Re: message duplicated if mailbox written while mutt is opening it
On Mon, Apr 16, 2001 at 12:24:34AM -0500, Tim Legant wrote: Yes and no. Mutt has to be compiled with the correct options to do it. Assuming you're on a system where qmail uses flock (not HP/UX or Solaris), configure mutt with --enable-flock. Then they're both talking the same language. Thank you for your help. Just as a follow-up in case anyone else has been reading this thread, I just peeked into INSTALL and saw this: Linux On recent Linux systems, flock() and fcntl() locks don't mix. If you use the --enable-flock switch on such systems, be sure to give the --disable-fcntl argument as well. So I will be using those two options. Thanks again.
Re: message duplicated if mailbox written while mutt is opening it
On Sun, Apr 15, 2001 at 02:11:46PM -0500, Tim Legant wrote: It's a maildir feature that this can't ever happen. It's a mbox "feature" that you have to use locking to prevent it from happening. Lol.. yep. This is occurring with regular mailboxes. I would have expected Mutt to do that locking though... no?
safety of externally appending to a mailbox
If I have a process which writes to a mailbox which Mutt already has open, is there any reason for this to be considered unsafe? For example I have some 'fetchpop' processes that write to several mailboxes; if I have one of them open Mutt doesn't appear to have any problem seeing the new message(s). I have never had any issues with this but I want to make sure there's no special way the append has to be done in order to do it cleanly. I ask because currently I have a hole poked into a firewall to allow smtp to come in from the upstream mail exchanger; I was considering a different approach which would let me close that smtp port: scp the remote file to a temporary name on the local box, then cat it to the "real" mailbox file. Any obvious problems with that approach? Thanks.
Re: Playing a new mail wav
On Sun, Apr 08, 2001 at 04:09:55PM -0400, Phil Sexton wrote: Would any of you know how to play a new mail wav file upon arival of new mail? I would like to: I'm using gkrellm for this (and other things): http://newweb.wt.net/~billw/ It would really be neat if I could play different wavs for different mailboxes. Afaik, gkrellm will not do this, though. :-\
Message-ID re-writing
Hello, using Mutt-1.2.5 here. I'm re-writing my Message-ID headers using qmail-queue, but while experimenting with that header for the benefit of a sendmail-using friend, I came across this phenomenon. If I use my_hdr to add a Message-ID header to the stuff I am sending, Mutt will _still_ put its own into the headers and I will end up with two of them in the outgoing message. However, if I use the "E" command before sending the message.. and manually add my own Message-ID header, Mutt will _not_ generate the extra header and it works as I would like it to. Anyone know why my_hdr doesn't work as I expected it to here? Please copy me on replies, not currently subscribed. Thanks!
Re: Message-ID re-writing
On Fri, Mar 23, 2001 at 12:59:52PM +0100, Thomas Roessler wrote: I'm re-writing my Message-ID headers using qmail-queue, Why are you doing such things? You'll ruin the possibility to get proper threading with message copies mutt may write to folders while sending. Actually, you're right. But that's because my "implementation" is broken, and I should have realized that. However, if I were to write the Message-ID *before* creating the message (which is what I'm trying to do with my friend's configuration using my_hdr), this shouldn't be a problem. FYI, the reason I'm doing such things :) is because I don't particularly like the Message-IDs that Mutt writes. Call me silly, but the second half of the ID -- since it uses an incrementing letter (A.. B.. C.. and so on) makes it easy for someone to tell how many messages you have sent in between the first one he sees, and the next one (assuming you haven't started a new instance of Mutt in the meantime, but the pid is right there too so that can be determined). Anyone know why my_hdr doesn't work as I expected it to here? I suppose it's a bug. I'll look at it. Thanks. (But don't expect a fix for the stable branch.) I don't.
Re: Message-ID re-writing
On Fri, Mar 23, 2001 at 05:44:12PM +0100, Jerome De Greef wrote: RFC 822: [...] This identifier is intended to be machine readable and not necessarily meaningful to humans. [...] That says it all ;) No, it doesn't. The problem is that it currently _IS_ meaningful to humans. If it were not meaningful to humans, in its current implementation in Mutt, I would be happy.
