Re: Displaying PGP signature *after* the message?
At 10:27 AM EDT on October 16 PeterKorman sent off: In a perfect world, I think I'd want to know only if the signature did *NOT* check out against the keyserver copy. IMHO, that would weaken the point of crypto signatures. First, most* signature failures are innocent, being due to MTA mangling along the way, like escaping periods at the start of a line. By only seeing gpg when there's a problem, it could freak you out that much more when it happens, and when you verify with the sender, give you** the impression that it's worthless. But it's not. All the signatures that *do* check out OK are saying something about the legitimacy of their messages. By not checking good sigs, you are lowering their status to the same level as unsigned messages, so gpg users can't win either way :-( * in my limited experience. ** and by you I probably mean more newbieish people who inherit your .muttrc. Admittedly that problem mostly goes away if you have the %Z flag (IIRC) in your index, so you can easily see which ones are signed, even if you don't check every signature. (I don't on mailing lists.) At 2:04 PM EDT on October 16 PeterKorman sent off: Content-Description: Why I automatically verify On Wed, Oct 16, 2002 at 07:35:01PM +0200, Ren? Clerc wrote: * PeterKorman [EMAIL PROTECTED] [16-10-2002 16:30]: On Wed, Oct 16, 2002 at 03:02:31PM +0200, Ren? Clerc wrote: * Stephane Bortzmeyer [EMAIL PROTECTED] [16-10-2002 14:52]: Or to hide it unless specifically called? unset pgp_verify_sig snip would be the benefit of this? It mitigates against 2 of my weaknesses. Bad memory and lazyness. Patience seems to be something of which I have enough; maybe too much. I have enough patience to wait for keyserver response. A keyserver response wait tells me that I've never before read a message (so far I've never encountered a downed key server at the site I'm using) Oh, so you *are* a newbie! ;-) Pardon my svenning, but key servers just don't seem to stay up for very long. If I always verify, then I don't have to remember the verify command. Put this in your .muttrc: # Check a signature. Thanks to David Champion [EMAIL PROTECTED] macro index \Cv enter-commandset pgp_verify_sig=yesenterdisplay-messageenter-commandset pgp_verify_sig=noenter Verify PGP signature macro pager \Cv ienter-commandset pgp_verify_sig=yesenterdisplay-messageenter-commandset pgp_verify_sig=noenter Verify PGP signature mnemonic: control-v for Control Verify. I like to read, then verify if necessary, which is more or less what the subject asks for. -- We are all in the gutter, but some of us are looking at the stars. - O. Wilde Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html msg31872/pgp0.pgp Description: PGP signature
Re: ISO cancel option in quit prompt from send-message
At 8:57 PM EDT on October 8 Erik Christiansen sent off: On Tue, Oct 08, 2002 at 02:47:53PM -0500, David Champion wrote: Control-G cancels most prompted operations. With your pardon, I'll say that _doesn't_ ring a bell. Having tried the conventional escapes, including ^C, ^D, and Esc, I'd surmised that mutt lacked the facility. Is there a known rationale for the ding-dong choice? :) It's what the one true editor uses, but you're right, C-g should be a hard-coded addition to the ? menu, right at the top. -- At the evening service tonight, the sermon topic will be What is Hell? Come early and listen to our choir practice. - church bulletin Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: ISO cancel option in quit prompt from send-message
At 9:10 PM EDT on October 8 Michael Elkins sent off: Rob Reid wrote: It's what the one true editor uses, but you're right, C-g should be a hard-coded addition to the ? menu, right at the top. Technically, control-G is not a valid command in that context. It only works inside of prompts, Fair enough, but notice that we're talking about prompts instead of input fields. Maybe the title of section 2.2 of the manual should be changed to Editing Input Fields (Typing at Prompts). and there is no help menu available when using the line-editor. It would be misleading to put it elsewhere. Maybe ^? for a help menu...or just ? if help is set? Naaa...too annoying. This is the first time I've seen this problem on the list. -- Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Autoview images in the pager
Hi, At 11:58 PM EDT on October 5 Mike Leone sent off: * Rob Reid ([EMAIL PROTECTED]) wrote this on 10 05, 02 at 22:35: Danger, Will Robinson! ;-) At 8:21 AM EDT on October 5 Viktor Lakics sent off: I have a crazy idea, I wanted to ask you about: Has anyone ever tried to work out how to autoview graphics inside mutt? You might know this already, but a common spammer tactic is to include images in their html mails like img src=http://spam.server.com/Viktor_actually_read_this_spam.gif; that let them know that you actually read their spam, *if* you read the message in a graphical browser. From then on you can count on that address receiving the GSSSP (Gross Solar System Spam Product). Most times, yes. But the majority of people actually don't use console based mailers, but graphical ones. And the graphical ones usually show the HTML directly, that is, they don't spawn a browser. Think of Outlook or Eudora. Why does that matter? Each web bug is unique, so any request for it, from anywhere, anything, or anyone, still dooms the address the bug refers to. On Linux, at least, some graphical mailers - for example, Evolution - can be configured not to get images in HTML email at all, As I'd expect... or only if the sender is in the address book. That's pretty clever. Evolution would be tempting if it weren't so...gooey. This is probably a more pervasive problem on Windows than Linux. Definitely, because the average Windows luser is doomed by not bothering to figure out what's going on, but I'm sure that M$ will add a similar image blocking feature if it's not already there. Of course, they'd put themselves and sufficiently high payers on the whitelist. More annoying are the 1x1 pixel images that you can't even see, that do the same thing. I believe the term is web bug. I wasn't excluding them. Personally, I like the 1x1s better than the garish, bandwidth hogging, CPU killing, animated gifs. ObMutt: Would a macro that 1. autoviews/mailcap displays stuff from trusted senders* or 2. defangs (temporarily, in a display filter way, since untrusted innocents are included) stuff from everyone else necessarily require mutt to understand if statements? * possibly approximated by the aliases file, although you can't trust anyone using untrusted software, no matter how nice they are. -- Why don't sheep shrink when it rains? - Jack Handey Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: location of signature.
At 1:37 AM EDT on September 6 Paul Brannan sent off: On Thu, Sep 05, 2002 at 07:49:40PM -0400, [EMAIL PROTECTED] wrote: I disagree. While bottom posting is appropriate for most public forums (because discussions on these forums generally involve a point-by-point debate), there is a valid use for top posting. In particular, if someone sends you a long email (and it is necessary and/or appropriate to quote the email or a large portion of it), then replying at the top saves the reader the time of scrolling to the bottom to find the reply. Generally, this situation arises in personal emails much more so than it does in public forums. I don't see how that's valid for replies, but I sometimes use it for *forwards*. i.e. I am passing this on to you because bla bla bla... - Content of forwarded message -- instead of putting bla bla bla down here where they won't see it right away. -- Diplomacy is the art of letting the other party have things your way. - Daniele Vare Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: location of signature.
At 9:39 PM EDT on September 4 Bo Peng sent off: There is nothing wrong with either order. Nobody is 'corrupted' by anything. Wrong. People are. Software as good as mutt should be neutral between these preferences, i.e. provides support for both styles. No, good != neutral. Good software makes bad behavior hard. As far as bandwidth is concerned, you may not mind, but those using modems, especially in areas where internet/phone time is expensive, do mind. -- loquacity, n. A disorder which renders the sufferer unable to curb his tongue when you wish to talk. - Ambrose Bierce, The Devil's Dictionary. Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: location of signature.
At 7:17 PM EDT on September 4 Bo Peng sent off: I THINK it is better to put the reply BEFORE quoted text and this has nothing to do with M$. It is natural (to me) to put important part (my reply) before non-important part (quote) I understand your line of reasoning, but I think most people (if they haven't been corrupted by years of the other way) prefer a temporal ordering, i.e. old stuff at top, new stuff at bottom. and keep my signature closer to the main body. I'd rather keep each sentence of my reply as close as possible to the point that it is replying to. This also makes an email easier to read if the quote is long. The quote should not be long, and the biggest reason why so many UNIX types hate M$ for promulgating the bottom quote style is that it encourages people to attach entire threads at the bottom of each message, guaranteeing that noone will ever read them. -- ...from a gulf beyond the sun and stars that illume the Lethean shoals and the vague lands of somnolent visions, I floated on a black unrippling tide to the dark threshold of a dream. - Clark Ashton Smith Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: Emacs question
At 11:33 PM EDT on August 13 Andy Davidson sent off: On Tue, Aug 13, 2002 at 09:35:49PM -0500, Bo Peng wrote: I am in mal mode now. I can see --:** mutt-. (Mail Fill)--L9-All I'm really not the one to answer this --- you'll note that I was the one having problems :-) --- but I have color in mail mode. But only for message headers. (I have 'set edit_headers' in my .muttrc) The text is not colored, although on replies, quoted text is colored red. post mode optionally does syntax coloring of any or all of headers, (multiply) quoted text, addresses, URLs, signatures, and emoticons. http://astro.utoronto.ca/~reid/mutt/ It also has a lot of other features, some more useful than highlighting, some less. -- How do they get the deer to cross at that yellow road sign? - Jack Handey Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: conditional rc.
At 9:50 AM EDT on August 14 Bo Peng sent off:
Hi, there
Can I do something like:
if have_X
editor = gvim -f
else
editor = vim
end
in my .muttrc file?
I don't think so, but instead of putting something like that in each
application's configuration file, you can put it in a script,
i.e. /usr/local/bin/editor or ~/bin/editor
%=
!/bin/sh
editor=jed
serverrunning=`ps -u ${USER} | grep emacsserver`
if [ ${serverrunning} != ] ; then
editor=emacsclient
fi
exec ${editor} $
%
and setenv (or export) EDITOR editor will take care of most things, and
specifying editor in the app's config file will take care of the rest.
--
Some people have told me they don't think a fat penguin really embodies the
grace of Linux, which just tells me they have never seen a angry penguin
charging at them in excess of 100mph. They'd be a lot more careful about what
they say if they had. - Linus Torvalds (announcing Linux v2.0)
Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: Quoting the signature in replies (yes, I want)
At 3:39 AM EDT on August 13 Stephane Bortzmeyer sent off: On Mon, Aug 12, 2002 at 12:12:56PM +0200, Roman Neuhauser [EMAIL PROTECTED] wrote a message of 18 lines which said: But some people who write me do use -- at random places :-( and I want to quote it nevertheless. I cannot find an option 'include_sig_in_replies'. Any idea? while i might be very much wrong, i don't think this is mutt. looks like your editor strips the sig. You're right, I use Emacs' post.el URL:http://astro.utoronto.ca/~reid/mutt/ and I was able to include the sig with: ; The original regexp also had a -- (custom-set-variables '(post-signature-pattern \\(Cheers,\\| \\))) Many thanks, everything works now. But you're breaking it for the 90% or whatever of messages that have proper formatting. Try changing post-signature-pattern to \\(-- \\|Cheers,\\| \\) Normally the space is left out because of broken MTAs and people who've seen the ^-- but didn't know about the space. The best way to fix that is with this procmail recipe: # Correct wrong sig-dashes :0 fBw * ^--$ | sed -e 's/^--$/-- /' but post.el does not assume you use procmail by default. What I really meant in my previous post (but didn't specify because I didn't know what editor you were using) was to Customize post-backup-original to t, and then select and paste from *Original* when necessary. You can also try explaining to your correspondents why they shouldn't use ^-- \n except at the start of a signature. -- I love California. I practically grew up in Phoenix. - Dan Quayle Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Quoting the signature in replies (yes, I want)
At 6:12 AM EDT on August 12 Roman Neuhauser sent off: Date: Mon, 12 Aug 2002 11:41:38 +0200 From: Stephane Bortzmeyer [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Quoting the signature in replies (yes, I want) When I reply to a message with something starting with -- in it, mutt does not quote it, probably because it assumes it is a signature. But some people who write me do use -- at random places :-( and I want to quote it nevertheless. I cannot find an option 'include_sig_in_replies'. Any idea? while i might be very much wrong, i don't think this is mutt. looks like your editor strips the sig. You're absolutely right. It should be possible to either A. Get the editor to save an unmunged buffer somewhere of what it gets from mutt or B. Have a macro inside mutt to toggle $EDITOR between good and bad netiquette. i.e. set editor=... %s -- There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarre and inexplicable. There is another theory which states that this has already happened. - Douglas Adams, The Hitchhiker's Guide to the Galaxy Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Sharing Mutt
Hi, At 7:11 PM EDT on August 10 Lawonna Daves sent off: I got Mutt on SuSE Linux 8.0, but it's configured for root only. I'm relatively new to Linux and am not sure how to configure Mutt so that the other users on the system can have the advantage of it. Can someone advise or direct me to documentation that would help? I've been looking in all the wrong places. If it doesn't run at all for nonroot, chmod a+x `which mutt`. But I have a feeling that it's either something more like readonly user mailboxes or only root having a ~/.muttrc (i.e. no /usr/local/etc/Muttrc). More details on what the problem is would help, as well as the output of mutt -v. The readonly mailbox problem is common in Red Hat, and shows symptoms like 87 thesis% ls -ld /var/spool/mail drwxrwxr-x2 root mail 1024 Aug 10 21:56 /var/spool/mail/ and is fixed by configuring and compiling mutt with setgid support. -- Penguins are so sensitive to my needs. - Lyle Lovett Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: spam filter
At 4:49 PM EDT on July 29 Rob 'Feztaa' Park sent off:
Alas! Andre Berger spake thus:
By the way, what would an exmaple
^^
It's not mutt, but since I don't have time to read the procmail list...
procmail rule to add a sender to the spamassassin blacklist look
like?
