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

Re: [Mutt] #2807: pager.c remembers TopLine when it shouldn't



#2807: pager.c remembers TopLine when it shouldn't

Changes (by myon):

  * priority:  trivial => minor
  * milestone:  => 2.0

Old description:

> {{{
> Since this change (http://mutt.kublai.com/hg/cvs/rev/b894fd4c9032)
> on Sep 08 2005, the builtin pager keeps its position
> even on OP_MAIN_NEXT_UNDELETED. So you cannot re-read
> the last message from the top of it with <space>.
> (See How-To-Repeat)
>
> This must be a side-effect of http://bugs.mutt.org/1210
>
> >How-To-Repeat:
> 1. :unset pager_stop
> 2. <change-folder> to a folder which has a long (more
>  than a screen-length) message as the last entry.
> 3. Hit <space> on the last message.
> 4. Repeat <space> until the message finishes.
> 5. You will be in index mode (because the message
>  is the last entry in the folder).
> 6. Hit <space> again. You will see the _bottom_ of
>  the message, not the top.
>
> >Fix:
> Forget TopLine when <next-page> (with $pager_stop)
> causes OP_MAIN_NEXT_UNDELETED.
> (though Brendan called it "a bogus detection")
> }}}

New description:

 {{{
 Since this change (http://mutt.kublai.com/hg/cvs/rev/b894fd4c9032)
 on Sep 08 2005, the builtin pager keeps its position
 even on OP_MAIN_NEXT_UNDELETED. So you cannot re-read
 the last message from the top of it with <space>.
 (See How-To-Repeat)

 This must be a side-effect of http://bugs.mutt.org/1210

 >How-To-Repeat:
 1. :unset pager_stop
 2. <change-folder> to a folder which has a long (more
  than a screen-length) message as the last entry.
 3. Hit <space> on the last message.
 4. Repeat <space> until the message finishes.
 5. You will be in index mode (because the message
  is the last entry in the folder).
 6. Hit <space> again. You will see the _bottom_ of
  the message, not the top.

 >Fix:
 Forget TopLine when <next-page> (with $pager_stop)
 causes OP_MAIN_NEXT_UNDELETED.
 (though Brendan called it "a bogus detection")
 }}}

Comment:

 There's more instances of this problem:

 From: Michael Gurski <debianbugs@xxxxxxxxxx>
 To: Debian Bug Tracking System <submit@xxxxxxxxxxxxxxx>
 Subject: mutt: Weird scroll on next message to current message's scroll
 amount
  behavior
 Date: Fri, 28 Jul 2006 12:24:03 -0400

 Package: mutt
 Version: 1.5.12-1
 Severity: minor

 I've noticed a really weird scrolling behavior with mutt periodically.
 What happens is I'll be reading a long message, such as a logcheck
 report, tab to the next message while partway scrolled through the
 current one, and mutt will read the next message and immediately
 scroll the same amount down in the new message.

 The even weirder part is that this will also happens if the initial
 message that I've scrolled down on is in one mail folder as the last
 message, and I go to a new folder to read a new message there.

 Not having looked in the sources, I can only guess how mutt's pager
 for messages is implemented, but this almost sounds like a variable
 representing position in the current message isn't being reset before
 opening the new message.

 All in all, a fairly minor annoyance, but since I've hit it a few
 times today, I figured I should at least report it.


 From: Dan Jacobson <jidanni@xxxxxxxxxxx>
 Subject: mutt: wasteful resending of whole last message to the screen
 Date: Sat, 02 Jun 2007 05:19:19 +0800

 Package: mutt
 Version: 1.5.13-3
 Severity: wishlist

 On the last message in ones mailbox, page thru it with spacebar, and
 just keep hitting spacebar even after you reach the end.

 Notice how the message is replayed for you leaving your on the last
 page, but with the whole message first uselessly sent to your terminal
 in the process.

 keep hitting spacebar and it happens again...

-- 
Ticket URL: <http://dev.mutt.org/trac/ticket/2807#comment:3>