only with none of the emails from the gpg --list-keys command. Trying to figure out why is really more of a python-learning-exercise to me now. jan
Suspending Mutt-1.20. with CTRL-Z
Greetings, I think my first post of this might have gotten lost in the shuffle. I found a behaviour in my Mutt-1.20. (on Ubuntu.Linux) that puzzles me. Again i have no idea where to start looking. CTRL-Z suspends mutt but fg doesn't seem to work. Mutt displays but pressing keys obviously still refers to the terminal below. Mutt-18.1 custom ubuntu-binary doesn't show this problem. jan
Re: Conditionally removing signature
Hi,
If you find this too OT, please accept my apolgies and ignore it. I will try to
ask somewhere pythonic.
Rocco Rutte wrote on 09.07.09:
#!/usr/bin/env python2.6
import subprocess, re
if __name__ == '__main__':
mail_re = re.compile(r'^[^]+([^]+).*')
a, p = set(), subprocess.Popen(['gpg', '--with-colons', '--list-keys'],
stdout=subprocess.PIPE)
for line in p.stdout:
if not line.startswith('uid:'): continue
m = mail_re.match(line.split(':')[9])
if not m: continue
a.add(m.groups(0)[0])
print 'group -group gpg -addr %s' % ' '.join(a)
I havn't gotten it to work yet. Could the regex be the culprit?
A test with
bash$ gpg --with-colons --list-keys | grep '^[^]+([^]+).*'
doesn't find any lines.
And the script prints the resulting strings to stdout and not to the config
file as Rocco wrote - right?
jan
--
Bitte beachten Sie auch die Rückseite dieses Schreibens!
Re: [ANNOUNCE] 1.5.20 released
Hello, Once more i found a behaviour in my Mutt-1.20. (on Ubuntu.Linux) that puzzles me. Any Pointers as to which way to look will help. CTRL-Z suspends mutt but fg doesn't seem to work. Mutt displayes but pressing keys obviously still refers to the terminal below. I hadn't noticed this before which i find surprising since i often use CRTL-Z. Mutt-18.1 custom ubuntu-binary doesn't show this problem. jan
Re: Mutt 1.20. --GPG--
Hi, * Jan-Herbert Damm wrote: GPG asks my passphrase but doesn't accept it. What gpg setup did you use? contrib/gpg.rc? It's not gpg that's asking but mutt. So maybe mutt doesn't call gpg at all or in a wrong way... Ok, it works fine now. The problem was in the direction pointed to by Rocco: I aktually wasn't telling mutt at all how to call gpg. I copied the run commands from the samples/gpg.rc into my .muttrc with only slightly detailing the path to pgpewrap. I find that building my own mutt improved my understanding a lot because i was forced to figure out the workings in detail. With the custom mutt install of ubuntu on the other hand it seems that these workings are being hidden from the user, which can lead to selling him for dumb (as the literal translation of a german figure of speech would phrase it). greetings jan I guess I'm running into the same trap as just a few days ago: the ubuntu set up with the /etc/Muttrc.d/* files being sourced from /etc/Muttrc. Whereas my /opt/mutt/etc/Muttrc does something else. - $ locate gpg.rc /etc/Muttrc.d/gpg.rc /home/jan/SRC/mutt-1.5.20hg/contrib/gpg.rc /opt/mutt/share/doc/mutt/samples/gpg.rc /usr/share/doc/mutt/examples/gpg.rc - Let me fumble with this for a while. I am sure the answer resides in there somewhere. jan, and thanks! -- Bitte beachten Sie auch die Rückseite dieses Schreibens!