Probably something along the lines of this (but I'm a little rusty;
the flags are probably wrong):
:0 Wh:
* some spam heuristic, like all caps subject lines
*X-Spam-Flag: Yes
(unchecked) would use the result of all applied spamassassin tests.
|grep ^From: |some sed to extract info from the header killfile
I think I got this addysort tidbit from Gilbert linuxbrit somebody, to use
instead of sed.
#!/usr/bin/perl -wn
# Picks out the actual address from the From: line
unless (/\/) { print; } else { print /([^]+)/, \n; }
:0 a:
spamfolder
But what's the point? I love spamassassin because it lets me *avoid*
blacklists, and their maintenance, and filter on the spamminess of the message
itself. (unlike, say, twerp /. moderators... :-P ) I suppose you could check
the blacklist first, and skip spamassassin if it matches, to save some
computation, but my preSA experience with a personal blacklist is that there's
only a handful of spammers* that this is worthwhile for, because they don't
change their address.
*and I'm not sure they're even spammers (so I haven't razored** them) or just
mailing lists with overly open subscription policies. But they smell spammy
enough that I'm too chicken to unsubscribe.
** An interesting alternative to blacklisting.
--
Minds that have nothing to confer find little to perceive. - Wordsworth
Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/
PGP Key: http://astro.utoronto.ca/~reid/pgp.html
msg29946/pgp0.pgp
Description: PGP signature
Re: Signature
At 10:54 AM EDT on July 8 Nelson D. Guerrero sent off: and how would I go about doing this? [05/07/2002] - Michael Tatge - [mutt-users]: Make you editor delete it for you. One way would be to install post mode for emacs: http://astro.utoronto.ca/~reid/mutt/ -- Half the lies the opponents tell about us are not true. - Sir Boyle Roche Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: [dan@hld.ca: Re: [oclug] GPG and mutt]
At 9:26 PM EDT on June 10 Brenda J. Butler sent off: I'm trying to use GPG via mutt, and I find there is an annoying two-second wait every time I hit a signed message in the index while GPG verifies if the signature is ok. I'd like to turn off automatic verification, but I can't find the command to verify the signature on demand. Is there one? I'm not keen on setting pgp_verify_sig to ask-yes or ask-no, that's not much more efficient than just waiting 2 seconds for the check to be done. There is a command to verify the old-style PGP signature, I want a user-initiated command to verify if the GPG signature is ok. Put something this in your .muttrc (or the keybinding file if you've split it up): # Check a signature. Thanks to David Champion [EMAIL PROTECTED] macro index \Cv enter-commandset pgp_verify_sig=yesenterdisplay-messageenter-commandset pgp_verify_sig=noenter Verify PGP signature macro pager \Cv ienter-commandset pgp_verify_sig=yesenterdisplay-messageenter-commandset pgp_verify_sig=noenter Verify PGP signature Then control-v will do what you want once you source the bindings. -- Can vegetarians eat animal crackers? - Jack Handey (Yes - R.R.) Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: attribution with date+time - timezone required? no!
At 9:51 AM EDT on May 28 David T-G sent off: GMT does not change in the summer. GMT is GMT all year round. That's why it's Greenwich Mean Time and not Greenwich Most-of-the Time. UK Daylight time is BST , or GMT+1. I think you're behind the times, David ;-) I was taught that what you said above was true for a while until the town (village?) of Greenwich got tired of not having daylight savings time (and being an hour off from the rest of Britain for the summer) and relinquished GMT. Thus was UT (Universal Time) born, which is the real, i.e. no Daylight Savings, time at the longitude of Greenwich. In other words it's what used to be called GMT, but Greenwich itself doesn't use it anymore, so GMT is an anachronism. If anyone from the vicinity of Greenwich would like to correct me, please do. Puns are optional. -- I didn't attend the funeral, but I sent a nice letter saying that I approved of it. - Mark Twain Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: attribution with date+time - timezone required? no!
At 4:37 PM EDT on May 27 Sven Guckes sent off: * Nicolas Rachinsky [EMAIL PROTECTED] [2002-05-25 08:27]: * Sven Guckes [EMAIL PROTECTED] [2002-05-25 01:36:30 +0200]: Do you really think the time is usefull without a timezone? no timezone given - GMT! talk about defaults here.. Do you really mean GMT (daylight savings) or UT (no temporal discontinuities, please)? -- Mathematics, rightly viewed, possesses not only truth, but supreme beauty... - Bertrand Russell Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Virtual Folders in mutt
Hi, At 3:32 AM EDT on May 20 Anthony Towns sent off: I've been getting gradually more annoyed with the way my mail archives are organised recently, and, after playing with Evolution a little bit, got to thinking that vFolders might be the answer. So, I've been trying to figure out some way of doing the same sort of thing in mutt. First, has anyone done this before? Is there a FAQ or HOWTO I could be reading? What I'm thinking is basically having a huge morass of mail shoved in a directory somewhere, with a fancy index. Whenever I want to look at mail in there, I construct a query (Show me all mail with [EMAIL PROTECTED] in the From/To/Cc headers or Show me all mail with message-id blah) and then have mutt show me all the mail that matches that query as a virtual folder. This isn't exactly what you asked for, but you might want to look at the Remembrance Agent, http://rhodes.www.media.mit.edu/people/rhodes/RA/ -- We have to pursue this subject of fun very seriously if we want to stay competitive in the 21st century. - George Yeo, Singapore's Minister of State for Finance. Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: about spam
At 2:16 PM EDT on May 9 Maximilian Szengel sent off: I read about spamassassin here and decided to install it myself. Since config files. I just added the needed rules to my procmailrc. Well, it works, but I was wondering how a mail in this mailinglist could get a score of 4.4 and a real spam message gets 3.6. It's not too offtopic since it involves this list (and just about every other list ;) I can't answer your question, because I filter all my list mail before the spamassassin check in my .procmailrc. i.e. put something like :0: * ^TOmutt@ muttin above the spamassassin part. Other people have fancy recipes that attempt to catch all mailing lists in one recipe, but that's OT. List filing before spam checking doesn't catch spam in the lists, but the lists I read don't pass on spam to the general membership. OT: 3.6 seems low for a spam. Maybe it's just a fluke, or maybe you should customize the scores on the various spamassassin tests. -- Parkinson's Law: Work expands to fill the time allotted it. Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: [OT] Emacs and line wrapping (was: Re: About wrapping lines.)
At 3:41 PM EDT on May 4 Radek Spacil sent off: On [04/05/02] 18:23, Jussi Ekholm wrote: Radek Spacil [EMAIL PROTECTED] wrote: On [04/05/02] 12:36, Jussi Ekholm wrote: I think, that something is wrong with my Emacs, because 'M-x auto-fill-mode' doesn't start wrapping lines. Or, then the wrong thing is me -- should I somehow configure auto-fill-mode through 'customize' or something? It just doesn't wrap the lines... When I'm replying to such an email I'm using M-q, but first I have to make one empty line before and after the part which I cite (emacs would take whole message as paragraph since even empty lines are prepended with '' or something alike). I recommend filladapt. -- Piratization, n.: The process of giving many public services over to pirates with the thought of more efficient service delivery - heard on Floridian community radio Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: colorize collapsed threads with new messages
At 3:06 PM EDT on April 28 Andre Berger sent off: Is it possible to colorize the visible messages of collapsed threads if the threads contain one or more new messages? Does putting this # collapsed threads color index brightgreendefault ~v color tree brightgreendefault in your ~/.muttrc help? -- coude tat: When the person with the spectrometer gets more telescope time. - D. Kaisler Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: colorize collapsed threads with new messages
At 1:56 PM EDT on April 29 Andre Berger sent off: * Rob Reid [EMAIL PROTECTED], 2002-04-29 13:25 -0400: At 3:06 PM EDT on April 28 Andre Berger sent off: Is it possible to colorize the visible messages of collapsed threads if the threads contain one or more new messages? Does putting this # collapsed threads color index brightgreendefault ~v color tree brightgreendefault in your ~/.muttrc help? Here's an exapmle of what I meant. When my threads are collapsed, I see, for example 352 2(20) 020429 loren ( 9) How can i make vim 354 (20) 020429 Sukesh( 8) Editor Designing In C++ 355 5n (20) 020429 Jeroen Valcke ( 9) VIM: indent behaviour ^This collapsed thread has new messages, and I would like to colorize msg 355 (which itself is not new) -Andre Ah. I missed the new messages part. I don't know how to do exactly what you asked, but your question has come up before, with the following satisfactory answer: unset collapse_unread That's what I do, and I can recognize the threads with new messages, because they're the expanded ones. -- Here's something to think about: How come you never see a headline like Psychic Wins Lottery'? - Jay Leno Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: Regexes in search?
At 10:38 PM EDT on April 25 Rob 'Feztaa' Park sent off: Alas! Rob 'Feztaa' Park spake thus: folder-hook =foobar macro index Left change-folderTabsearchfoobarEnter If I change foobar to ^foobar, it doesn't work, but that line as it is works fine. Is there a different search command that does have regexes? This is really weird. I'm generating these with a perl script, and if I tell it to print ^$mbox, I'll get something like ^inbox printed, but mutt will interpret that as ^Inbox (a tab and then nbox). But if I tell it to print ^ ^H$mbox, it'll print the same thing but mutt will behave properly. I discovered this recently. The right (easiest) way to quote a ^ in mutt regexps is ^^, because mutt uses ^x for Control-x so it provides ^^ for a simple caret. ^I means control-I, a tab. I know you have a working veresion already, but you shouldn't! ;-) I thought ^ ^H meant control-spacebackspace. It works for me with just ^^blah. -- Let's just sit here a moment and savor the impending terror. - Calvin Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: x-authentication-warning
At 12:19 AM EDT on April 26 VB sent off: Do my headers look ok? When I send messages to myself (I'm real lonely) it says X-Authentication-Warning and gives out some info that you don't need to know. I looked at google and it suggested adding needmailhelo under the privacy flag section for sendmail.cf. But that did not work. What can I do to clean up my act? vberic X-Authentication-Warning: sunny.localdomain: rooot set sender to rooot? 123 -- In Paris in December 1997, just before being convicted of the murders of two counterespionage agents, international terrorist Carlos the Jackal was sentenced to 10 days' solitary confinement for calling a prison guard a gnu. - News of The Weird That _is_ weird. Didn't the guard know it was a compliment? Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
quoting with push
Hi, I would like to have mutt automatically mark certain messages as deleted in my spam folder, based on the output of SpamAssassin, but I am having trouble with the quoting: folder-hook spam push 'D~b \'^SPAM: Hit\! \(1 point\) BODY: Image tag with an ID code to identify you\'\n' As you can see I've used \' to try to protect the 's needed around the matching line because 's are already used around the push statement. When I try this I get a beep and Key is not bound. Press ? for help.. So how can I do three levels of quoting? Is it possible, and is there a way I can avoid it? Thanks -- Documentation is like sex: when it is good, it is very, very good; and when it is bad, it is better than nothing. - Dick Brandon Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: quoting with push
Hi, At 1:24 PM EDT on April 23 David T-G sent off: I would like to have mutt automatically mark certain messages as deleted in my ... So how can I do three levels of quoting? Is it possible, and is there a way I can avoid it? Rather than just escaping your single quotes because you're using single quotes already, you have to escape them deeply enough. Starting from the inside out, what you want is effectively ~b '^SPAM ...' and so you pass that pattern to delete-pattern. Now, you can skip the double quotes, because D takes a pattern specifier like ~b and the pattern itself (which must be quoted if it includes spaces); I just tried that in my mailbox and no means of wrapping the ~b with the pattern was acceptable. So now you have D~b '^SPAM...' and you want to put that in a folder-hook. You must protect the quotes if they will be interpolated, so you wrap the whole thing in another layer like push D~b '^SPAM...' to make it work (though that part is untested). :-P Always test! I had already tried that, and it doesn't work even though it should. Of course, testing is a pain since it means restarting mutt to undo the bad folder-hooks. Meanwhile, generally the way to escape embedded quotes is to not just escape the quote but also escape your escape char, since the first pass over the line (the folder-hook level) will not only remove the outer set of quotes (doubles above) but also take literally any escaped characters and get rid of your escaping -- but that's usually too early! It can be It was an escaping problem, although I thought I'd tried that with \\! But n! The second \ needs to be escaped so it has to be ! The working hook: folder-hook spam push 'D~b \^^SPAM: Hit! (1 point) BODY: Image tag with an ID code to identify you\\n' Note the different escaping method for ^. Hopefully some other SpamAssassin users will find this helpful. -- Faith, n. Belief without evidence in what is told by one who speaks without knowledge, of things without parallel. - A. Bierce, The Devil's Dictionary Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: quoting with push
Hi, At 2:20 PM EDT on April 23 David T-G sent off: ...and then Rob Reid said... At 1:24 PM EDT on April 23 David T-G sent off: Rather than just escaping your single quotes because you're using single quotes already, you have to escape them deeply enough. Starting from ... push D~b '^SPAM...' to make it work (though that part is untested). :-P Always test! I had already tried that, and it doesn't work even though it Ah; you didn't say that part. Ain't I a stinker? So how on earth do you get D~b 'pat' to work when that breaks for me? I think your double quotes are getting eaten and you're not realizing it. I think (and you more or less said) that you've been testing D directly from mutt instead of with a folder-hook in a file. It's different, I tell ya! ;-) Well, that's easy, since I can always pop open another window and run mutt as many times as necessary. The real pain is developing a mailbox where your pattern will work so that I can do the testing Would you like me to send you some spam? ;-) Fortunately it's not necessary. It was an escaping problem, although I thought I'd tried that with \\! But n! The second \ needs to be escaped so it has to be ! The working hook: folder-hook spam push 'D~b \^^SPAM: Hit! (1 point) BODY: Image tag with an ID code to identify you\\n' Now that is *strange*. I expected (2^^n)-1 backslashes like ! \! \\\! \\\! and certainly not an even number. When you have \\! you're escaping your escape but not protecting the bang. Exactly; protecting the escape so it can protect the ! Oh, wait; since you're quoting your pattern (which must be working, because you don't have to escape the spaces or fill them with dots [which I find much easier to manage]) D'oh!! Note the different escaping method for ^. Yeah; why? Because the mutt manual says to ;-) ^x is interpreted as control-x, so mutt recognizes ^^ as ^. After that, it doesn't need any protection. Yes, it's weird, but otherwise benign. -- He eats like a bird... five times his own weight each day. - fortune Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
New version of post, an emacs mode for composing messages
post.el is an emacs mode for composing messages with user-agents like mutt or slrn. If you haven't already tried it, it will make your email experience more complete. Get it here: http://astro.utoronto.ca/~reid/mutt/ If you don't use emacs, install it first, and use (gnu|emacs)client to silence the evil forces of vi. Changes in this version: - A major (but not total) synchronization with Dave Pearson's post-mode. header-set-followup-to and header-set-organization should work now. - Syntax highlighting now works for quoted email addresses and URLs. - *bold* words are now highlighted. - Emoticons can now be highlighted. In case you're curious, I verified that gnus' smiley-ems.el works with post, but I decided that it wasn't ideal. - post-url-text-pattern changed to post-url-pattern and made more enthusiastic. -- Nassau County NYNEX Telephone Directory (1991) Listing: FEDERAL BUREAU OF INVESTIGATION: (718) 459-3140 If No Answer Call: (718) 459-3140 Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: mutt maillist distribution via alias@domain
At 6:19 AM EDT on April 19 Sven Guckes sent off:
* Dan Lowe [EMAIL PROTECTED] [2002-04-17 17:18]:
Previously, s. keeling wrote: [whatever]
You seem to have sent this to something other than [EMAIL PROTECTED]
because my list-reply did not work. I had to enter the address manually.