Re: Message-ID re-writing
On Fri, Mar 23, 2001 at 06:39:48PM +0100, Jerome De Greef wrote: (hum, I think I don't get the point on this one) I noticed. ;) The point is, it gives away information that maybe some people don't want to give away. Using a randomly-rotated 3- or 4-character string would be far better than using an incrementing single character. Regardless of what types of kind of sarcasm about paranoia you can throw at this, I think there are enough people who would appreciate such an improvement -- especially one which has no detractors other than folks who just are not willing to change things. Many people (besides myself) rewrite Message-IDs for exactly this reason. This change would make that no longer necessary. Do you have people looking at your User-Agent string? Yes/no? How do you know? Well if you aren't concerned then you can leave the string in, or take it out; that's why this option exists. Someone somewhere didn't like it, and now we have the user_agent option. Did you disparage that idea at the time too? ;)
Message-ID re-writing
Hello, using Mutt-1.2.5. I'm re-writing my Message-ID headers using qmail-queue, but while experimenting with that header for the benefit of a sendmail-using friend, I came across this phenomenon. If I use my_hdr to add a Message-ID header to the stuff I am sending, Mutt will _still_ put its own into the headers and I will end up with two of them in the outgoing message. However, if I use the "E" command before sending the message.. and manually add my own Message-ID header, Mutt will _not_ generate the extra header and it works as I would like it to. Anyone know why my_hdr doesn't work as I expected it to here? Thanks.
Re: display sent messages
On Mon, Jan 01, 2001 at 05:40:47PM +0100, Jens Nestel wrote: I just sent an e-mail an I want to view it again, but how can I display the contents of sent mail again ? From man muttrc: record Type: path Default: "" This specifies the file into which your outgoing messages should be appended. (This is meant as the primary method for saving a copy of your messages, but another way to do this is using the "my_hdr" command to create a Bcc: field with your email address in it.) The value of record is overridden by the "force_name" and "save_name" variables, and the "fcc-hook" command. So, make sure you've set the "record" variable in your .muttrc to a mailbox name into which you wish to store copies of your sent mail. HTH.
Re: my_hdr stripped when recalling a postponed message
On Tue, Sep 12, 2000 at 06:36:11AM -0400, David T-G wrote: At the compose screen, I chose to 'e'dit my message again, and the R-P: header was gone. Argh!! Are you really supposed to be able to define R-P:: yourself, perhaps? I don't know of any reason why you shouldn't be able to. :) I've been doing it for a long time now, using qmail as my MTA, it allows me to correctly set the Return-Path and effectively _not_ give away a local userid in the header. For instance if my local username were "jimb" and I didn't want this to appear anywhere in the headers, setting Return-Path has been the best way to handle this. With qmail you could set a QMAILUSER env var to do the same thing but if you need to change it on the fly from a running app that doesn't help. :P Thanks for your response! :)
Fw: my_hdr stripped when recalling a postponed message
Anyone have any ideas on this? (I didn't get any responses.) - Forwarded message from Jim Breton [EMAIL PROTECTED] - From: Jim Breton [EMAIL PROTECTED] Date: Tue, 29 Aug 2000 20:27:23 + To: [EMAIL PROTECTED] Subject: my_hdr stripped when recalling a postponed message Is this a bug, or intended behavior? 1) compose a message with a user-defined Return-Path my_hdr 2) postpone it 3) recall it 4) note that the user-defined Return-Path is gone The Return-Path line is saved in the "postponed" file when you postpone it; but, when you read it back in for editing, it gets wiped out. Thanks. - End forwarded message -
my_hdr stripped when recalling a postponed message
Is this a bug, or intended behavior? 1) compose a message with a user-defined Return-Path my_hdr 2) postpone it 3) recall it 4) note that the user-defined Return-Path is gone The Return-Path line is saved in the "postponed" file when you postpone it; but, when you read it back in for editing, it gets wiped out. (I'm not on the list, please include me in any responses.) Thanks!