Mutt 1.20. --GPG--
Hello, Testing Mutt 1.20. GPG asks my passphrase but doesn't accept it. I don't really know where to start looking for figuring this out. from config.log: --- $ ./configure -C --prefix=/opt/mutt --enable-pop --enable-imap\ --enable-smtp --enable-debug --with-gss --with-gnutls --with-sasl --with-idn --enable-hcache --- output of version: - $ /opt/mutt/bin/mutt -version Mutt 1.5.20hg (2009-06-26) [...] System: Linux 2.6.27-14-generic (i686) ncurses: ncurses 5.6.20071124 (compiled with 5.6) libidn: 1.8 (compiled with 1.8) hcache backend: GDBM version 1.8.3. 10/15/2002 (built Jun 15 2006 21:19:27) Einstellungen bei der Compilierung: -DOMAIN +DEBUG -HOMESPOOL +USE_SETGID +USE_DOTLOCK +DL_STANDALONE +USE_FCNTL -USE_FLOCK +USE_POP +USE_IMAP +USE_SMTP -USE_SSL_OPENSSL +USE_SSL_GNUTLS +USE_SASL +USE_GSS +HAVE_GETADDRINFO +HAVE_REGCOMP -USE_GNU_REGEX +HAVE_COLOR +HAVE_START_COLOR +HAVE_TYPEAHEAD +HAVE_BKGDSET +HAVE_CURS_SET +HAVE_META +HAVE_RESIZETERM +CRYPT_BACKEND_CLASSIC_PGP +CRYPT_BACKEND_CLASSIC_SMIME -CRYPT_BACKEND_GPGME -EXACT_ADDRESS -SUN_ATTACHMENT +ENABLE_NLS -LOCALES_HACK +HAVE_WC_FUNCS +HAVE_LANGINFO_CODESET +HAVE_LANGINFO_YESEXPR +HAVE_ICONV -ICONV_NONTRANS +HAVE_LIBIDN +HAVE_GETSID +USE_HCACHE -ISPELL SENDMAIL=/usr/sbin/sendmail MAILPATH=/var/mail PKGDATADIR=/opt/mutt/share/mutt SYSCONFDIR=/opt/mutt/etc EXECSHELL=/bin/sh -MIXMASTER --snip--- PS: after make install i manually changed permissions and group of /opt/mutt/bin/mutt_dotlock, because install failed at it: -- $ ls -l mutt_dotlock -rwxr-sr-x 1 jan mail 24118 2009-06-27 16:13 mutt_dotlock -- hints greatly appreciated jan
Re: Mutt 1.20. --GPG--
Rocco Rutte wrote on 02.07.09: Hi, * Jan-Herbert Damm wrote: GPG asks my passphrase but doesn't accept it. What gpg setup did you use? contrib/gpg.rc? It's not gpg that's asking but mutt. So maybe mutt doesn't call gpg at all or in a wrong way... I guess I'm running into the same trap as just a few days ago: the ubuntu set up with the /etc/Muttrc.d/* files being sourced from /etc/Muttrc. Whereas my /opt/mutt/etc/Muttrc does something else. - $ locate gpg.rc /etc/Muttrc.d/gpg.rc /home/jan/SRC/mutt-1.5.20hg/contrib/gpg.rc /opt/mutt/share/doc/mutt/samples/gpg.rc /usr/share/doc/mutt/examples/gpg.rc - Let me fumble with this for a while. I am sure the answer resides in there somewhere. jan, and thanks!
colors
Hello, My Ubuntu-custom version of Mutt-1.18. uses a different coloring scheme than the latest snapshot that i am testing now. In both cases i can't figure out where these default colors are being defined. /etc/Muttrc doesn't elaborate it and neither does my .muttrc. The Manual describes how to set colors but AFAIK not where the defaults are being defined. I would be grateful if somebody can point me to documentation on this. jan
Re: colors
Rocco Rutte wrote on 27.06.09: * Jan-Herbert Damm wrote: My Ubuntu-custom version of Mutt-1.18. uses a [...] coloring scheme Where are these default colors being defined? I'd say it'd be really strange if neither /etc/Muttrc nor .muttrc contained color statements _and_ you get colors. Sorry, I oversaw the very last line in /etc/Muttrc. It sources /usr/lib/mutt/source-muttrc.d which again invokes the sourcing of /etc/Muttrc.d/* which contains a file covering nothing but the colors. I guess that is ubuntu/debian style. And yes my home-built Mutt-1.20. is mono. jan
Mutt 1.20. notification of new mail twice
Hello, I am testing Mutt 1.20. - using mbox format It notifies me of new mail in some folder as usual. I read the mail(s), the N in the index view is gone, as usual. I leave the folder. Hitting c now notifies me of new mail in the folder i just left. But there is no new mail there, and no N in the index view, either. After leaving the folder again this does not happen again. Mutt 1.18. has no such behaviour on my system. I commented out set mark_old=no in muttrc to see wether it makes a difference. It doesn't. Has anyone made a similar experience? jan -- Bitte beachten Sie auch die Rückseite dieses Schreibens!