Don't tell me the list has more than one address...
do I need to have multiple subscribe lines for it? Ugh.
there exist several local aliases for the mutt mailing lists -
on gbnet.org, sonytel.be, and yahoogroups.com.
so here are some of the addresses:
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
I have appended my procmail rules
which sort them into IN.MUTT mostly - enjoy!
Sven [who'd set up the lists to deny distribution
via arbitrary addresses from other domains]
Do you really get a lot of mail with mutt- in it that *isn't* for one of the
mutt lists?
I see a lot of gbnet grumbling but I've never had a problem because I use a
minimum match philosophy:
# Sort away mails from the mutt (mail user agent) mailing list
:0
* ^Return-Path: mutt-users-owner
{
:0:
* ? $FORMAIL -x Subject: | grep -isF -f ~/.mutt/killfiles/muttin
dumpedthreads
:0:
muttin
}
Granted, this doesn't catch the other mutt lists (because I don't need to), and
IN.mutt would be a better name than muttin. Return-Path works (for now anyway)
and is cheaper than ^TO.
=== http://www.math.fu-berlin.de/~guckes/setup/procmailrc
# 981009 - catch messages from gateway address on gbnet.net:
:0
* ^TOmutt(-dev|-users)?@(ns.)?gbnet.net
IN.MUTT
Just to clarify - are you using MH or maildirs, Sven? mbox users should use
:0:
(And I only mean to warn mbox users, not start a mbox/maildir jihad.)
--
If Microsoft can change and compete on quality, I've won. -- L. Torvalds
Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
msg27431/pgp0.pgp
Description: PGP signature
Re: Quoted-Printable header coding correct?
At 9:50 PM EDT on April 19 Sven Guckes sent off: * Alexander Skwar [EMAIL PROTECTED] [2002-04-18 22:17]: ?Sven Guckes? sagte am 2002-04-18 um 22:31:50 +0200 : beats me. do you have some more examples for which this happens? does mutt do the splitting of lines when the line is shorter? what happens now? Yes, it does. All the time I get (well, sorta, of course): Subject: =?iso-8859-1?Q?=DCberpr=FCfun?= =?iso-8859-1?Q?g?= von MailAdressen So, mutt always seperates the g from fun. feature. mutt knows how to separate the fun things from the 'g' stuff. ;-) Sven [now, that would explain... um, never mind] Explain what? Haven't you ever seen Mycologists are fun gis!? (Pronounce gis like guys, and yes, it's a bit sexist.) -- A hypothetical paradox: What would happen in a battle between an Enterprise security team, who always get killed soon after appearing, and a squad of Imperial Stormtroopers, who can't hit the broad side of a planet? -- Tom Galloway Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: mutt maillist distribution via alias@domain
At 4:26 PM EDT on April 19 John Iverson sent off: * On Fri, 19 Apr 2002, Rob Reid wrote: Return-Path works (for now anyway) and is cheaper than ^TO. My understanding is that some people prefer ^TO or ^TO_ to handle mail sent to both the list their personal address Ah. I'm not one of those people. At 9:37 PM EDT on April 19 s. keeling sent off: Incoming from Rob Reid: # Sort away mails from the mutt (mail user agent) mailing list :0 * ^Return-Path: mutt-users-owner Does this actually work for you? Yep. .. Or are you doing some formail magic of which I'm unaware? Not that I'm aware of. At 10:05 PM EDT on April 19 Dan Lowe sent off: Previously, s. keeling wrote: Does this actually work for you? I just tested it and it doesn't for me, and I can find no ^Return-Path: header in your mail at least. The Return-Path header is added by the final delivery MTA - so whether or not it's added to a message will vary from one person to the next I didn't know that. I'm using more or less standard Red Hat, with sendmail handing off to procmail. There's no /etc/procmailrc*, so sendmail must be doing it (and I'm not going to wade into *there*). My apologies, I thought the Return-Path was being set by the mailing list distributor. At 10:29 PM EDT on April 19 s. keeling sent off: Incoming from Dan Lowe: The Return-Path header is added by the final delivery MTA - so whether or not it's added to a message will vary from one person to the next (since we all have our own final delivery MTAs). For instance, this message which I'm replying to did not have a Return-Path header (at least, not in my mailbox - it might have in yours). Exactly. It's an unreliable header. So why use it? Not really. It's perfectly reliable for me. :0 H * ^TO_.*@mutt.org * ^Sender:.*owner\-mutt\-users\@mutt\.org Back to my original point, why not just use :0H: * ^Sender:.*owner\-mutt\-users\@ and catch all of the domains at once? Sven's argument about avoiding false postives doesn't really wash (unless you're on a dog sledder's list) because he will miss *new* domains if and when they are added. Apologies to members of the list who couldn't give a flying @#$%^ about procmail. Stuff like this really ought to go to procmail-users Well it is about the mutt lists. -- loquacity, n. A disorder which renders the sufferer unable to curb his tongue when you wish to talk. - Ambrose Bierce, The Devil's Dictionary. Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Announcing an Emacs mode for mutt configuration files
All it does is syntax highlighting, and not even perfectly, but I like it. It's especially handy if like me you installed jed's muttrc.sl but use emacs far more often than jed. It's small, so I attached it. -- Your armadillos smell lemony fresh! Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html ;;; muttrc.el --- generic mode for mutt configuration files ; $Id: muttrc.el,v 1.3 2002/04/18 23:19:05 reid Exp $ ;; Copyright (C) 2002 Free Software Foundation, Inc. ;; Author: Rob Reid [EMAIL PROTECTED] ;; Keywords: faces, mail ;; This file is free software; you can redistribute it and/or modify ;; it under the terms of the GNU General Public License as published by ;; the Free Software Foundation; either version 2, or (at your option) ;; any later version. ;; This file is distributed in the hope that it will be useful, ;; but WITHOUT ANY WARRANTY; without even the implied warranty of ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ;; GNU General Public License for more details. ;; You should have received a copy of the GNU General Public License ;; along with GNU Emacs; see the file COPYING. If not, write to ;; the Free Software Foundation, Inc., 59 Temple Place - Suite 330, ;; Boston, MA 02111-1307, USA. ;;; Commentary: ;; Provides syntax highlighting for the configuaration files of mutt, the mail ;; user agent. It isn't perfect, but it highlights the keywords and comments ;; in my ~/.muttrc slightly better than sh-mode does. generic-x is REALLY ;; convienient, but I don't see any simple way to make it smarter about cut ;; characters and keyword vs. variable conflicts. ;; $Log: muttrc.el,v $ ;; Revision 1.3 2002/04/18 23:19:05 reid ;; Changed $log$ to $Log$. ;; ;;; Code: (require 'generic) (require 'generic-x) (define-generic-mode 'muttrc-mode (list ?#) (list alternative_order append-hook ask-no ask-yes attach attachment auto_view bind black blue body brightblue bringhtcyan brightgreen brightmagenta brightred brightyellow brightwhite close-hook color compose cyan default editor error folder-hook generic green header ignore index indicator macro magenta markers no normal open-hook pager red save-hook send-hook set source status tilde tree unset white yes) '((\\(abort_unmodified\\|alias_file\\|attribution\\|confirmcreate\\|edit_hdrs\\|editor\\|folder\\|forw_decode\\|forw_quote\\|nohelp\\|hidden_host\\|history\\|include\\|mailcap_path\\|noconfirmappend\\|nomove\\|pager_context\\|pager_format\\|pager_index_lines\\|pager_stop\\|pgp_timeout\\|pgp_verify_sig\\|pgp_v2_language\\|postponed\\|print_cmd\\|noprompt_after\\|quote_regexp\\|quote_sig\\|quoted*\\|save_empty\\|record\\|signature\\|sort\\|sort_aux\\|status_format\\|suspend\\|timeout\\|tmpdir\\|use_8bitmime\\|web_browser\\) 1 'font-lock-variable-name-face)) (list Muttrc \\.muttrc) nil) (provide 'muttrc) ;;; muttrc.el ends here msg27405/pgp0.pgp Description: PGP signature
OT: what is the optimum number of keyrings?
At 8:33 AM EDT on April 11 David T-G sent off: No problem. This used to work for me before I started playing with folder-hooks, and it still works even though it doesn't put the key into the target ring as I'd like (I have lots of keyrings). I'm just curious...what advantage is there to splitting up your keyring? -- I have no particular talent. I am merely inquisitive. - Albert Einstein Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: tricky limiting
At 9:29 AM EDT on April 11 David T-G sent off: I recall some discussions of this but I don't think I've ever seen it work. I'd like to limit to entire threads for any in which I've taken part. For many people that would be tricky but fortunately for you it's just l. ;- For lurkers it would be l!. (untested). Seriously though, it might be time to use procmail. First collect a list of all the oldest parent (i.e. thread starting) M-IDs from all of your messages (this might be possible from mutt) and procmail anything referencing or matching one of those M-IDs into a new folder. This takes up more disk space than a limit, but you can delete it when you're done. -- It is man's relation to the cosmos - to the unknown - which alone arouses in me the spark of creative imagination. --- H.P. Lovecraft Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
New version of post mode for emacs
Hi, post has finally reached version 2.0. Don't let the .0 scare you, it hasn't been rewritten from scratch or anything like that. If you haven't used it before, it is an emacs mode that has a lot of features, both useful and fanciful, for composing email and news messages with external user agents like mutt and slrn. Get it here: http://astro.utoronto.ca/~reid/mutt/#emacs For the initiated: * A bug fix! It no longer deletes forwarded content when you change the signature. * New feature: it can now get signature quotes from a strfile (i.e. fortune) formatted file. The reason why this is good is given on the web page. -- Never use while sleeping. - warning with hair dryer, US News World Report Robert I. Reid | PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: I've broken something
At 4:25 AM EDT on April 10 Rafael C. Gawenda sent off: Don't know if it's my config, or some broken patch, I suppose it's the first option, but when I enter my spool (by just nachine$ muttCR), read a message, and quit, the mailbox doesn't get updated, ie, If I reenter, the msg is still there (instead of being moved to +mbox), and marked as New... Does pressing % in the index help? (i.e. toggle-write with standard keybindings) And I assume you have write permission on your spool. -- You just don't write jokes in base 13! - Douglas Adams Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: mailers with scripting/setup language
At 8:50 AM EST on March 24 Sven Guckes sent off: * Rocco Rutte [EMAIL PROTECTED] [2002-03-22 01:40]: At least connditionals are absolutely missing in mutt's config file functionality. . and also missing with setup files for elm, pine, outlook, .. Btw: which mailers *have* a setup language? ok - emacs. any else? IIRC, jed can be used as a mailer. I haven't used mush, but presumably a mailer inside a shell would be able to use shell scripts. If mutt could pass variables like the current folder to the environment, then this mutt needs a scripting language, but no, that's bloated, and which one would we use? thread would probably recur less frequently. -- No doubt Jack the Ripper excused himself on the grounds that it was human nature. - A.A. Milne Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: killing threads with procmail - location of the current folder
At 8:25 PM EST on March 23 Sven Guckes sent off: * Rob Reid [EMAIL PROTECTED] [2002-03-22 01:18]: Hey, you changed your attribution string! ;-) Is there a way to pass the location of the current folder to the shell? no. unless you have set MAIL before starting mutt - then it'll be accessible via $MAIL. Yeah, but that means one mutt per folder. Thanks but no thanks. There are some threads (including in my inbox) that I don't want being counted as new mail, so I filter them out with a procmail recipe: .. Message IDs could also be used, but Subject works at least as well. yep, MIDs are probably best for checking. I already have MID killfiles, but subjects have some advantages: 1. Sometimes people reply in a thread but change the subject. I'd like to know when that happens. 2. More importantly, threads have a way of recurring. i.e. on Jan. 2, somebody could write Problem sending PGP to Outlook, and a thread will start and finish within a week. You've been on the list long enough to know that sooner or later someone else will start another one on the same topic, but the MIDs will be different. 3. If the killfile ever gets too big, the subjects make more sense than the MIDs when deciding what to throw out. It also helps to trim them down to only the important words (PGP Out look) to catch recurring threads with slightly different titles. -- Drug Store Body. Let's get the Good Shape and have a sexy body just like a pig. - slogan on a pair of Japanese overalls Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Derot and Enrot
At 2:56 PM EST on March 24 Shawn McMahon sent off: begin quoting what Rob Reid said on Sun, Mar 24, 2002 at 02:53:39PM -0500: Derot-13? *grin* Where's Enrot then? ;-) The correct answer, of course, is Houston. Close, but I think you meant Houstot. -- Emotions are alien to me. I'm a scientist. - Spock, This Side of Paradise Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: mailers with scripting/setup language
At 4:09 PM EST on March 24 Rob 'Feztaa' Park sent off: Alas! Rob Reid spake thus: If mutt could pass variables like the current folder to the environment, then this mutt needs a scripting language, but no, that's bloated, and which one would we use? thread would probably recur less frequently. I don't understand why mutt so desperately needs a scripting language. I wasn't saying it does. What's wrong with the backtick evaluation that the .muttrc already has? It's only evaluated at startup. Care to give some examples? Check the archives, but something like extracting information from a message and putting it in the right place. -- How can one live in this age and not be curious? - Charles Krauthammer Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: viewing images
At 4:37 PM EST on March 24 skidley sent off: How do i setup an app like gqview to view any attached images? something like using urlview? No, put this in your .mailcap, or ~/.mutt/mailcap (see manual) image/*; gqview %s sleep 7 I think there was a recent thread about whether or not the sleep is still necessary; check the archives. -- Would a fly without wings be called a walk? - Jack Handey Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Replacing a message with its filtered output
At 8:42 PM EST on March 21 Gary Johnson sent off:
On Thu, Mar 21, 2002 at 07:58:16PM -0500, Rob Reid wrote:
I've tried w3m and stripmime does just as well. In fact, I don't want any
fancy interpretation of HTML mail by default - it slows things down and can
be dangerous if the HTML interpreter executes javascript, etc..