Re: Different mail accounts
On Sun, Feb 06, 2000 at 03:09:43PM +, Edmund GRIMLEY EVANS wrote: Wouter Hanegraaff [EMAIL PROTECTED]: and a bit below there is a header like this Sender: [EMAIL PROTECTED] However, I have multiple accounts because I don't want certain people and mailing lists to spam me on my real mail address. And if my real address is still somewhere in the headers, I might as well have just one account. The From_ line and the Sender: line are presumably added by your MTA (sendmail/exim/...). I can't speak about Exim or Sendmail since I haven't tested it (would be a piece of cake to do so though), but the "Sender" line in qmail's case is generated solely by the MUA, the MTA does not put one in. Pine, for example, puts an "X-Sender" or "Sender" (configurable) into the headers if the "From" line does not match the user who executed Pine (which happens when you use "Roles," the built-in multiple account support. This, as it happens, is one of the major reasons I switched to Mutt as my MUA. Pine really pisses me off (pardon the expression) when it throws those headers in there when my whole point of changing my From line to begin with is to keep my "good" addresses out of my mail! While qmail does not put in [X-]Sender lines, it does add a "Return-Path" header which indicates the userid who executed the "sendmail" wrapper which Mutt uses. But in this case, you can override this with a "my_hdr Return-Path: " command. Anyway to my point. :) Test with your respective MTAs, if you are sure it is the source of [X-]Sender lines you may be able to override them by supplying your own from Mutt. Worth a shot anyway.
index is reversed on :source .muttrc
I'm wondering if this is a bug or something. I am sorting my index by threads/reverse-date-received, and any time I source my .muttrc the order of my index (message list) is reversed. If I re-read the mailbox (c ! for example) the listing returns to normal. Is it supposed to do this?
Re: mutt/eterm
How about setting: export TERM=xterm prior to running Mutt? On Fri, Feb 04, 2000 at 06:19:14AM -0500, mike irwin wrote: can anyone tell me if, and how to set up mutt to be run in eterm? i was just using xterm before, but now i would like to use eterm with trans on (if possible). when i try to run mutt in eterm it gives me this message: unknown terminal: /usr/bin/Eterm check the TERM environment variable. also make sure that the terminal is defined in the terminfo database. alternatively, set the TERMCAP environment variable to the desired termcap entry. any help is greatly appreciated. -- mike irwin [EMAIL PROTECTED] - "...all that we see or seem, is but a dream within a dream."
Re: Problem verifying gpg signatures with pgp
On Fri, Feb 04, 2000 at 06:51:25AM -0600, Christopher Uy wrote: Does this happen with any message you sign, no matter how simple it is? I know I had problems verifying signatures for a while, but the problem was intermittent and ultimately turned out to be related to a procmail bug and a line in my signature file. :-/ Yah pretty much. I am using qmail for my MTA and procmail for my MDA, but I'm not using any "recipes" at all. Just normal delivery to /var/spool/mail/. Strange how the PGP sigs are verifiable by both GPG and PGP, yet the GPG sigs can only be verified by GPG. There must be something more than meets the eye happening. :) Shouldn't be a GPG bug because making a sig from the shell works fine. I guess I'll have to look at the arguments Mutt sends to GPG and start changing them one by one until I find out which one breaks it. Will post to the list once I figure it out of course. :) Thanks.
Problem verifying gpg signatures with pgp
I'm using gpg 1.0.0 with the IDEA and RSA modules loaded. It was all compiled from source. Same goes for PGP 5.01. With GPG as my "pgp_default_version" I have no problem decrypting and verifying signatures and messages made by either GPG or PGP. When I use PGP with mutt however (pgp_default_version = pgp5), I can decrypt/verify PGP messages fine, and even decrypt GPG messages. But verifying signatures gives me a "Bad signature from..." error. FWIW, I think the problem is caused somewhere in the signing (not in the verification) stage. I say this because if I pipe the GPG-signed message to "pgpv" it still fails. I wouldn't have posted this to the mutt list, except this only is a problem for me in Mutt -- when I sign a file with GPG in a shell, I can verify it fine with PGP. Only when signing from within Mutt does it appear broken. Does anyone have any insight into what may be going on?
Re: Folders don't display new status/Signing messages
You have to set the "mailboxes" line in your .muttrc. http://www.mutt.org/doc/manual/manual-3.html#ss3.10 Sorry to piggyback on your post, but this brings up a question I have about this definition. Is it possible to specify "all mailboxes in my $folder directory?" I tried: = =* =/* and ~/mail/* /home/user/mail/* None of them worked. Can it be done? On Wed, Feb 02, 2000 at 03:04:17PM -0800, Joshua Haberman wrote: For some reason, mutt fails to display the "N" next to folders with new messages in them on the folder index. The default folder_format string begins with %N and should do the trick, and I additionally tried manually setting the folder_format to include a %N but to no avail: no N ever appears. I know Mutt knows when messages are new, because the message index screen correctly marks N's next to new messages. What do I need to do to get this N to also appear in the folder index? Also, I notice a few people who are signing messages to the list: are there any netiquette rules indicating whether it's helpful/polite to sign messages posted to public forums? What are people's feelings on this? Joshua -- GPG Public Key: http://joshua.haberman.com/files/pubkey.pgp
Re: Tagging on To line??