Re: Saving to maildir with date without sending
Hello Mutters, Yannick Delbecque wrote on 20.06.09: I use mutt to take notes. I'm trying to configure mutt to skip the sending part of the process: I wish I were more competent on mutt's internals because i do think this is an interesting question. It surprises me that those familiar with the code haven't commented on it... jan
Re: [mutt] Re: [ANNOUNCE] 1.5.20 released
Hello all, with your hints I was able to build Mutt-1.20. successfully. I chickened out by running apt-get build-dep mutt as indicated in http://wiki.mutt.org/?MuttFaq/Charset. The apt-get command installed elinks on my system and now opening html-mails via the v-command starts elinks instead of w3m like before. I wonder where this is being determined. My mailcap has entries for w3m, elinks and lynx in this order. BTW: encoding mysteries haven't all been solved but i am at peace for now. Applying this suggestion (from the mentioned link) to .vimrc was damaging. set encodingterminal charset: follows current locale set termencoding= set fileencodings= charset auto-sensing: disabled set fileencodingauto-sensed charset of current buffer It resulted in a replacement of lots of umlauts with questionmarks in innoscent files(!) jan
F1 - key-binding trapped
Hello, this .muttrc entry (in my /etc/Muttrc): macro generic,pager F1 shell-escape less /[...]/doc/mutt/manual.txt seems to be trapped by Gnome somehow. F1 (from within mutt) starts the gnome-terminal-help on my ubuntu 8.10. How to change this? jan
Re: [mutt] Re: [ANNOUNCE] 1.5.20 released
Hi, Rocco Rutte wrote on 24.06.09: The apt-get command installed elinks on my system and now opening html-mails via the v-command starts elinks instead of w3m like before. I wonder where this is being determined. My mailcap has entries for w3m, elinks and lynx in this order. See $mailcap_path. I'm almost certain mutt uses the first match in the first file in that setting... which maybe is a system-wide file in your case. If ~/.mailcap isn't the first one, use it was the first entry but didn't exist. The first existing entry was /etc/mailcap but making a file .mailcap with just the line concerning w3m does work (!) jan
Re: [mutt] Re: [ANNOUNCE] 1.5.20 released
Hi, Rocco Rutte wrote on 18.06.09: * Jan-Herbert Damm wrote: At first it can be tricky to get all the configure flags right. You can compare the output of mutt -v of both installed versions, check for differences and ask here again if you can't figure out where a flag/option comes from. I found quite a few discrepancies comparing mutt -v output. These are the compiler-flags used in Mutt-1.18.-ubuntu-build but not used in my Mutt-1.20.-home-build and of which i don't know what configure-option they are triggered by: +HAVE_GETADDRINFO +COMPRESSED +HAVE_LANGINFO_CODESET jan
Re: [mutt] Re: [ANNOUNCE] 1.5.20 released
Hello, Thanks for explanations! I tried another build and found that my switches revealed more problems that i want to gnaw on for a while. +HAVE_LANGINFO_CODESET Literally, this means that your system has the langinfo.h header and the nl_langinfo(CODESET) function is available as well. This allows mutt to automatically detect the default charset the system wants to use---without this ability, mutt is forced to assume some default charset (namely ISO-8859-1). This is particularly disturbing because my Mutt-1.20. hat great problems displaying numerous characters (such as the lines showing the threads and german umlauts etc.) But :set ?charset renders utf-8. Since i also have strange encoding behaviour in Mutt-1.18. (i may not use german umlauts in muttrc else running into trouble) i hope to find a common root to these issues. I will report back when/if i find out more. greetings jan
Re: [mutt] Re: [ANNOUNCE] 1.5.20 released
Hello, Adam Wellings wrote on 15.06.09: Mutt 1.5.20 is out at the usual locations. It's on Sourceforge Thanks. Now i found it. I would like to try it out. To be safe i want to keep my working Mutt 1.5.18 in /usr/bin and install Mutt 1.5.20 in /usr/local/bin. ./configure (without any options) and make went well. Can i safely sudo make install? jan
Re: [mutt] Re: [ANNOUNCE] 1.5.20 released
Hello, Rocco Rutte wrote on 16.06.09: Jan-Herbert Damm wrote: I would like to try Mutt 1.5.20 I usually choose ~/opt/mutt/ so that I can easily install without sudo and deinstall with rm. I am glad i followed this advice. Thanks Rocco! It turns out i have to resolve some issues. Encoding and gpg don't work as expected. I will do some investigation. It is my first manual install of mutt. The old version is apt-get installed. jan
Re: [ANNOUNCE] 1.5.20 released
Hello, Mutt 1.5.20 is out at the usual locations. please, where are these locations? i couldn't find this version on mutt.org. maybe just cvs? jan
search in index (mbox)
Hello, i am confused about mutt's searching behaviour. if i use the / (slash)- command in the index of my inbox i can search for a name of somebody who sent me a mail. But the same command doesn't work in my sent box. There the search only seems to match strings in the subject line. i am sure there is a simple explanation. help appreciated. jan
Re: search in index (mbox)
Hi, Kyle Wheeler wrote on 09.06.09: On Tuesday, June 9 at 08:51 PM, quoth Jan-Herbert Damm: i am confused about mutt's searching behaviour. snip ... The *default* $simple_search is: ~f %s | ~s %s snip Presumably, in your sent box, you'd want it to also search the recipients (since, like I said, all the messages in your sent box were sent by YOU), so you could change it to being: ~f %s | ~s %s | ~C %s Check the manual's (or the muttrc man page's) description of simple patterns for more details on what those letters mean. checked it, changed it, happy now! it is beautifully logical but tricky to figure out on your own. thank you, Kyle.