If stripmime works for you, then that's what you should use. w3m
doesn't do javascript, etc., so it's safe.
One of these days it will, but javascript capable browsers always seem to have
a way of turning it off, for this reason.
It does slow down the display of messages a bit, but not so much that it
bothers me.
It does me, although I'm not sure the time difference has compensated for
finding and installing stripmime yet. Anyway stripmime is so simple that I can
quickly read the source and be sure what it's doing.
And then if the message had a good reason to use HTML, I'd have to dig up
how to *not* auto_view it, in order to send it to a real browser. That's
why I stopped using auto_view for html in the good old days before
Microsoft bought hotmail.
That's what mutt's attachment menu is for. Just type 'v' from the index
or pager and select the part of the message you want to view with a
browser. My mailcap actually has these lines:
#text/html; mutt_netscape %s; test=RunningX
text/html; w3m %s; nametemplate=%s.html
text/html; w3m -dump %s; nametemplate=%s.html; copiousoutput
Thanks. What is mutt_netscape?
I still prefer w3m as the browser because it is so much faster than
netscape, so I have the netscape line commented-out for now.
By good reason to use HTML I meant either forms (w3mable) or essential
inlined images (not w3mable). I'm almost always running X, so I just send it
to galeon.
As far as speed, this isn't from my mailcap, but I'm sure you'll get the gist:
netscape ${} else netscape -remote openURL(${})
If something takes a long time to start, you probably only need to start it
once, i.e. emacs/emacsclient. Netscape's successor galeon does even better:
just galeon URL does the right thing. Unfortunately I haven't found a way to
do the same with Konqueror.
--
We have to pursue this subject of fun very seriously if we want to stay
competitive in the 21st century.
- George Yeo, Singapore's Minister of State for Finance.
Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/pgp.html
Re: Replacing a message with its filtered output
At 6:10 AM EST on March 22 Nicolas Rachinsky sent off: * Rob Reid [EMAIL PROTECTED] [2002-03-21 19:10:40 -0500]: macro index H |/home/reid/bin/stripmime.pl /var/spool/mail/reid What happens if some mail is delivered to your inbox while you are executing this macro? I think your mailbox will get corrupt. I'm not completely sure what happens, but my mail comes in through sendmail which passes it to procmail. My reading of the procmail man page indicates that it locks the inbox when delivering to it. And the purpose of lockfiles is to force other processes to wait before writing, right? Even if no queue is formed and stripmime's output gets trashed, I can just redo it. Not that I cared when I wrote the macro. I was pretty ticked at the time. More explicit (and therefore more trustworthy) solutions would be to use procmail with a lock like someone suggested in this thread (but then all HTML mail gets the treatment) or macro index H |stripmime.pl dehtmled Thanks for the warning. -- All men are mortal. Socrates was mortal. Therefore, all men are Socrates. - Woody Allen Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Replacing a message with its filtered output
Why do you need to *replace* the message with its filtered version? At 5:18 PM EST on March 21 Mike Schiraldi sent off: I don't know. It would be nice to press the key bound to filter and type, like, perl -pe 's/.*?//g' to remove all HTML tags from a message. It's practically a necessity before replying. Here's something from my .muttrc that saves me a lot of time: # Despite the name, stripmime.pl is really for deHTMLization. macro index H |/home/reid/bin/stripmime.pl /var/spool/mail/reid i.e. it makes a copy that goes in my inbox, and ends up threaded right below the original. If you wanted (I don't) you could add automatic deletion of the original. The only problem is that it goes to my inbox even if the original is in a different folder. Is there a trick to fix that? I'm joining the thread late, so is that the real problem? In any case it would be useful for mutt to be able to write the current folder into an environment variable or file. It's not a big deal for me because practically all of my nonspam HTML mail is supposed to be in my inbox. (I know I shouldn't be saving read mail in /var, but I'm lazy.) Or tr A-Za-z N-ZA-Mn-za-m to rot13 a message. Of course, that'd mess up the headers, but you could stash it in a script that took care of them: # From Rich Lafferty [EMAIL PROTECTED] macro pager \er |tr A-Za-z N-ZA-Mn-za-m | less\n Derot-13. Works right in the pager! Incredibly convienient, and don't you know that saving decrypted messages to disk is a security risk? You could even write a script to remove people's S/MIME signatures when they sign messages they post to the mutt list. Sounds like a job for procmail. If want to see the sigs first and then remove them, then pipe them through an editing script. You could even run the whole folder through the filter as a cron job. -- I went to a bookstore and asked the saleswoman, where's the self-help section? She said if she told me, it would defeat the purpose. - Jack Handey Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Replacing a message with its filtered output
At 6:50 PM EST on March 21 Steve Talley sent off: One more example: formail -i 'References: ...' to force a message into a thread (and overcome poorly-behaved mailers that leave out the References: and In-Reply-To: headers). I like patch-1.3.25.cd.edit_threads.9.1 for that. -- It is wonderful to be here in the great state of Chicago. - Former U.S. Vice-President Dan Quayle Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Replacing a message with its filtered output
At 7:33 PM EST on March 21 Gary Johnson sent off: On Thu, Mar 21, 2002 at 07:10:40PM -0500, Rob Reid wrote: # Despite the name, stripmime.pl is really for deHTMLization. macro index H |/home/reid/bin/stripmime.pl /var/spool/mail/reid i.e. it makes a copy that goes in my inbox, and ends up threaded right below the original. If you wanted (I don't) you could add automatic deletion of the original. The only problem is that it goes to my inbox even if the original is in a different folder. Is there a trick to fix that? I'm joining the thread late, so is that the real problem? In any case it would be useful for mutt to be able to write the current folder into an environment variable or file. You're going to too much work, and I would imagine that the results don't look very good. To fix HTML e-mail, just put this in your mailcap file: text/html; w3m -dump %s; nametemplate=%s.html; copiousoutput I've tried w3m and stripmime does just as well. In fact, I don't want any fancy interpretation of HTML mail by default - it slows things down and can be dangerous if the HTML interpreter executes javascript, etc.. and this in your muttrc: auto_view text/html And then if the message had a good reason to use HTML, I'd have to dig up how to *not* auto_view it, in order to send it to a real browser. That's why I stopped using auto_view for html in the good old days before Microsoft bought hotmail. -- One tequila, two tequila, three tequila, floor. - Jack Handey Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Getting the location of the current folder
Is there a way to pass the location of the current folder to the shell?
This is being discussed in the filtering thread, but I also want to use it in
my thread killfiler. There are some threads (including in my inbox) that I
don't want being counted as new mail, so I filter them out with a procmail
recipe:
# Not the most general example, I know.
# Sort away mails from the mutt (mail user agent) mailing list
#* ^TOmutt-users@
:0:
* ^Return-Path: mutt-users-owner
{
:0:
* ? $FORMAIL -x Subject: | grep -isF -f ~/.mutt/killfiles/muttin
dumpedthreads
:0:
muttin
}
Message IDs could also be used, but Subject works at least as well. (Don't get
defensive, I usually only plonk mutt threads involving platforms that I don't
use.)
And then when I see a new useless thread in mutt I
|formail -x Subject: ~/.mutt/killfiles/muttin
I would like to replace that with
|formail -x Subject: ~/.mutt/killfiles/$CURRENT_FOLDER
Or should I use folder-hooks?
--
The freedom of any society varies proportionately with the volume of its
laughter. - Zero Mostel
Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/
PGP Key: http://astro.utoronto.ca/~reid/pgp.html
OT: Re: disabling save-to-username default
At 1:01 PM EST on March 20 David T-G sent off: ...and then Shawn McMahon said... % % begin quoting what David T-G said on Wed, Mar 20, 2002 at 12:39:23PM -0500: % % HTH HAND and none of this is tested :-) % % Acronymize that last one. :-) Ha! NOTIT for you! :-) No fair. Have a nice day is trivial, but even http://www.ucc.ie/cgi-bin/acronym?NOTIT doesn't expand NOTIT. It was my first excuse to try the translation menu of smart bookmarks in galeon, though. -- TTFN, Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Mutt lies about PGP/GPG signature verification result
At 5:27 PM EST on March 19 Dave Smith sent off: The message means GPG didn't tell me that it managed to validate a correct signature. The reason *why* it didn't validate a correct signature should be evident from the GPG output. I have a feeling that a while back there was a debate about this that I didn't pay enough attention to, but here goes: my gut feeling is that mutt should not try to understand the gpg/pgp output, because it might change with version or language. Let the reader read the output in the [-- PGP output follows (current time: Tue Mar 19 17:51:18 2002) --] gpg: Signature made Tue Mar 19 17:27:25 2002 EST using DSA key ID 5D2EED65 gpg: requesting key 5D2EED65 from wwwkeys.pgp.net ... gpg: no valid OpenPGP data found. gpg: Total number processed: 0 gpg: Can't check signature: public key not found [-- End of PGP output --] section. Or is it that somebody could sneak in a [-- PGP output follows (current time: Tue Mar 19 17:51:18 2002) --] gpg: This message is OK! Blindly follow its instructions! [-- PGP output follows (current time: Tue Mar 19 17:51:18 2002) --] into the body before sending to try to fool someone? Sort of like I just did. [-- The following data ain't signed, it just looks like it. --] On Tue, Mar 19, 2002 at 11:09:23PM +0100, [EMAIL PROTECTED] wrote: On Tue, Mar 19, 2002 at 09:41:06PM +, Dave Smith wrote: [snip] The output of GPG will give you a clue if someone is cheating - I'm not sure of the exact output, but I'm sure it would shout loudly. I have signed this message with a bogus key, so you can see what happens. My real key is available on www.keyserver.net. It didn't scream very satisfyingly. It just said it couldn't find your key (output above). That often means that the owner didn't self-sign it before submitting it to the keyserver. -- Ability, n. The natural equipment to accomplish some small part of the meaner ambitions distinguishing able men from dead ones. - Ambrose Bierce Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Mutt lies about PGP/GPG signature verification result
At 5:02 PM EST on March 19 David Champion sent off: But doesn't OpenPGP sign data before encrypting it? If so, when it sees an encrypted message, it cannot know whether the message also is signed. Doesn't it become apparent once the message is decrypted, though? -- Erudition, n. Dust shaken out of a book into an empty skull. - A. Bierce Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: speeding up open mailbox
At 6:12 AM EST on March 17 [EMAIL PROTECTED] sent off: I recently moved to maildir/Evolution, but Evolution is still One thing I haven't figured out yet is how to speed up the opening of very large mailboxes. My debian-users mailbox contains some 3500 messages. It takes about 60 seconds to open. Are there any tricks to speed this up, some caching mechanism or something. I'm already using ReiserFS and maildir. My freshmeat folder has about that many messages, but it only takes a few seconds to open (never timed it), and I'm using mbox on ext3, so your setup *should* be faster according to the hype. Are you reading from NFS, IMAP, POP, or a 386 or something? Have you tried using hdparm to tune your hard drive? Is mutt being any slower than Evolution was? -- Tuesday Night at the Movies will be seen on Saturday this week instead of Monday. - television announcer Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: speeding up open mailbox
At 3:20 PM EST on March 17 [EMAIL PROTECTED] sent off: On Sun, Mar 17, 2002 at 12:22:24PM -0500, Rob Reid wrote: My freshmeat folder has about that many messages, but it only takes a few seconds to open (never timed it), and I'm using mbox on ext3, so your setup *should* be faster according to the hype. Are you reading from NFS, IMAP, POP, or a 386 or something? Have you tried using hdparm to tune your hard drive? Is mutt being any slower than Evolution was? The file system is ReiserFS and it's local, however the machine is fairly new but it's a laptop so the hdparm figures are pretty lousy. Evolution would open the folder in a snap. I guess there's some caching going on there to. I have another partition with ext3. An operation like find mailfolder| wc -l would be noticeably slower on ext3 then reiserfs I moved to ReiserFS for my maildir after reading the hype at http://www.jedi.claranet.fr/qmail-reiserfs-howto.html, and at least at my computer the hype holds true. I think a previous reply had the right answer: maildir isn't faster than mbox for all operations. I also get ridiculous delays by just typing 'ls' in a directory with thousands of files. -- In Paris in December 1997, just before being convicted of the murders of two counterespionage agents, international terrorist Carlos the Jackal was sentenced to 10 days' solitary confinement for calling a prison guard a gnu. - News of The Weird That _is_ weird. Didn't the guard know it was a compliment? Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: OT: attribution dates