Hi. Using: ~t string works for me. (With the T command which you've already got.) On Mon, Jan 31, 2000 at 10:26:38PM -0700, Charles Curley wrote: I have tons of email I'd like to tag based on the contents of the To: line. Searching (T) doesn't seem to do it. How can I do that? Thanks. -- -- C^2 No windows were crashed in the making of this email. Looking for fine software and/or web pages? http://w3.trib.com/~ccurley
Re: qmail and From:
I've been doing this on my qmail system; however, qmail still puts a "Return-Path" header on the message which indicates the local username who called "sendmail." I haven't found any way around this yet. You can set the $QMAILUSER and $QMAILHOST env vars to specify what you want in the Return-Path header but once you do that, all msgs you send will use it -- even if you generate your own From lines. I used to get around this with Pine by delivering to the local smtp daemon instead of calling sendmail; this way, the MTA wouldn't know what userid was connecting to it and therefore would generate a Return-Path that matches the From line. But Pine would put in an X-Sender line when you use a From address different from your local userid, so that was not a perfect solution either. I am trying to figure out if there's a way to force qmail to generate the Return-Path header based on the From line, but haven't seen any way to do this so far. If it can't be done, then the only other solution I can see would be to run nullmailer (or equivalent) as a middle-man, passing it your mail locally and having nullmailer hand it to the local smtpd. If anyone has any better ideas, I'm all ears. :) I can't stand having my local username in the headers when I am trying to "hide" it with my own headers. On Sat, Jan 29, 2000 at 06:20:35PM +0200, Mikko Hänninen wrote: Patrick Walsh [EMAIL PROTECTED] wrote on Fri, 28 Jan 2000: As far as I can tell setting the From: when using qmail with mutt requires that the environment variables MAILUSER, MAILHOST, MAILNAME be set before starting mutt, along with QMAILINJECT=f. It doesn't require them, though can specify them if you like. Is there a way to set the From: within a mutt session when using qmail as the MTA? Certainly. To set the real name part, look at the $realname variable. The hostname part can be changed with $hostname, $hidden_host variables. If you want more control over the From: header, you can always specify it with something like my_hdr From: Your Name [EMAIL PROTECTED] Hope this helps, Mikko -- // Mikko Hänninen, aka. Wizzu // [EMAIL PROTECTED] // http://www.iki.fi/wiz/ // The Corrs list maintainer // net.freak // DALnet IRC operator / // Interests: roleplaying, Linux, the Net, fantasy scifi, the Corrs / An unbreakable toy is good for breaking other toys.
folder-hook not executed with mutt -f cmdline
Hi, it seems that my folder hooks are not run when I load a specific folder from the command line. For instance, I have a folder hook that looks like this: folder-hook =mutt "my_hdr From: Jim Breton [EMAIL PROTECTED]" But if I load mutt like this: mutt -f ~/mail/mutt It does not set the my_hdr as specified above. Loading the folder from the list does cause it to be set. Is this behavior intentional?
Re: qmail and From:
Haha -- guess what. :) Just on a whim I tried setting the Return-Path with a "my_hdr" command and qmail did honor it! Perfect! I was all excited about it, but you beat me to the punch. :) Thanks! On Sat, Jan 29, 2000 at 08:42:58PM +0200, Mikko Hänninen wrote: Jim Breton [EMAIL PROTECTED] wrote on Sat, 29 Jan 2000: I've been doing this on my qmail system; however, qmail still puts a "Return-Path" header on the message which indicates the local username who called "sendmail." I haven't found any way around this yet. Ahh. That's the envelope sender. I thought about mentioning it, but you specifically asked about the From: header, so I left out the envelope sender to avoid potention confusion. :-) You can set the $QMAILUSER and $QMAILHOST env vars to specify what you want in the Return-Path header but once you do that, all msgs you send will use it -- even if you generate your own From lines. I use MAILUSER and MAILHOST to set it to what I want, but those ought to work too (according to the qmail-inject man page). I set these variables at login time, so they're set for all emails I send. You could also create a wrapper script to mutt that will set them for you. It's by far the easiest solution, I think. Alternatively, you could also set a Return-Path: header for your emails from within Mutt, and qmail should (by default) respect that. I've not tried that approach myself though. I am trying to figure out if there's a way to force qmail to generate the Return-Path header based on the From line, but haven't seen any way to do this so far. I don't think you can do exactly this. You need to specify the envelope sender *some* way -- either via the environment variables, or the Return-Path: header, or on the command line (sendmail -fsender@address). The latest developement version of Mutt has support for doing the latest too, I think. If anyone has any better ideas, I'm all ears. :) I can't stand having my local username in the headers when I am trying to "hide" it with my own headers. You should be able to hide it with one of the options above; if you still have trouble feel free to write back. :-) Mikko -- // Mikko Hänninen, aka. Wizzu // [EMAIL PROTECTED] // http://www.iki.fi/wiz/ // The Corrs list maintainer // net.freak // DALnet IRC operator / // Interests: roleplaying, Linux, the Net, fantasy scifi, the Corrs / Red meat isn't bad for you. Fuzzy blue-green meat is bad for you.
