<<< Date Index >>>     <<< Thread Index >>>

atime/mtime (Was: mutt: 5 new changesets)



On Sat, Jun 20, 2009 at 00:00:03 -0700, Brendan Cully wrote:
> http://dev.mutt.org/hg/mutt/rev/9ae13dedb5ed
> changeset:   5922:9ae13dedb5ed
> branch:      HEAD
> tag:         tip
> user:        Rocco Rutte <pdmef@xxxxxxx>
> date:        Fri Jun 19 22:49:54 2009 +0200
> summary:     Fixup atime for mbox/mmdf also when mailbox is unchanged but has 
> new mail. See #1362.
> 
> http://dev.mutt.org/hg/mutt/rev/bd59be56c6b0
> changeset:   5921:bd59be56c6b0
> branch:      HEAD
> user:        Rocco Rutte <pdmef@xxxxxxx>
> date:        Fri Jun 19 19:27:37 2009 +0200
> summary:     Don't mangle atime/mtime for mbox folders without new mail upon 
> sync. Closes #1362, #3271.

I applied these two on top of 1.5.20. But I think changeset
5922:9ae13dedb5ed doesn't make sense to me.  If I leave any mail marked
as new (which I do if I intend to yet read it or work on it) in any
folder, mutt marks the whole mbox as new.  That leads to the peculiar
situation that if I try to 'c'hange folders, mutt keeps leading me back
to the previously visited one, as if I had not seen it yet - back and
forth.  If I mark the current folder with '%' (don't write changes), I
can finally get to my other folders.

I think this behavior is not desired.

When I back out said changeset, behavior seems back to "normal", what I
consider normal.

Any thoughts?
Moritz