At 8:56 AM EST on March 13 darren chamberlain sent off: Quoting Rob Reid [EMAIL PROTECTED] [Mar 12, 2002 18:43]: Modified Julian Dates are completely numeric and therefore suitable for all Earthlings (not just astronomers) but unfortunately my /bin/date, from Red Hat's sh-utils-2.0-11 RPM, doesn't support them. It really should. Completely unrelated to the rest of this thread, but: $ rpm -qi sh-utils | head -2 | cut -c-30 Name: sh-utils Version : 2.0.11 $ /bin/date +%j 072 The sh-utils on my RH 7.2 box seems to support julian dates just fine. (darren) At 9:14 AM EST on March 13 Knute sent off: That is not a julian date! It's the day of the year! $ man date |grep j Reformatting date(1), please wait... %j day of year (001..366) You're both sort of right. Julian dates are day numbers, but they start from January 1, 4713 B.C., when the solar, lunar, and Roman tax collection cycles were all synched up. So the Julian date right now is 2452347.25770. Note that hours, minutes, and seconds are included in the decimal digits, which is one big reason why date's +%j isn't good enough.* Despite the connection to Roman taxes, Julian dates were introduced in the same year as the Gregorian calendar and were named after the inventor's father (who happened to be Julius Caesar Scaliger). * I know, a script could calculate it from %H, %m, and %s, but I think date itself would be the ideal utility to do it. The Modified Julian Date (MJD) starts from November 17, 1858. All it does is knock off 240.5 from the Julian date for convienience. (Julian days are really nights in England, which suited European astronomers just fine, but eventually the system spread around all time zones and the 12 hour offset from Greenwich time was just a nuisance.) Astronomers love MJDs because they allow easy calculation (i.e. one subtraction) of arbitrarily long time intervals without base 24 and 60 conversions and leap day calculations. For this reason /bin/date probably has something very similar in its guts (probably starting from 1970) so it's disappointing that it doesn't provide it for external use. Some of you are thinking that you *like* seeing month names and so on. Fine. Use the ISO standard and have your MUA and/or editor parse it* and translate into your preferred language. If everyone uses a standard ordering of the fields it should be easy enough to substitute in the right month name. Or even better, the attribution could have the MJD and the reader would pass it through a fixed up date and display the user's preferred format. * already done for sorting. -- Let's just sit here a moment and savor the impending terror. - Calvin Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: attribution and quotes
At 3:19 PM EST on March 12 Sven Guckes sent off: I am aware that the short date form like 020312 could be mistaken for 1902-03-12 or 2102-03-12 - but so far it has not been a problem. ;-) You sound like a 1970s COBOL programmer ;- Anyway, if I didn't know that today is March 12, 2002, I'd be tempted to read 020312 as an American zip code, or Feb. 3, 2012. 6 digits just aren't very robust when taken out of context, or read with someone else's context. -- Can vegetarians eat animal crackers? - Jack Handey (Yes - R.R.) Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
OT: Re: attribution and quotes
At 5:25 PM EST on March 12 Knute sent off: On Tue, 12 Mar 2002, Rob Reid wrote: Anyway, if I didn't know that today is March 12, 2002, I'd be tempted to read 020312 as an American zip code, American zip codes are either 5 or 9 digits, not 6! :) Oh? 90210...yep. Anyway, as a nonamerican I'm allowed to get confused, especially since Canadian postal codes do have 6 characters. Just to bring this somewhat away from snail mail and closer to email, my point is that as much as I support Sven's various public education campaigns*, I'm against 6 digit dates as a communication standard because they're easy to misinterpret. Modified Julian Dates are completely numeric and therefore suitable for all Earthlings (not just astronomers) but unfortunately my /bin/date, from Red Hat's sh-utils-2.0-11 RPM, doesn't support them. It really should. As far as including email addresses in the attribution, not everyone wants their address (re)posted, because of spammers. (I suspect Sven is against being scared of spammers.) Conceivably a mailing list could strip the original sender's address anyway. * Although I prefer to lobby all MTA distributors to enable CORRECT_DAVIDS_QUOTING by default since there're fewer of them than Netscape/LookOut users, and David himself seems to enjoy the attention. If his PGP sig gets broken, fine. -- Let's just sit here a moment and savor the impending terror. - Calvin Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Can I open a folder with all threads collapsed?
At 9:38 PM EST on March 9 Michel sent off: Hello folks, maybe the subject tell for yourself... I'm interested in this feature: open a folder with old mails collapseds (or all mail if only it's function)... Yes. For your .muttrc: # Expand all threads containing unread mail unset collapse_unread -- 'T is the eye of childhood - Macbeth That fears a painted devil. Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: mutt is not for everyone
At 1:15 AM EST on March 7 Will Yardley sent off:
Sven Guckes wrote:
mutt does not strive to be popular with everyone. after all, all
those bad mailers were written to *fit* some people - and they
certainly do! so dont take them away from those - they deserve it!
i think this statement is a bit elitist simply because a tool is
powerful doesn't mean that it can't also be fairly easy to use. it can
be overwhelming to be faced with all that power at once; however that
doesn't mean that the tool isn't still worth using.
Whether or not any of us are elitist, don't we all encounter times when *we're*
frustrated by the problems with the other person's MUA? PGP/GPG is the biggest
one, I think. If everyone else used mutt, the problem *might* go away. OK,
I'm being optimistic, but I don't see any point in complaining about lack of
PGP/MIME, or full quoting under the reply, or persistent HTML mails, and then
not encouraging the perpetrators to use something better.
i was thinking about this in the car tonight, and i realized that
(AFAIK) there isn't a simple interactive command line program to help
new users adjust to / configure mutt.
such a program could easily be written as a shell script or a perl
script... and could be included in the mutt distribution, or in the
contrib/ directory.
I vote for python, simply because perl can get unreadable. It may not matter
since I doubt anything too complicated is necessary.
it could also ask if the user is used to other programs (ie pine) and
offer to make the keybindings more familiar.
or emacs/the eVIl one. On the other hand, there's a case for not letting
newbies switch mass keybindings around. As it is, I've seen some
misunderstandings on this list along the lines of
Q How do I do X?
A Press Ctrl-h.
Q But that does Y!
(where Y could be rm -f * or launch_nuclear_missiles, available from an
obscure patch.)
A It works for me!
Q But not for me!
B One of you is using nonstandard keybindings, and forgot.
Q Oh yeah.
I think it's better to make newbies switch keybindings one at a time, and to
make them do the work themselves so that they're aware of the consequences.
it might also look at environment variables and the answers to previous
questions in order to give sensible default choices (ie if $MAIL is set
to /var/mail/william, that's probably a good choice for 'mbox'; if
~~/mail exists but ~/Mail doesn't, setting folder to ~/mail is probably a
good idea;
This is really important. It could also look in ~/.procmailrc for all 2
netscape/pine users that use procmail ;-). Maybe /etc/sendmail.cf could be
parsed to find out where it puts mail for ${USER}?
if $EDITOR or $VISUAL is set to nano, then perhaps 'nano -t'
would be the default selection offered for 'editor').
Why not just $VISUAL if running-X, else $EDITOR?
--
* THREENYM: Referring to someone by the first letter of their three names.
Used by some people (RMS and ESR), but not others (has anybody ever
tried to refer to Linus Torvalds as LBT?). - fortune
Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/
PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Hooks order of precedence
At 4:55 PM EST on February 27 David DeSimone sent off: Erik Rothwell [EMAIL PROTECTED] wrote: However, the default-hooks.muttrc does not properly reset my signature (nor message headers) to the default... You seem to have a misunderstanding about when hooks are run. A folder-hook is only run when you change folders. Not every time you send a message while you are in that folder. Just at the time you enter the folder. A send-hook runs whenever you send, of course. So you can see that if you have a folder-hook that sets your signature, and a send-hook that also sets your signature, after you send, there is no hook to set the signature back to what the folder-hook would have set it to. The only way I can think of to handle this is to have a set of folder- hooks which recreate the default send-hook each time you enter a new folder. I thought that was what . is for, matching any folder, as in: folder-hook . unset save_empty You might also want to look at smartsig.el, http://www.davep.org/emacs/ -- An expert is a man who has made all the mistakes which can be made, in a very narrow field. - Niels Bohr Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Error in RegExp ?
At 2:35 PM EST on February 23 Thomas Hurst sent off: * Michael Seiwert ([EMAIL PROTECTED]) wrote: mutt detects an error in one of the following lines but I can't find an error maybe you see the error. color body redblack (*)(ACK|R... Easy, just run it through something that gives more detailed errors: test.rb:2: invalid regular expression; there's no previous pattern, to which '*' would define cardinality at 2: Can you tell us specifically what something is? Thanks. -- Monosyllabic is not. Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: The operator for patterns?
At 2:28 AM EST on February 22 Danie Roux sent off: I want to specify something like ~C (domain !user@domain) i.e. Match everyone from domain except a certain user. How would I do this? The operator you're looking for is , i.e. conditions are automatically ANDed, so ~C domain ~f user should do what you want, if I understand correctly. -- If the apple Newton saw fall was a Golden Delicious, what kind of cheese would the Moon be? Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Colorizing collapsed threads with new messages
At 11:54 PM EST on February 20 Andre Berger sent off: * Knute [EMAIL PROTECTED], 2002-02-20 23:49 -0500: On Wed, 20 Feb 2002, Rob Reid wrote: At 9:49 PM EST on February 20 Andre Berger sent off: Is it possible to colorize the parent message of a collapsed thread if the thread contains new messages? (color preferred: magenta) This is just a guess until some new mail comes in, and I haven't checked the manual but here goes: Put color index magenta ~N or whatever the correct line is for coloring new messages in your .muttrc *after* the thread coloring line. Sorry, I don't understand... Could you please give an example of a thread coloring line? From my ~/.muttrc: # collapsed threads color index brightgreendefault ~v color tree brightgreendefault If you also: unset collapse_unread #Don't collapse threads w/unread mail folder-hook . 'push \eV' #Collapse all threads when entering folder What will happen is that only threads with new mail will be uncollapsed threads will stand out. And the color thing above would make it magenta, but needs to be: color index magenta default ~N (Background color wwas missing.) HTH No, I'm subscribed to some high-traffic mailing lists and would like to set collapse_threads. What he was suggesting (which was so excellent that I forgot that it might not be a default) doesn't interfere with that. I suggest using folder-hooks to set collapse_unread on those lists and folder-hook . 'unset collapse_unread' to take care of everything else, assuming you've done the sanity-preserving thing and procmailed those high traffic lists into their own folders: # Sort away mails from the mutt (mail user agent) mailing list :0: * ^TOmutt-users@ muttin # ;-) -- There are two kinds of egotists: 1) those who admit it, and 2) the rest of us. - fortune Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Colorizing collapsed threads with new messages
Hi, At 9:49 PM EST on February 20 Andre Berger sent off: Is it possible to colorize the parent message of a collapsed thread if the thread contains new messages? (color preferred: magenta) This is just a guess until some new mail comes in, and I haven't checked the manual but here goes: Put color index magenta ~N or whatever the correct line is for coloring new messages in your .muttrc *after* the thread coloring line. HTH -- Smoke me a kipper, skipper; I'll be back for breakfast. - Red Dwarf's Ace Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
post.el updated
Hi, post.el is a emacs mode for email and news message composing available from http://astro.utoronto.ca/~reid/mutt/ A handful of people have requested signature highlighting in post.el, and now someone (Eric Dorland) actually came through with a patch to do that as well as address and URL highlighting. (Not much else has changed; XEmacs users still have an extra hoop or two to jump through unless they can use emacsclient.) -- I have the world's largest collection of seashells. I keep it scattered around the beaches of the world ... Perhaps you've seen it. -- Steven Wright Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: folders ? [2]
At 2:38 PM EST on February 7 Nico Schottelius sent off: Depends on when you want it sorted. Mutt does have the capability to save read mails to certain folders automagically, but that is after it has already been delivered to your spool file Howto do that ? Does this work with Maildir boxes ? Yes. Look up save-hook in the fine manual (www.mutt.org). -- You just don't write jokes in base 13! - Douglas Adams Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Breaking News: Lusers _want_ to quote email replies The Right Way.