Re: doubled messages by copying
Hello, You just want to move messages? Then use s (save-message), which copies the message to whatever mailbox you specify, then marks the copy in the current mailbox as deleted. Thank you Monte, Christian, Gary and Noah. Embarrassingly i wasn't aware of save-message. I realize that the german explanation of its function (nachricht in datei speichern = save to a file) confused me into thinking save-message would produce some file outside of mutts mbox format. Monte's Macro is a nice find for me. I will keep my eyes peeled for the original problem of doubled messages anyhow. jan
doubled messages by copying
Dear List, after using the C-command to copy messages from my inbox to some other folder i *sometimes* find these messages twice in the other folder. why? jan
Re: doubled messages by copying
Hello, after using the C-command to copy messages from my inbox to some other folder i *sometimes* find these messages twice in the other folder. Joost Kremers wrote on 11.05.09: Perhaps because C doesn't move to the next message after copying, so that you sometimes inadvertently copy the same message twice? I doubt that, because i delete the original in the inbox right afterwards. Sadly i cannot reproduce the strange behaviour on purpose. I want to *move* read messages to another mailbox (mbox format) for archiving. I'm using C plus d because I haven't yet found a smarter way to move mails among mboxes. jan
what is the benefit of imap?
Hello, i have a meta-question. imap-issues are being discussed on this list quite frequently. i wonder what the benefit of imap generally is? if you have access to broadband flatrate internet (as many people in agglomerations here in germany do) i can't think of the benefit of managing my mails on a remote server. if you consider this question silly or off-topic please give me a hint where i could find an answer. thanks. jan
folder-hook command behaves strangely
Hello, I try to tell my muttrc to use different attribute strings in replies in different mailboxes (mbox-format): # begin .muttrc-entry: # attribute string set attribution=%n schrieb vor kurzem: # # attribute string for mbox mutt-users: folder-hook mutt-users set attribution=%n wrote on %d: # end When I change into mutt-users folder I get an error-message telling me that wrote is an unknown variable. BUT using # folder-hook mutt-users set attribution=%n: # instead, works! I read TFM and reckoned it might have to do with not defining a default. (and in fact coming back to the inbox the attribute string is still the one defined for the mutt-users box.) But entering: # folder-hook . set attribution=%n schrieb vor kurzem # did not help. Now mutt complains about schrieb as unknown variable. EVEN when i open the mutt-users box where I expect it to complain about wrote. If I now disable the original set attribution=%n schrieb vor kurzem-entry i get the expected complaint about wrote as invalid variable. What am i missing out on? jan HINT: I posted about my encoding problems before. Maybe that problem is related because if I put a german UMLAUT into the attribution-string for example set attribution=%n schieb kürzlich I have crumbled up umlauts in my reply mails!
Re: folder-hook command behaves strangely
Christian Brabandt wrote on 19.02.09: I have made the experience, that it usually works better, if you quote the command using double or single quotation marks. So this works for me: folder-hook mutt-users 'set attribution=%n wrote on %d:' Wow! That did it. Thanks, Chrisitian -- Was länge währt wird endlich gut. Was lange gärt wird endlich Wut.
