Re: What does check-new in browser menu actually do?
At the risk of following up to myself (one foot in internet hell!), having done another little investigation, here's what I found. The code responsible for the check-new command is in browser.c, in functions _mutt_select_file (public) and examine_mailboxes (static). examine_mailboxes ultimately passes the job to mutt_buffy_check, but calls the latter with the force argument set to 0, which means nothing will happen unless the mailbox checking timeout has expired. Since this code executes in response to an explicit user request, this seems strange. What is the reason for not passing force = 1 here? -- Please *no* private copies of mailing list or newsgroup messages. Rule 420: All persons more than eight miles high to leave the court.
What does check-new in browser menu actually do?
I often spend lots of time in the browser view, in fact it is my default interface to mutt. When I get bored I hit the check-new key combo. I think I have never once seen new mail detected that way, though. When I get _really_ bored I hit the y key to return to the index of the last mailbox I was in. And lo, many times _immediately_ after doing that I see a buffy notification on the bottom telling me about new mail in other mailboxes. So, I have a sneaking suspicion that check-new doesn't do what it's supposed to, or else I quite misunderstand its purpose. I cannot get any more detail from the documentation; I could of course check the source code but I feel lazy today. Can the experts help? As a bonus question, why can't mutt automatically check for new mail in the background when it is in the browser, the same way as it does in the index? It's yet another irritating inconsistency. -- Please *no* private copies of mailing list or newsgroup messages. Rule 420: All persons more than eight miles high to leave the court.
folder browser, check-new and all that
Hi all In connection to my earlier postings, I did some more testing and I'm pretty sure that I tracked down a kind of pathological test case to reproduce the issue. I also found that the source of the problem I'm seeing is actually connected to my first post about the check-new function. Here is what I did. Set up two mailboxes foo/, bar/. Enter foo/ and leave back to the browser (important). While having mutt open *and* being in the browser, send mail to the two boxes. Just for the record, I did echo lala | mail -s to_foo user@localhost echo lala | mail -s to_bar user@localhost and in my .procmailrc as first entries :0 * ^Subject:.*to_foo.* foo/ :0 * ^Subject:.*to_bar.* bar/ Now, invoke check-new (in my case hit `n': 'bind browser n check-new'). And now the thing: (1) The 'N' flag appears *only* for bar/, not for the last-visited foo/. (2) Choose any mailbox but foo/, hit Return to enter it, and find yourself in foo/. This works vice versa if bar/ was visited last. I also have set timeout=2 set mail_check=20 The issue turns up always, regardless of whether I wait $timeout+$mail_check seconds before invoking check-new or not, but *only* if I do use check-new. So, from what I see, mutt (at least mine) does not flag the last-visited mailbox correctly when I use check-new and does open it, although I selected another one. And now here another, related problem. After reading some more docs I began to imagine that actually most users seem to work like so: have one mailbox open (not the browser) and change directly to other mailboxes, not via the browser. I tried that and I was surprised. That works exactly as I expected. I got informed by a message New mail in ... after $timeout+$mail_check seconds automatically and without pressing a key. This brings me back to my fist posting. Apparently mutt's check-for-new-mail-and-inform-me machinery works like a charm ... but from what I see only if I'm in the message index of an open mailbox, and not in the browser, i.e. the 'N' flag does not appear automatically just as the New mail in ... messages do. Why is this so. Is this intended behavior? Does anybody else have this issue (besides Raffi Khatchadourian) or am I just not using mutt the mutt way? Thanks. s.
Re: $check-new
On Thu, Sep 26, 2002 at 03:40:07AM +0200, Sven Guckes wrote: don't forget about comp.mail.mutt! http://www.google.com/search?q=mutt+%22new+mail%22scoring=d Ah, TSM must be the problem, thanks. Johan
Re: $check-new - TSM problem?