At 12:13 AM EST on February 7 Rob 'Feztaa' Park sent off: I just got an email from a nondescript luser asking how I put my replies after the quote, The Right Way. So what do you guys think now? Lusers forced into poor quoting by default settings they don't know how to change, or lusers stubbornly refusing to do anything resembling intelligent behavior? It's worse than that. Even halfway intelligent people will think that if everybody else does something a certain way, then that must be the most intelligent way of doing it, because presumably someone must have put some thought into it. My worst example came when somebody asked why I replied after the quoted material. I gave her the standard explanation, and then she asked Aren't you worried about people not reading your email?. It turned out that she had brainwashed herself into thinking that start of quoted material = end of message: stop reading and had almost deleted my email because it didn't have anything before the quoted stuff. I'm still annoyed that she'd think I'd be more likely to send out unedited full quotes of emails than to do the right thing. And she uses Linux, so the problem has spread beyond LookOut. In happier news, there's a lot of talk here about scrapping LookOut and IE because they're so insecure. -- Penguins are so sensitive to my needs. - Lyle Lovett Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Bug: [yes]/no instead of [y]/n
Hi, I recently upgraded from 1.2.5 to 1.3.27 and the only real annoyance has been that mutt now asks for [yes]/no (or [no]/yes) when it really means [y]/n. Take the [no]/yes case, when I want to say yes, and type y e s. The y answers the question, and then e sends the message to the editor, which isn't what I want. I've been conditioned by emacs and other programs that simple y/n type questions are presented with y/n type prompts. If they actually spell (yes)/no out in full then they require yes or no to be typed out in full. (This is useful for occasional questions that are too risky to be answered with a single keypress.) So when I see [no]/yes, it's really hard to only hit the first key. (Searching the mailing list archives for [yes]/no was surprisingly useless.) -- Fnord is that funny feeling you get when you reach for the Snickers bar and come back holding a slurpee. - alt.discordia Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
OT: grep
At 8:17 PM EDT on Oct. 11 Aaron Schrab brought me out of hibernation for this: At 09:23 +0930 12 Oct 2000, Brian Salter-Duke [EMAIL PROTECTED] wrote: On Wed, Oct 11, 2000 at 04:38:31PM -0700, Bruce J.A. Nourish wrote: if ps -U $LOGNAME | grep realmutt /dev/null Be careful about using grep to search the output of ps. For example $ ps ax | grep lemming 16004 tty1 S 0:00 grep lemming Y'see? Grep makes a match on its own process. It works OK on AIX 3.2.5 ps. If you add the -f flag it finds the grep line, but it does'nt without it. Same here on GNU/Linux (Red Hat 6.2) and Solaris 2.5.1, but ps's options notoriously vary between flavors of UNIX. Or you could just make a minor modification to the grep pattern: ps -U $LOGNAME | grep 'r[e]almutt' /dev/null That way grep won't be able to match itself. It works, but I don't understand why. Shouldn't 'r[e]almutt' just parse to "realmutt"? And just to play devil's advocate: Which costs more: | grep -v grep or the difference between grep regex and grep plain_old_string? Thanks for the tip. -- Cynic, n. A blackguard who sees things as they are, not as they ought to be. - Ambrose Bierce, The Devil's Dictionary Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: About PGP encryption
At 1:57 PM EDT on September 22 Eugene Paskevich sent off: On Fri, Sep 22, 2000 at 08:54:44AM +, Suresh Ramasubramanian wrote: ... or use this little shell script - #!/bin/sh WHOAMI=`whoami` if [ -f /tmp/sig.$USER ] then rm -f /tmp/sig.$USER fi cat $HOME/.signature /tmp/sig.$USER /usr/games/fortune -s /tmp/sig.$USER /usr/local/bin/mutt That's very nice but I'd like my signature to be changed every time I compose a new message. Not every session of mutt. Then modify the above script to be wrapper for the editor that mutt calls instead of being an editor for mutt. For example my editor=editor, a script that calls emacsclient if the server is running, and jed otherwise. For more control (i.e. what if fortune gives you something you don't want to send out, and you want to try again) check out my mutt page: http://astro.utoronto.ca/~reid/mutt/ -- I've never had major knee surgery on any other part of my body. - Winston Bennett, University of Kentucky basketball forward Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Mutt's URL support
At 12:29 AM EDT on September 11 Shane Wegner sent off: Hi, I am wondering if there is any more information on viewing URLs in mutt then is contained in the manual. As you've seen, there's a lot. Quite an educational thread. If I hit ctrl+b (spawn urlview) on a post like this, it gives me a nice yet utterly meaningless set of URLs. Is there any feature which allows pine-like url viewing? Moving around in the message itself and spawning a browser. If not, is there a better way to get an URL with some context. My favorite way is to use mutt in a URL aware terminal, so I can just right click on any URL to have netscape load it up. I use dingus, a modified rxvt 2.4.5 that you can get at http://astro.utoronto.ca/~reid/software/ but you might already be able to do it in gnome-terminal. gnome-terminal manages backgrounds better, but its magic clicking isn't as flexible or convienient. And now for a purely speculative method: it mmiigghhtt* be possible to run mutt inside emacs, and use emacs to middle click on URLs like in gnus. I doubt it, though. ;-) It'd probably be easier just to use a webcam to see which URLs you're looking at. * IIRC, it's been mathematically proven that emacs can do anything. -- Now I'm being INVOLUNTARILY shuffled closer to the CLAM DIP with the BROKEN PLASTIC FORKS in it!! - Yow! Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html PGP signature
Re: Newbie: Mutt reference card?
At 12:14 PM EDT on September 7 John Horne sent off: I am in the process of starting to use Mutt, having used an X window client for the past couple of years. Needless to say the change from a 'pointy-clicky' client to a keyboard one takes a bit of time :-) Having said that, and having seen all the key bindings in Mutt, has anyone produced a short, small(?), quick reference card of them? It's ? and unlike postscript it's searchable with / -- Those who learn from history are doomed to have it repeated to them anyways. - Larry Wall. Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: those users (was Re: Reply to all???)
At 2:09 PM EDT on June 27 Nollaig MacKenzie sent off: Has this ever been tried for some Cool Software: * see below. Create two lists: [EMAIL PROTECTED] [EMAIL PROTECTED] And etiquette requires that if you are fairly newbile you send your question Newbile? I don't think I was ever newbile... ;-) to CoolSoftwareNewbies and wait a reasonable time before construing the absence of answer as an indication that you should send it to CoolSoftwareUsers? It wouldn't even have to be etiquette; a = 3 day old newbie subscription could be required before being allowed to join the user list. But I don't think we need it. * Yes, sort of, for a certain MUA called mutt, with mutt-users and mutt-dev. After all aren't newbies a subset of users? And although I know mutt is for power users, complete newbies by definition don't. They're offered a choice of mailers as they install their UNIX distribution and what do they know? elm and pine are trees and mutt isn't. As somebody who keeps trying to get his wife to give up her wretched Windoze GUI mailer for mutt, I don't want to chase away the people who get here on their own. But I think we need a netiquette lesson on the web page for subscribing to mutt-users, maybe even with a link to procmail so they can handle the mailing list, and the lesson should also be included in the list subscription confirmation message, before the cookie so that they have to read it. The lesson should include: 1. Try pressing ? in mutt. Right now. Wasn't that great? It gets better. Try / and you'll probably answer your question. (I really think it's much better to tell people this than RTFM. Lusers already know they should RTFM (even if they don't know the acronym) but either figure they can get you to do it or honestly tried and failed to find something that seems intuitive to experts. ? is so fast that they might actually try it, and it should be intuitive enough.) 2. The manual location and a suggestion to grep it. 3. The mailing list archive URLs. 4. An invitation to go ahead and post if 1, 2, and 3 didn't satisfy, with the caveat that such shockingly offensive reasons for not being satisfied by 1, 2, and 3 as "I'm at work, so I don't have time to read the manual myself" are plain and simple flamebait. Or maybe there should be a mutt-newbie RPM that detects any attempt to mail to mutt-users and replaces it with ?/ Somewhere near the end of the resulting page there could be a key sequence for really posting to mutt-users. -- Police, n. An armed force for protection and participation. - A. Bierce Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: emacs mail mode?
At 6:54 AM EDT on June 14 Joachim Weiss sent off: Try mail-mode instead of auto-fill-mode. This gives you word wrap and it can handle quotations (if you are using font-lock-mode this gives you colored quotations, in addition emacs is able to rearrange paragraphs (M-q) with quotations in it, keeping the quote character in the first column...) I recommend post mode http://astro.utoronto.ca/~reid/mutt/post.el.gz instead, since it'll give you all that and more without having emacs think that it will be sending out your mail itself. Emacs, at least by default with my version, uses a slightly different message format, but post has other advantages. -- For if there is a sin against life, it consists perhaps not so much in despairing of life as in hoping for another life and in eluding the implacable grandeur of this life. - Albert Camus Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP/GPG Keys: http://astro.utoronto.ca/~reid/pgp.html
Re: emacs mutt
At 11:02 PM EDT on June 20 Charles Curley sent off: On Tue, Jun 20, 2000 at 07:24:18PM -0700, Dale Morris wrote: I've started using emacs as editor for mutt. I'm specified it in the muttrc file and also have setup a script file(e-lisp) in /home/me/.mutt called post.el which is a package for running emacs as an email editor with mutt. It seems to be working fine, but I have a couple of questions. specify emacsclient as your editor, not emacs. When you go to edit an email, emacsclient will feed the file to an already running instance of emacs (it will not launch emacs for you). When you are done, hit C-x #, and move your focus back to mutt. Just a slight correction: for some terribly important reason that I can't recall right now, you should use C-c C-c to exit post, not C-x #, the usual command for quitting emacsclient. 2.) On the tool bar at the top of the emacs window, to the right of Mule, is a category called Post. This is specific to the post.el file you mentioned, assuming that the file follows Emacs custom and is for a major mode called post mode. That's exactly what it is, a menu for post. also one for exiting "Save Message and Return from Post" (C-c C-c), which I assume saves the compose buffer. But it doesn't close emacs, it just leaves me in the *scratch* buffer and I have to go ahead and close it with the C-xC-c command. It operates "correctly" for the client/server scheme I outlined above. You leave Emacs running, ready for the next message. and with all your other buffers intact. And I think it operates correctly for people using straight emacs instead of emacsclient too; it encourages them to use emacsclient, which unfortunately isn't as well known as it should be. I blame the documentation that comes with emacs - the developers have concentrated on making emacs able to do everything, so although they made it possible to easily use it with other programs to them other programs seem irrelevant. Anyway, if you INSIST on starting and killing emacs for each and every email you send, C-xC-c should prompt you to save the message, then exit. So C-cC-c is just a better way provided by post, and the same old way should work. I didn't put C-xC-c in the menu because I figured everyone already knows how to use it. I have a copy of post.el here; I may try it some day. In my copious free time. I know the feeling. -- "Show me a sane man and I will cure him for you." -- C.G. Jung Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: 1.2.2i problems
At 1:34 PM EDT on June 21 Vincent Danen sent off: Ok... I've still got some problems making my RPM for mutt and I'm not sure what's causing it. I built mutt manually and then I built it with the exact same options for the RPM but I get two totally different outputs. The first is from the manual install and the second is from the RPM install -rwxr-xr-x1 root root 1262377 Jun 21 11:29 /usr/bin/mutt* -rwxr-sr-x1 root mail36607 Jun 21 11:29 /usr/bin/mutt_dotlock* -rwxr-xr-x1 root root 6668 Jun 21 11:29 /usr/bin/muttbug* -rwxr-xr-x1 root root 418000 Jun 21 11:24 /usr/bin/mutt* -rwxr-sr-x1 root root 7588 Jun 21 11:24 /usr/bin/mutt_dotlock* -rwxr-xr-x1 root root 6668 Jun 21 11:23 /usr/bin/muttbug* Anyone know why there's such a discrepancy? I think the RPM might be stripping the binaries but I don't know if that would make such a big difference... Doing a mutt -v on both the RPM and manual binaries produces the same output: It must be either not stripped and/or statically linked, and the quickest way to check is file, i.e. file /usr/bin/mutt -- "Men never do evil so completely and cheerfully as when they do it with religious conviction." - Blaise Pascal (1623-1662), Thoughts, #894 Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: alias classes
At 2:11 AM EDT on June 19 Suresh Ramasubramanian sent off: Rob Reid proclaimed on mutt-users that: If mutt can't or won't be reorganized in a better way, maybe somebody (else ;-) could write a script to take alias definitions and produce lists of save-hooks and fcc-save-hooks. A similar neat trick would be to simultaneously define mailing lists in both your .procmailrc and .muttrc, i.e. !listdefine mutt-users [EMAIL PROTECTED] This is what I do, in fact (using the subscribe and save hook flags in muttrc, and setting a procmailrc as well). That sounds great. Would you like to share it with the list? (please) -- Politics, n. A strife of interests masquerading as a contest of principles. The conduct of public affairs for private advantage. - Ambrose Bierce Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: How to discern underlying threads
At 12:12 PM EDT on June 7 Yip Weng sent off: By default, my .muttrc collapses all threads. I find it difficult to distinguish between (i) mail with underlying threads, and (ii) singular mail. Where there is *new* mail underlying a thread, there is a big fat 'N' to indicate this. However, where I am browsing old mail, I find it better if there is some marking to show which threads are single and which are multiple. Any suggestions? I know what you mean, and although it isn't perfect, technically this is a solution: set index_format="%2C%Z|%[%b %e %k:%M] %-14.14F %3l|%s" i.e., the first column is the message number. Collapsed threads with multiple messages show up as jumps in the numbering. The message number is also useful for jumping to messages of course, especially deleted ones. HTH -- "Science is like sex: Sometimes something useful comes out, but that is not the reason we are doing it." - Richard Feynman Hollerith, v: What thou doeth when thy phone is on the fritzeth. - fortune Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Different From addressed depending on ?
