headers
Thomas Roessler | 1 Sep 2002 10:12

[2002-09-01] CVS commit messages

This message was generated and sent automatically.  It contains a
summary of the CVS commits over the last 24 hours.  These changes
should be propagated to the public repository within at most a day
or two.  Most probably, they have already been propagated.

No changes were committed to the repository over the last 24 hours.
Permalink | Reply | 0 comments |
headers
Peter Surda | 2 Sep 2002 22:02
Picon

outbox, postponed, IMAP & bad decisions

Hello!

I have mutt-1.4i-2.rhl7 (I suppose 1.4 with some small RH-specific patches)
and a BIG problem.

I use IMAP for everything, also for outbox. Quite often the connection to IMAP
breaks after no data has been transferred for a while (like when I'm writing a
long email). This isn't a bad thing per se. But, sometimes mutt only realizes
this in a critical moment (when saving a message to postponed or outbox) and
then simply ignores it, effectively not making a copy of the message, but
continuing with the action anyway (e.g. it sending the message). So basically
I lose my message.

I find it very annoying and I think it is pretty clear this should work
differently. I am confident it isn't overly difficult to fix. I am too tired
at the moment to look at the code. I remember vaguely this was already
discussed on the list but don't remember if it was solved.

Please tell me if:
- this was already fixed in cvs
- someone is working on it
- I should try fixing myself
- I should use a different email client (well, don't tell me this, I don't
  know what I'd do without mutt).

Bye,

Peter Surda (Shurdeek) <shurdeek <at> panorama.sth.ac.at>, ICQ 10236103, +436505122023

--
(Continue reading)

Permalink | Reply | 4 comments |
headers
GUUG bug Tracking System | 3 Sep 2002 20:18
Picon

Processed: your mail

Processing commands for control <at> bugs.guug.de:

> merge 1328 1329
bug#1328: mutt-1.4i: Mutt deletes files where it should not.
bug#1329: mutt-1.4i: Mutt deletes files where it should not.
Merged 1328 1329.

>
End of message, stopping processing here.

Please contact me if you need assistance.

Herr der Kaefer
(administrator, GUUG bugs database)
Permalink | Reply | 0 comments |
headers
Michael Elkins | 3 Sep 2002 20:15

Re: outbox, postponed, IMAP & bad decisions

Peter Surda wrote:
> Please tell me if:
> - this was already fixed in cvs
> - someone is working on it
> - I should try fixing myself
> - I should use a different email client (well, don't tell me this, I don't
>   know what I'd do without mutt).

This is a known bug.  If you look at the bug tracking database
(bugs.guug.de) you will see several open reports about this issue.  So
far nobody has stepped forward to work on the numerous IMAP bugs.
Permalink | Reply | 3 comments |
headers
Brendan Cully | 3 Sep 2002 20:41
Gravatar

Re: outbox, postponed, IMAP & bad decisions

On Tuesday, 03 September 2002 at 11:15, Michael Elkins wrote:
> Peter Surda wrote:
> > Please tell me if:
> > - this was already fixed in cvs
> > - someone is working on it
> > - I should try fixing myself
> > - I should use a different email client (well, don't tell me this, I don't
> >   know what I'd do without mutt).
> 
> This is a known bug.  If you look at the bug tracking database
> (bugs.guug.de) you will see several open reports about this issue.  So
> far nobody has stepped forward to work on the numerous IMAP bugs.

I'm sorry about that. I've been AWOL since taking my latest job a year ago.
But I gave notice Friday (woohoo! but also probably stupid). I'm going on a
trip to Europe in three weeks but expect to be able to get a lot done when
I get back, if no one else has done any serious work before then.

-b
Permalink | Reply | 0 comments |
headers
Peter Surda | 4 Sep 2002 04:02
Picon

Re: outbox, postponed, IMAP & bad decisions

On Tue, Sep 03, 2002 at 11:15:14AM -0700, Michael Elkins wrote:
> > Please tell me if:
> This is a known bug.  If you look at the bug tracking database
> (bugs.guug.de) you will see several open reports about this issue.  So
> far nobody has stepped forward to work on the numerous IMAP bugs.
Well this is the only one that bugs me.

The problem is that you may trigger a situation with no-way-out. If you have
everything over imap and edit a long file and between that the imap connection
breaks:
- if you try to postpone it, you lose the msg
- if you try to send it, it sends, but you lose it anyway
- basically you have no other options

This sucks badly. I just tried fixing it for several hours, without success.
Making the current IMAP code reconnect automatically is really complex and I
can't even find a reliable way to tell if the connection is there or not.

The code is too linear, like
- try to do something
- if there is an error, close and quit everything you can
So at the end you lose all unsaved data without recovery options.

And the "stupidest bug" award goes to the guy who put a:
      mutt_write_fcc (fcc, msg, NULL, 0, NULL);
into send.c. Behold, there is absolutely no test for the return value. It even
doesn't have anything to do with IMAP.