Re: newbie install
hello, i was surprised to read that mutt can handle smtp-sending on its own now. because i also had to fiddle with msmtp (which is rather simple). So just install mutt on it's own - there don't seem to be any dependencies according to synaptic. I don't know about synaptic. But using apt-get install mutt exim4 was installed automatically with mutt on my sytem. i was using exim after quite an effort to configure it for weeks until i noticed that there were subtle authentifaction issues and *some* of my outgoing mail was refused by the smtp-server with my exim4 seemingly freezing the bounces so that i never got a hint of there being a problem. I switched to msmtp and now it works. everything would have been a lot simpler had i known about mutts smtp feature. what should I learn from this? better building mutt from the source? jan
Re: Clarification on tag-prefix-cond vs. tag-prefix
Hello, And Simple patterns seem to be specifically associated with certain keywords replacing ~,%,= expressions. Keywords? Replacing? No, the ~ or % or = don't get *replaced*. We must be talking past each other here. indeed, I was being unprecise. The manual (Chapter Advanced Usage) is saying this about simple *searches* not simple *patterns*. I still find it confusing if not contradictory, because the searching by keywords (= not valid patterns) is explained directly below the heading Simple Patterns. Thus i conclude a seemingly invalid pattern must be the announced simple pattern as in simplified pattern. --snip from the manual-- 2.2. Simple Patterns Mutt supports two versions of so called ``simple searches'' which are issued if the query entered for searching, limiting and similar operations does not seem to be a valid pattern (i.e. it does not contain one of these characters: ``~'', ``='' or ``%''). [...] Table 4.2. Simple search keywords ++ | Keyword | Pattern modifier | |-+--| | all | ~A | |-+--| --snip--- but i guess i am splitting hairs as we say in german. Yet i stumble on things like this and start in the wrong tracks... jan
Re: Clarification on tag-prefix-cond vs. tag-prefix
Kyle Wheeler: ... However, because I use tag-prefix-cond, when there aren't any tagged messages (i.e. the pattern ~r 3m didn't match anything), mutt will stop processing that hook and none of the rest of it will happen. Does that make sense? yes it does! thank you so much for posting this help on archiving. i learned a lot from reviewing it very slowly. if you don't mind, can you explain the pattern ~r 3m equally well? i know i could figure it out by RTFM but it would take forever. Jan
Re: encoding in replies
Derek Martin wrote: What's the value of send_charset? It is: send_charset=us-ascii:iso-8859-1:utf-8 Changing it manually to utf-8 only didn't change anything. Here's a clue: Was lange g?rt wird endlich Wut. The third word in the second line contains a character which is not present in the character set used to encode this message, or it contains one which can not be translated to UTF-8. This is bewildering. The first line of the quote is what I manually added to the signature which is taken from my .signature file. I think i wrote that file not with vim but with gedit. To be sure i deleted it and wrote it again in vim and ran another test. The problem was gone. Then i reopened the .signature file in gedit, changed something, saved it, tested again but the problem did not reapear. So i havn't been able to reproduce the problem, still don't have a clou what is really happening but for the moment there is no more problem. It is unsatisfying... I will do some more thinking but meanwhile declare this issue as partly solved. Thanks again. jan
encoding in replies
Hello everybody, I suspect I have a similar encoding problem like the one discussed today. (in the thread: quoting urls in replies. It might be trivial and i apologize if it is. I did try to RTFM and i have been pondering this with growing desperation: The problem in short: When I write mails with german umlauts to myself, I get them back with the umlauts displayed as questionmarks. I use vim to write my mails (i tell mutt to use it via an entry export EDITOR=/usr/bin/vim in .bashrc) when I write the mail umlauts are displayed nicely. but if I postpone the message and resume writing afterwords the umlauts are allready gone. (I guess that has to to with mutt opening a new tmp-file). The output of :set all in vim shows among other things encoding=utf-8 The output of locale in bash: LANG=de_DE.UTF-8 LC_CTYPE=de_DE.UTF-8 LC_NUMERIC=de_DE.UTF-8 LC_TIME=de_DE.UTF-8 LC_COLLATE=de_DE.UTF-8 LC_MONETARY=de_DE.UTF-8 LC_MESSAGES=de_DE.UTF-8 LC_PAPER=de_DE.UTF-8 LC_NAME=de_DE.UTF-8 LC_ADDRESS=de_DE.UTF-8 LC_TELEPHONE=de_DE.UTF-8 LC_MEASUREMENT=de_DE.UTF-8 LC_IDENTIFICATION=de_DE.UTF-8 LC_ALL= No $charset ist set in .muttrc And one more really strange thing: All of this happens on a custom gnome desktop environment. If I switch to a Terminal without X. Bash displays all the capitalized umlauts and the small a-umlaut but not the small u-umlaut and the small o-umlaut, it simply presents a space instead (whereas vim or mutt show some graphical characters that I have never seen before for these and only for these umlauts) I am aware that I might have explained all this badly. That is because i don't understand the problem. If somebody can tell me the right RTFMs to read i will happily do that homework first. Thanks jan -- Was lange g?rt wird endlich Wut.