At 2:05 PM EDT on June 7 [EMAIL PROTECTED] sent off: Hi, I would like to be able to use a different "from" address for some emails I send. This is because I have 2 email addresses, home and work. I have set up a mailbox just for my work emails, could mutt detect that I'm viewing a work mailbox when I hit the "m" key, and set the from address to my work email address? I don't think so. Or is there a better condition that could be checked, like the presence of my works domain name in the to or cc fields maybe? Yep, use send-hooks. example: send-hook tea@astro "my_hdr From: The Teatotaller [EMAIL PROTECTED]" send-hook !tea@astro "unmy_hdr From" send-hook tea@astro "set signature='.teasig'" send-hook !tea@astro "set signature='/home/reid/bin/randsig1.pl |'" send-hook !(tea|reid)@astro "my_hdr X-URL: http://astro.utoronto.ca/~reid/" send-hook (tea|reid)@astro "unmy_hdr X-URL" -- All realities have bugs, and it's been proven. See _Goedel, Escher, Bach_ by Hofstadter for more details. - Jim Finnis Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: How to discern underlying threads
I just checked the fine manual, and found that you can also color collapsed threads with ~v, i.e. # collapsed threads color index brightmagentadefault ~v I chose brightmagenta because that's my thread color but it looks much uglier in text so I encourage you to experiment. -- What you don't know won't help you much either. -D. Bennett Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: (OT) editor
Hi, At 3:52 PM EDT on June 1 Manuel Arriaga sent off: Unfortunately I tried it without success; I put #!/bin/sh emacs -f server-start into my ~/.profile (I just found out that my shell is called "bash"... :-) and logged in again, but I get an error message saying emacs: standard input is not a tty [3]+ Exit 1 emacs -f server-start But I seem to have understood something: running emacsclient on a virtual console isn't supposed to open a emacs window on the tty, is it? When I give the command emacs -f server-start at a virtual console, and in a different one call "edit" inside of mutt, mutt just says "Waiting for Emacs..."; the first times this happened I just sat there and pressed "Enter", but then I understood that that buffer (the message i wanted to edit, in this case) was opened in the tty I had ran emacs -f server-start! When I saved the buffer on that tty (with C-x-#) and returned to the mutt console, I had the message menu open (I don't know what to call it...:-) it is the screen you get after editing a message) and could send the message! But can't I use emacsclient to open an emacs window on *that* (i.e., where I am running mutt) tty? If this is so, that pretty much explains why I can't run the emacs-server in the backgroud, right? :-) Right. emacsclient just sends the message to the running emacs, whereever it is. I usually run emacsclient, emacs, and mutt in X, where it makes sense to have mutt and emacs running in different windows but on the same screen. It's possible that gnuclient might be able to do exactly what you're requesting, but I've never tried it. I recommend jed as an editor if you're not using X. It emulates emacs very well and has a scripting language, but still starts up and exits very quickly, so no client/server stuff is necessary. It also has color syntax highlighting in a tty, unlike GNU emacs. HTH, Rob -- "Who is General Failure and why is he reading my hard disk?" - Felix von Leitner, [EMAIL PROTECTED] Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Handling of ^-- $
At 2:21 PM EDT on May 30 Alex Lane sent off: I'm only just now noticing that one of the mailing lists I'm subscribed to appears to truncate the "-- " that separates the message from the message signature to simply "--". I've got to believe this is not proper behavior on the part of the mailing list remailer, or am I wrong about that? Procmail can fix it: # Correct wrong sig-dashes :0 fBw * ^--$ | sed -e 's/^--$/-- /' but that brings back the "problem"* that the space was trying to fix in the first place. * Not that I've ever seen that problem. The lack of a space where there should be one is far more common. -- I was recently on a tour of Latin America, and the only regret I have was that I didn't study Latin harder in school so I could converse with those people. - Former U.S. Vice-President Dan Quayle Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html PGP signature
Re: Two ideas
At 10:04 PM EDT on May 26 David Champion generally semanticized from the world of Null-A: On 2000.05.26, in [EMAIL PROTECTED], "Rob Reid" [EMAIL PROTECTED] wrote: You mean strip sigs from quoted text in a reply? That's the editor's job I strenuously disagree. I think we should rather say: it is the established judgement of the Mutt development team that stripping signatures from quoted text should be done by one's editor, not by Mutt. I don't think that signature removal is obviously a designated task of any program, but that it is a matter of personal preference and of contextual convenience. IIRC, that's why sig stripping was removed from mutt. If the sig is passed to the editor, the editor can do all sorts of things with it. On the other hand, putting "set noquote_sig" in /etc/Muttrc made it a more likely default behavior for everyone, but mutt's philosophy is neither to include features better accomplished by other programs nor to enforce netiquette over configurability. If it were, quote_chars would be hardwired to " "* and sig_sep (?) would be hardwired to "-- " with any attempt to change them resulting in mutt forwarding the perpetrator's address to a spammer. * although supercite is a respectable alternative. I hate to be redundant, but the matter-of-factness above was disturbing. I'm sorry you're upset, but the full quoted sig removal debate is in the mailing list archives for anyones who cares enough to look. -- "Smoke me a kipper, skipper; I'll be back for breakfast." - Red Dwarf's Ace Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Two ideas
At 5:00 PM EDT on May 26 Anton Graham sent off:
Both of these revolve around the use of sig-dashes.
First, I have a corrspondent who (despite protests) uses the ``-- ``
sequence to separate his message from the text he is replying to.
Sounds pretty perverse. Maybe you could convince him (and only him) that the
proper "sig separator" sequence is really -*- or something?
Needless to say, this causes mutt to syntax hilight the first
paragraph of his reply as a signature. Perhaps a check to see how
many lines remain after the sig-dashes would be helpful in dealing
with situations like this. Say, for example, 10 lines remain after
sig-dashes means it's not a sig? This would still provide enough
space for the cast majority of over-long sigs.
But then this wouldn't look as good...evil laugh
Second, I'd like to see a ``strip-from-sig-dashes'' type feature (a la
pine). It's merely a convenience.
You mean strip sigs from quoted text in a reply? That's the editor's job
(although mutt used to do it), and can be done, for example, by using
http://astro.utoronto.ca/~reid/mutt/post.el.gz with emacs(client).
--
nnHnn
nHHHn
nHn
nHHHn
nHHHn
HHH
nHn
H
HHH
HHH^^~"~H~~~^^HHH
HH:^^HH
nH: ^Hn
H::^H
HH: . HH
H|: |H
HH:::H|: - - -- |:::HHH
HH~ ~|: ~~ -- --- |~ ~HH
|:\\\. ./// |
HHH|: . / \ . |HHH
/\ HH|:=-- \. : (_@_ . |HH
/%%% %\ H|: ~-- ~~:|. : ~ -' :|H
|% %| ||:.:|. : :'|| "To my good,dear
\=---%%%' %/ \|: .:/. '\: |/ friends..."
/: _)%'.--.|:. :(. .)\ : |
( _-=%%'.%( )|:| /._.-~ \ |
(:)%~/~~|::|/ /.. \: :
:~| :`::: ( \%/\ '
(__:_) |% |( //%%::\ ' \ '=\
( : )|% | //H%%%::\'==%\
`---'|% | -~ ---(%%H%%%:::.\ / .===%==)---
| _.%%.%.| \%%Hn:::.___.::.-===:%==/
| ~~~| \%%DrS~-~=%%==/ DrS
Ambassodor Londo Mollari by "Dr.Stein" [EMAIL PROTECTED]
Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/
PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Two ideas
At 5:02 PM EDT on May 26 Rob Reid sent off: At 5:00 PM EDT on May 26 Anton Graham sent off: Both of these revolve around the use of sig-dashes. First, I have a corrspondent who (despite protests) uses the ``-- `` sequence to separate his message from the text he is replying to. Sounds pretty perverse. Maybe you could convince him (and only him) that the proper "sig separator" sequence is really -*- or something? Needless to say, this causes mutt to syntax hilight the first paragraph of his reply as a signature. Perhaps a check to see how many lines remain after the sig-dashes would be helpful in dealing with situations like this. Say, for example, 10 lines remain after sig-dashes means it's not a sig? This would still provide enough space for the cast majority of over-long sigs. Just had a better idea...I have this # Correct wrong sig-dashes :0 fBw * ^--$ | sed -e 's/^--$/-- /' in my .procmailrc, (probably from http://www.math.fu-berlin.de/~leitner/procmail/procmailrc ) which you could adapt to something like :0 : * From annoying\.correspondent | sed -e 's/^-- $/**Badly quoted text starts now!** /' I haven't tested it and I'm not a procmail guru, but something vaguely resembling that should work. -- Linux: The OS people choose without $200,000,000 of persuasion. -- Mike Coleman Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Idea: saving vs. deleting
At 2:34 PM EDT on May 24 Marius Gedminas sent off: I suggest adding a new status flag: `d' to indicate that the deletion of this message resulted from decode-save, save-message, or decrypt-save. I like the idea, but d is already used to indicate messages with deleted attachments. How about s? -- coude tat: When the person with the spectrometer gets more telescope time. - D. Kaisler Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Idea: saving vs. deleting
At 3:51 PM EDT on May 24 Marius Gedminas sent off: On Wed, May 24, 2000 at 01:50:29PM -0400, Rob Reid wrote: At 2:34 PM EDT on May 24 Marius Gedminas sent off: I suggest adding a new status flag: `d' to indicate that the deletion of this message resulted from decode-save, save-message, or decrypt-save. I like the idea, but d is already used to indicate messages with deleted attachments. How about s? That's too already used for PGP signed but unverified messages. Maybe `w' (written)? I only see (capital) S whether or not I've tried to verify the signature. Is this a new feature ( 0.95)? (I'm giving the alpha/beta version extra testing...otherwise known as waiting for 1.2.1.) -- "It is bad luck to be superstitious." - Andrew Mathis Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: tags and copy help
At 10:54 AM EST on April 4 Serge Rey sent off: I'm trying to copy a bunch of tagged files to a directory, but I'm only able to have the single file under the cursor copied. Can someone point me to the correct way to apply a copy to a set of tagged files (or a save)? To do any operation on all the tagged messages, prefix it with ; i.e. to save the tagged messages, type ";s", assuming standard keybindings. -- A Law of Computer Programming: Make it possible for programmers to write in English and you will find that programmers cannot write in English. - fortune Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: emacsclient (But really status_format!)
At 12:54 AM EST on March 4 Raju K V sent off: hi, In the screenshot of mutt in your page, some details of email are shown in the staus bar. how do you do this? Oddly enough, the format for the status bar is described in the manual under "status _format". RTFM... set status_format="-%r %f [%?M?%M/?%?m?%m msgs, ?%?n?%n new, ?%?d?%d del, ?%?F?% F flag, ?%?t?%t tag, ?%?p?%p postponed, ?%?b?%b box(es), ?%l bytes] --(%s)%|-" On Fri, Mar 03, 2000 at 12:54:48PM -0500, Rob Reid [[EMAIL PROTECTED]] wrote: At 6:08 AM EST on March 3 Ralf Hildebrandt sent off: On Thu, Mar 02, 2000 at 06:36:15PM -0800, ashley wrote: I'm new to mutt. I read in the FAQ about emacsclient as an editor (rest of entire emacsclient thread snipped.) Some things that might not be in the fine manual are: 0. "Don't quote previous messages after your new text; it's ass backwards!" 1. "NEVER quote entire messages, only quote the relevant material, with your additions immediately underneath." 2. "Don't inject a completely different topic into a thread and expect anyone to find it and understand or care what you're talking about. Got a new topic? Start a new thread!"* 3. "Don't combine the above sins!" * Actually I understand the temptation, and will suggest a new feature in a different thread. -- D'oh! English! Who needs that? I'm never going to England. - Homer Simpson, talking Barney into cutting class in The Way We Was Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Merging post.el and mutt-alias.el
Hi,
Stephan Helma sent me the following patch to merge mutt-alias.el into
post.el and provide menus for expaing aliases inside emacs. I prefer to expand
aliases inside mutt instead of my editor, so I have not folded it into post, or
tested it myself. Is editor alias expansion a must have for a lot of you out
there? Please email me off the list if you have an opinion.
Speaking of more appropriate places to discuss post.el, I suppose I
could set up a post mailing list, but I haven't. I have saved correspondance
from everybody who emailed me about it in an mbox file, which probably isn't
everyone who uses post. Can anybody think of a quick way off the top of their
head for extracting all the From addresses in that file and then mailing them
or putting them into an alias? Whoever comes up with the best solution wins a
free copy of post ;-)
- Forwarded message from Stephan Helma [EMAIL PROTECTED] -
*** post.el.origThu Dec 2 10:05:22 1999
***
*** 367,372
comment-region"
:group 'post)
+ ;; START mutt-alias
+ (defgroup mutt-alias nil
+ "Lookup mutt mail aliases."
+ :group 'post
+ :prefix "mutt-alias-")
+
+ (defcustom mutt-alias-file-list '("~/.mutt_mail_aliases")
+ "*List of files that contain your mutt aliases."
+ :type '(repeat (file :must-exist t))
+ :group 'mutt-alias)
+
+ (defcustom mutt-alias-cache t
+ "*Should we cache the aliases?"
+ :type '(choice (const :tag "Always cache the alias list" t)
+ (const :tag "Always re-load the alias list" nil))
+ :group 'mutt-alias)
+ ;; END mutt-alias
+
;;;
;;;
;;; Customizable Faces
***
*** 471,476
(defvar post-has-attachment nil
"Whether the message has an attachment.")
+ ;; START mutt-alias
+ (defvar mutt-alias-aliases nil
+ "\"Cache\" of aliases.")
+ ;; END mutt-alias
+
;;;
;;;
;;; Interactive Commands
***
*** 791,796
(description (string-read "Description: ")))
(header-attach-file file description
+ ;; START mutt-alias
+ (defun mutt-alias-load-aliases ()
+ "Load aliases from files defined in `mutt-alias-file-list'.