line wrap
Quick question. Do people generally find it annoying if I don't use line wrap in my editor (vim)? I see that most other people on this list appear to be using line wrap in their editors, I always left mine disabled figuring the client's reader/pager would handle the wrapping. Is this a bad idea?
Re: folder-hook not executed with mutt -f cmdline
Yep you were right. Thanks. :) Btw, this leads me to another question. Suppose I have two folders, one called "mutt" and the other one is "mutt-sent" If I want to trigger a command in the "mutt" folder but not in "mutt-sent" how can I do this? The hook seems to make a match on either one. I've tried "mutt" and "=mutt" for the hook. There must be some regexp pattern that would get this right. Or am I being too picky, and should just rename my folders? ;) On Sat, Jan 29, 2000 at 12:57:39PM -0600, Jon Miner wrote: * Jim Breton ([EMAIL PROTECTED]) [000129 12:40]: folder-hook =mutt "my_hdr From: Jim Breton [EMAIL PROTECTED]" But if I load mutt like this: mutt -f ~/mail/mutt A guess (untested because I don't have time right now to test it) is that it's noticing that "~/mail/mutt" != "=mutt", even though folder is (presumably) set to ~/mail. What happens if you type mutt -f =mail? jon -- .Jonathan J. Miner--Division of Information Technology. |[EMAIL PROTECTED] University Of Wisconsin - Madison| |608/262.9655 Room 3149 Computer Science| `-' "The best way to get a drink out of a Vogon is to stick your finger down his throat..." -- The Book, on one of the Vogon's social inadequacies. From _The_Hitchhikers_Guide_To_The_Galaxy_ by Douglas Adams
[jamesb-mutt@alongtheway.com: look Ma, no Pine]
Sorry if anyone gets this twice. I sent it yesterday but for some reason it didn't show up on the list, and a couple msgs I wrote just a few mins ago have already shown up so I figure it's pretty safe to say this msg got lost somehow. I'm also taking this opportunity to add another question. :) - Forwarded message from Jim Breton [EMAIL PROTECTED] - As some of you may remember I just "discovered" Mutt only a few days ago. Well, those few days have been awesome, Mutt is unbelievable! I do have a few questions though. :) For one, I'm always wondering if anyone has any patches for matching on something other than the beginning of a Subject line ;-) Also: is it possible to change header settings without changing folders (where I am using folder-hooks to change headers)? Something like a macro. I haven't tried any run-time reconfiguration yet, I've been doing everything in my .muttrc with hooks so far. But let's say I wanted to send a msg from this address (jamesb-mutt) but I was in a folder which ran a hook that changed my address to something different. There must be some way for me to re-config without changing folders and without editing the headers by hand? Lastly, is there an address-book-ish feature where I can type a common-name like "Jim" and it will expand the address to "[EMAIL PROTECTED]"? Would aliases in my .muttrc do this? Sorry for the newbie-ish questions, I can figure it all out but I've had to hammer out tons of mail over the last few days and have only had time to "get to know" the most superficial Mutt features so far. Thanks! P.S. If I have my spoolfile in /var/spool/mail/$USER but I don't want to have to browse to it constantly in the folder list, is it safe to make a symlink in my ~/mail/ directory which points to the file in /v/s/m/? If that's a bad idea I could just have my mail written to my home dir but I'd rather not do that yet because I have a few other things that depend on the file being in /v/s/m/ at the moment. - End forwarded message -