So until it's fixed I set mutt to do a bcc instead of fcc and added a procmail
rule to put it into the "outbox". How to do it with postponed so that it works
(Continue reading)

Permalink | Reply | 1 comment |
headers
Thomas Roessler | 4 Sep 2002 10:12

[2002-09-04] CVS commit messages

This message was generated and sent automatically.  It contains a
summary of the CVS commits over the last 24 hours.  These changes
should be propagated to the public repository within at most a day
or two.  Most probably, they have already been propagated.

No changes were committed to the repository over the last 24 hours.
Permalink | Reply | 0 comments |
headers
Emil Sit | 4 Sep 2002 15:27
Picon

Re: outbox, postponed, IMAP & bad decisions

> And the "stupidest bug" award goes to the guy who put a:
>       mutt_write_fcc (fcc, msg, NULL, 0, NULL);
> into send.c. Behold, there is absolutely no test for the return value. It even
> doesn't have anything to do with IMAP.

You can always try the following patch. It is also available
from the BTS under bug #265 but you have to wade through a
bunch of spam to get to it. (Perhaps bugs.guug.de should filter
messages with spamassasin or bogofilter before adding messages
to bug logs.)

Index: send.c
===================================================================
RCS file: /home/roessler/cvs/mutt/send.c,v
retrieving revision 2.77.2.1
diff -u -r2.77.2.1 send.c
--- send.c	2002/01/30 22:50:59	2.77.2.1
+++ send.c	2002/03/20 19:11:36
 <at>  <at>  -1529,7 +1529,21  <at>  <at> 
        * message was first postponed.
        */
       msg->received = time (NULL);
-      mutt_write_fcc (fcc, msg, NULL, 0, NULL);
+      if (mutt_write_fcc (fcc, msg, NULL, 0, NULL) < 0)
+      {
+	/* sleep for error message from things like mx_open_mailbox */
+	mutt_sleep (2);
+	if (! (flags & SENDBATCH))
+	{
+	  mutt_error _("Could not write Fcc!");
(Continue reading)

Permalink | Reply | 0 comments |
headers
Pieter Donche | 4 Sep 2002 15:29
Picon

install problem mutt-1.4

Hi,

I have SUN Blade 2000, Solaris 9, gcc-3.2...
First installed libiconv, latest version 1.8 in /usr/local
Downloaded mutt-1.4i
# ./configure --with-iconv=/usr/local
(added --with-iconv=/usr/local because the Solaris iconv 'was not good
enough' and configure stopped)
Configure runs to the end, but has said:
...
checking for iconv.h... yes
checking whether iconv.h defines iconv_t... yes
configure: warning: Unable to find an iconv function. See INSTALL for help
...

Why is that ????

I do have

# ls -la /usr/local/lib/libiconv*
-rw-r--r--   1 root     other        719 Sep  4 13:23 /usr/local/lib/libiconv.la
lrwxrwxrwx   1 root     other         17 Sep  4 13:23 /usr/local/lib/libiconv.so -> libiconv.so.2.1.0
lrwxrwxrwx   1 root     other         17 Sep  4 13:23 /usr/local/lib/libiconv.so.2 -> libiconv.so.2.1.0
-rwxr-xr-x   1 root     other    1117830 Sep  4 13:23 /usr/local/lib/libiconv.so.2.1.0
-rw-r--r--   1 root     other    1115587 Sep  4 13:23 /usr/local/lib/libiconv_plug.so

# ls -la /usr/local/include/icon*
-rw-r--r--   1 root     other       4294 Sep  4 13:23 /usr/local/include/iconv.h

(all installed today..)
(Continue reading)

Permalink | Reply | 6 comments |
headers
David Champion | 4 Sep 2002 18:53
Favicon

Re: install problem mutt-1.4

* On 2002.09.04, in <Pine.GSO.4.40.0209041528290.16365-100000 <at> hcoss.uia.ac.be>,
*	"Pieter Donche" <Pieter.Donche <at> ua.ac.be> wrote:
> Hi,
> 
> I have SUN Blade 2000, Solaris 9, gcc-3.2...
> First installed libiconv, latest version 1.8 in /usr/local
> Downloaded mutt-1.4i
> # ./configure --with-iconv=/usr/local
> (added --with-iconv=/usr/local because the Solaris iconv 'was not good
> enough' and configure stopped)

Have you tried without libiconv? I use Solaris's native iconv (with no
--with-iconv); it works quite fine.

I'm using 9 beta still, but it should be basically the same.

--

-- 
 -D.			We establised a fine coffee. What everybody can say
 Sun Project, APC/UCCO	TASTY! It's fresh, so-mild, with some special coffee's
 University of Chicago	bitter and sourtaste. "LET'S HAVE SUCH A COFFEE! NOW!"
 dgc <at> uchicago.edu	Please love CAFE MIAMI. Many thanks.
Permalink | Reply | 1 comment |

Gmane