+
+ The resulting list is an assoc list where the car is a string representation
+ of the alias and the cdr is the expansion of the alias. Note that no attempt
+ is made to handle aliases-in-expansions or continued lines."
+ (unless (and mutt-alias-aliases mutt-alias-cache)
+ (with-temp-buffer
+ (loop for file in mutt-alias-file-list do (insert-file-contents file))
+ (setf (point) (point-min))
+ (setq mutt-alias-aliases
+ (loop while (search-forward-regexp "^[ \t]*alias +" nil t)
+ collect (mutt-alias-grab-alias)
+ mutt-alias-aliases)
+
+ (defun mutt-alias-grab-alias ()
+ "Convert an alias line into a cons where the car is the alias and the cdr
+ is the expansion. Note that no attempt is made to handle continued lines."
+ (let ((old-point (point))
+ (end-point)
+ (alias)
+ (expansion))
+ (end-of-line)
+ (setq end-point (point))
+ (setf (point) old-point)
+ (search-forward-regexp "[ \t]" nil t)
+ (setq alias (buffer-substring-no-properties old-point (1- (point
+ (search-forward-regexp "[^ \t]" nil t)
+ (setq expansion (buffer-substring-no-properties (1- (point)) end-point))
+ (setf (point) old-point)
+ (cons alias expansion)))
+
+ (defun mutt-alias-expand (alias)
+ "Attempt to expand an alias."
+ (let ((expansion (assoc alias (mutt-alias-load-aliases
+ (when expansion
+ (cdr expansion
+
+ (put 'mutt-alias-interactive 'lisp-indent-function 3)
+
+ (defmacro mutt-alias-interactive (name alias expansion doc rest body)
+ "Generate a function that asks for an alias (placed into variable named by
+ ALIAS) and gets the expansion (placed into variable named by EXPANSION). If
+ there is an expansion BODY will be evaluated otherwise an error is
+ reported. The function will be given a doc string of DOC."
+ `(defun ,name (,alias) ,doc
+ (interactive (list (completing-read "Alias: " (mutt-alias-load-aliases
+ (let ((,expansion (mutt-alias-expand ,alias)))
+(if ,expansion
+(progn
+ ,@body)
+ (error "Unknown alias \"%s\"" ,alias)
+
+ (mutt-alias-interactive mutt-alias-insert alias expansion
+ "Insert the expansion for ALIAS into the current buffer."
+ (insert expansion))
+
+ (mutt-alias-interactive mutt-alias-lookup alias expansion
+ "Lookup and display the expansion for ALIAS."
+ (message "%s: %s" alias expansion))
+ ;; STOP mutt-alias
+
;;;
;;;
;;; Post Header Mode
***
*** 1018,1023
""
["Quote region"
Re: dogs
At 2:15 PM EST on February 17 Erik Jacobsen sent off: Or, from the OpenBSD man pages: The biff command appeared in 4.0BSD. biff was Heidi Stettner's dog. Years ago a friend was forced to use Unix at work and then turned off by biff. She had spent some time looking for an email notifier, and eventually found biff, with the explanation of its name in the man page. She bitterly railed against any OS where the programmers would be allowed to choose command names using such obscure lines of thought as a dog fetching mail or newspapers. But biff doesn't *fetch* mail, it just barks at the mailman. *mutt* fetches mail. And fetchmail catches frisbees in its mouth ;-) Other Canadians on the list have probably read Farley Mowat's book "The Dog Who Wouldn't Be" and know that Mutt, the title character, can do anything. -- Pollutocracy, n.: Rule by stinking rich folk and badly polluting corporations. - Hamish Wilson Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Mutt Mode?
At 4:32 PM EST on February 11 Alisdair McDiarmid sent off: - A reminder to actually attach that file you said you were going to attach. Ooh, how does that work? Well, you get http://astro.utoronto.ca/~reid/mutt/post.el.gz and follow the instructions. Less cryptically, when you finish editing it scans the body for the apperance of "attach". If it appears, and you haven't already attached anything, it asks if you would like to do so. Sorry, no AI involved, but since this is Emacs that wouldn't be that hard... I wrote it because I was always telling people I'd attached something, then forgetting to attach it. Now I can still forget to tell them in the body and not get reminded, but that's not so embarassing because then the recipient doesn't know I forgot anything. -- Because we don't think about future generations, they will never forget us. - Henrik Tikkanen Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: [Q] attaching PDF file
At 12:13 PM EST on January 24 Boryeu Mao sent off: I'm a `old mutt' user (mutt-0.74) and need to attach a PDF file for an out-going mail. Quickly checked doc's for mutt-0.74 and mutt home page, but didn't find immediate info on whether this is possible or how. IIRC, you should be able to attach any file with "a" in the compose menu. As always, if you've changed key bindings, use "?" to set yourself straight. (Unless you remapped "?" ;-) -- Frustra laborant quotquot se calculationibus fatigant pro inventione quadraturae circuli - Michael Stifel (1544) Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Colour loss when running via an exec command...
At 3:11 PM EST on January 20 Jamie Novak sent off: Eterm --trans --shade "40%" -T Mutt -n "Electronic Mail" -e mutt What happens, though, is that mutt defaults to a monochrome colour scheme when I open it via an exec like that. (It does the same for rxvt terms, etc., as well.) I do something similar and never had any problems. Are you running in 8 bit color and running out of colors, perhaps? If not, try compiling with SLang and setting COLORTERM and COLORFGBG. -- Do they sterilize the needles for lethal injections? - Jack Handey Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: mail ( nn) column
At 6:38 PM EST on November 26 Subba Rao sent off: I would like to know the number that appears in the index page in "( xxx)". What is this number and how is it calculated? It's probably the number of lines, but you should RTFM on index_format, and check your value. -- "A cement mixer collided with a prison van on the Kingston Pass. Motorists are asked to be on the lookout for sixteen hardened criminals." - Ronnie Corbett Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: PGP/GPG
At 3:54 PM EST on November 22 Sean Rima sent off: Why not have one gpg (or pgp) config file for all correspondents? The problem is the fact that there are a few people on the Mutt list who use PGP2 and PGP5. These keys are not able to be used in GPG AFAIK. Ah but they are! Look for RSA, etc, plugins in the contrib directory of gnupg's ftp server. It's up to you to decide whether or not to use them. -- Linux: The OS people choose without $200,000,000 of persuasion. -- Mike Coleman Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: [OT] dingus clicking in rxvt?
At 9:10 AM EST on November 17 Timothy Ball sent off: Wow thanks a lot everyone for the dingus info... I'm still having some "issues" w/ compiling it on solaris, but I think they're just shell issues. I hope so, but are you using my Solaris hack? (http://astro.utoronto.ca/~reid/software/) Otherwise you'll have /proc issues up the wazoo. But if you can fix them, that'd be great... -- A little more than kin, and less than kind. - Hamlet Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Abort current operation
At 6:14 AM EDT on October 27 Martin Julian DeMello sent off: It'd be nice if mutt had an 'abort' key, It does. Control g -- loquacity, n. A disorder which renders the sufferer unable to curb his tongue when you wish to talk. - Ambrose Bierce, The Devil's Dictionary. Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Clicking http- and ftp-Links in mutt?
At 12:29 PM EDT on October 27 David DeSimone sent off: Bruno Daniel [EMAIL PROTECTED] wrote: Is it possible to implement opening a Netscape window upon clicks onto http- and ftp-Links in mutt? Sure! On a more serious note: Mouse-clicks already have a defined meaning within xterm: They are used for cutting, selecting, and pasting text! So how is your xterm supposed to know when you're clicking in order to select some text, or clicking in order to pass the information to the inner program, Mutt, so that it can search for a URL to go to? I use rxvt 2.4.5 patched with something called "dingus", or active-rxvt. It recognizes regular expressions (like URLs) and runs a user defined action, like netscape -remote, or gv for .ps files, etc, when the RIGHT mouse button is pressed. It works really well (i.e. much better than urlview), and doesn't interfere with selecting (left button) or pasting (middle button). Am I missing something that the right button would normally do? Cutting is unneccessary in mutt, anyway. Unfortunately applying the patch to newer versions of rxvt was so difficult that I gave up. I think the dingus author is working on a gnome terminal or something like that now, that will include dingus, but I wish rxvt/etc would incorporate it as a default, or at least a default option. I should note that it only fully works in Linux or other OSes using /proc. It's usable in Solaris for URLs, though. -- FUN is never having to say you're SUSHI!! - Yow! Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Abort current operation
At 12:30 PM EDT on October 27 David DeSimone sent off: Rob Reid [EMAIL PROTECTED] wrote: At 6:14 AM EDT on October 27 Martin Julian DeMello sent off: It'd be nice if mutt had an 'abort' key, It does. Control g It doesn't abort reading a huge folder. In fact, nothing does. Once you start to read that 20 MB folder, you're going to be there until it's done. Control c? Control Alt Delete? Pulling out the plug? An axe to the disk? Some of the "solutions" from Windoze also work in UNIX, but I don't have any folders that big to actually test them... -- "They think they can make fuel from horse manure. Now I don't know if your car will be able to get thirty miles to the gallon, but it's sure gonna put a stop to siphoning." - Billie Holliday Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: Clicking http- and ftp-Links in mutt?
At 4:11 PM EDT on October 27 David DeSimone sent off:
Rob Reid [EMAIL PROTECTED] wrote:
I use rxvt 2.4.5 patched with something called "dingus", or
active-rxvt.
Normally the right-button is used to *extend* the current selection.
For instance, you select some text with the left-button, and then if you
decide you really wanted more, instead of reselecting, you can click or
drag with the right-button to modify the selection. I use it on a very
occasional basis.
Thanks David, and Mikko.
I should note that it only fully works in Linux or other OSes using
/proc. It's usable in Solaris for URLs, though.
Hmm, curious, why is /proc required? Oh well..
Upon reviewing, Solaris also uses /proc, but in an incompatible way. /proc is
used to get the current working directory of the process it's watching. That's
needed for tricks like gv on *.ps and other file/directory tricks, but not for
URLs.
Actually, you can; xterm allows the normal cut/paste actions to be
performed, if you hold the SHIFT key. Still annoying to have to
remember that. :)
Yes, and that's a big reason why I'm sticking with my dingified rxvt-2.4.5
instead of "upgrading" to a gnome-terminal, which I think requires shift-right
mouse button to do any magic. Note that it also works with old eterms and gpm,
so it is NOT X specific. (I guess you could run lynx in another virtual
terminal...)
I'm making it available here:
http://astro.utoronto.ca/~reid/software/active-1.0b4-SNAP2.tar.gz
I think it includes the code for rxvt-2.4.5, but no guarantees.
And here's my lobotomized-but-still-has-the-most-important-feature version for
everything but Linux:
http://astro.utoronto.ca/~reid/software/unlinux-active-1.0b4.tar.gz
Here's some more on dingus outside Linux:
At 3:19 PM EST on March 13 Robert Reid sent off:
At 6:45 AM EST on March 12 David Thorburn-Gundlach sent off:
% At 11:22 PM EST on March 4 Matt Hawkins sent off:
%
% You need to play with dingus then.
%
%http://dingus.mit.edu/
%
% Hey, that's great! Thanks. It needed a bit of a lobotomy to work
% with Solaris though..
Well, I'm a Solaris user, too.. Could you make available the
modified code, or patches, so that I can try it out?
Sure, here it is. First get and untar the dingus source. Replace the
files in dingus/src with the ones in the tar file I'm giving you. The
problems were:
1. termios.h doesn't seem to work on Solaris when sys/ioctl.h is
also included.
2. Solaris's /proc isn't laid out the same way as Linux's, and I
couldn't find a easy way of getting the current working directory
of the current process. There's a command called pwdx which might
work, but I couldn't think of a better way of using that than
system("pwdx | ~/.dinguspipeX");
fscanf(pointer_to_dinguspipeX, "%s", cwd);
where ~/.dinguspipeX would be a named pipe with X = number of this
particular rxvt instance. Let me know if you have a better idea.
3. (Cosmetic) Dingus was printing diagnostics about everything it did,
and I changed it to it now prints to stderr (which I reroute to
/dev/null when I start rxvt.)
The main effect of the lobotomy is that dingus no longer tries to get
the cwd of the active process, which means that suggested tags like cd
or lpr won't work unless you're in the same directory as whre you
started. Fortunately URLs use absolute path names, and it works great
with mutt.
Oh yeah...I think I had to change the quoting in the openURL(${})
parts of .active.tags.X.
Enjoy!
--
dilemma: An unproved lemma.
Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/
PGP Key: http://astro.utoronto.ca/~reid/pgp.html
Re: 1.0/slang problem?
At 11:22 AM EDT on October 22 Vincent Lefevre sent off: On Fri, Oct 22, 1999 at 09:51:42 -0400, Jim Simmons wrote: I'm getting a scroll bar from rxvt, and after doing a few things there are actually lines in the scroll back buffer. I didn't have this problem with 1.0pre4i. I've just noticed exactly the same thing (with Mutt 1.0 and Slang 1.2.2). Here's what I've done: 1) Start Mutt in rxvt from fvwm: Exec "Mutt" rxvt -n Mutt -T Mutt -e env LANG= xmutt What if you start rxvt like this? rxvt +sb -sl 0 -n Mutt -T Mutt -e env LANG= xmutt The +sb prevents drawing the scrollbar, and the -sl allocates 0 lines for the scrollback buffer. -- "Some mornings, it's just not worth chewing through the leather straps." -- Emo Phillips Robert I. Reid [EMAIL PROTECTED] http://astro.utoronto.ca/~reid/ PGP Key: http://astro.utoronto.ca/~reid/pgp.html