On Thu, Sep 26, 2002 at 02:28:53PM +0200, Sven Guckes wrote: * Johan Svedberg [EMAIL PROTECTED] [2002-09-26 09:58]: Sven Guckes wrote: don't forget about comp.mail.mutt! http://www.google.com/search?q=mutt+%22new+mail%22scoring=d Ah, TSM must be the problem, thanks. TSM? TSM Tivoli Storage Manager The backup system we use. Read in the manual that this could be the problem, but now the admin says that it's not and mutt is compiled the right way on the system. So I'm confused again. I don't have anything like biff touching the files (I think)... Johan
Re: $check-new - TSM problem?
* Johan Svedberg [EMAIL PROTECTED] [2002-09-26 13:41]: Ah, TSM must be the problem, thanks. TSM? TSM Tivoli Storage Manager The backup system we use. Read in the manual that this could be the problem, but now the admin says that it's not and mutt is compiled the right way on the system. So I'm confused again. I don't have anything like biff touching the files (I think)... maybe you should just rely on your procmail logfile... i remember there some nice (perl) script which can visualize the (new) mails within the folders since last time you checked. but this is not in sync with the current state within the folders, of course... Sven
$check-new
Hi, all. I'm using 'mutt -y' alot to monitor my mailboxes, and I have this to be able to see where I have new mail: bind browser $ check-new Although I'm experienceing some buggy behaivor by mutt. Sometimes (I can't see a pattern) it doesn't give the mailbox the N flag marking new mails, although there really are new mails in that box because I can see the filesize of the mailbox change and when I enter it there are new mails in it. Has anyone else seen this? -- Johan Svedberg, [EMAIL PROTECTED], http://www.acc.umu.se/~winkle
Re: $check-new
* Johan Svedberg [EMAIL PROTECTED] [2002-09-25 13:15]: I'm using 'mutt -y' alot to monitor my mailboxes, and I have this to be able to see where I have new mail: bind browser $ check-new Although I'm experienceing some buggy behaivor by mutt. Sometimes (I can't see a pattern) it doesn't give the mailbox the N flag marking new mails, although there really are new mails in that box because I can see the filesize of the mailbox change and when I enter it there are new mails in it. Has anyone else seen this? yes, it has been reported about a dozen times now. - archive Sven
Re: $check-new
On Wed, Sep 25, 2002 at 03:30:31PM +0200, Sven Guckes wrote: * Johan Svedberg [EMAIL PROTECTED] [2002-09-25 13:15]: I'm using 'mutt -y' alot to monitor my mailboxes, and I have this to be able to see where I have new mail: bind browser $ check-new Although I'm experienceing some buggy behaivor by mutt. Sometimes (I can't see a pattern) it doesn't give the mailbox the N flag marking new mails, although there really are new mails in that box because I can see the filesize of the mailbox change and when I enter it there are new mails in it. Has anyone else seen this? yes, it has been reported about a dozen times now. - archive My search skills must suck, because I can't find anything. :-( Do you (or anyone else) have the strength to explain? :) Johan Svedberg
Re: $check-new
* Johan Svedberg [EMAIL PROTECTED] [2002-09-25 19:04]: On Wed, Sep 25, 2002 at 03:30:31PM +0200, Sven Guckes wrote: * Johan Svedberg [EMAIL PROTECTED] [2002-09-25 13:15]: I'm using 'mutt -y' alot to monitor my mailboxes, and I have this to be able to see where I have new mail: bind browser $ check-new Although I'm experienceing some buggy behaivor by mutt. Sometimes (I can't see a pattern) it doesn't give the mailbox the N flag marking new mails, although there really are new mails in that box because I can see the filesize of the mailbox change and when I enter it there are new mails in it. Has anyone else seen this? yes, it has been reported about a dozen times now. - archive My search skills must suck, because I can't find anything. :-( Do you (or anyone else) have the strength to explain? :) don't forget about comp.mail.mutt! http://www.google.com/search?q=mutt+%22new+mail%22scoring=d Sven
