After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 140020 - Song skips when position is moved maximum right
Song skips when position is moved maximum right
Status: RESOLVED FIXED
Product: rhythmbox
Classification: Other
Component: User Interface
0.8.8
Other Linux
: Normal normal
: ---
Assigned To: RhythmBox Maintainers
RhythmBox Maintainers
Depends on:
Blocks:
 
 
Reported: 2004-04-14 11:11 UTC by Tom kalmar
Modified: 2011-07-26 19:46 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
fix various bits of slider behaviour (9.93 KB, patch)
2008-05-20 07:46 UTC, Jonathan Matthew
committed Details | Review

Description Tom kalmar 2004-04-14 11:11:25 UTC
1. Play a song
2. move the position slider maximum left (with Mouse)
3. the Song skips
4. moving (the mouse) further left skips the next song
I expected the Song not to skip at least not the next Song too. The position
slider should be set to 0 after Skipping.
Comment 1 Colin Walters 2004-04-15 19:14:03 UTC
I can reproduce occasional small skips in the sound.

However I can't reproduce the switch to a different song.  What version of
Rhythmbox and GStreamer are you using?

Comment 2 Tom kalmar 2004-04-15 23:56:18 UTC
I use Rhythmbox 0.6.10 and gstreamer 0.6.4 from debian unstable.
Comment 3 Tom kalmar 2004-05-22 21:33:44 UTC
Same behaviour with Debain Version 0.8.4...
to make myself clear with skip i dpn't mean corruption in output. I mean the
next Song starts playing. When moving the position slider left the song starts
repeating as expected.
moved right the song gets changed ...
thats not directly wrong but when i move the mouse even further right the next
song starts playing. Thats totaly unexpected... the Slider position should be
set to 0:00 or only one song should be skipped. 
Comment 4 Sebastien Bacher 2005-01-14 11:51:21 UTC
do you still get this issue with 0.8.8 ? it doesn't change the song here
Comment 5 Tom kalmar 2005-01-15 09:58:16 UTC
still starts the next song ... but at least only one Song not unlimited as before 

Comment 6 Sebastien Bacher 2005-06-19 20:48:55 UTC
do you consider that a bug? Seems to be the right behaviour to me
Comment 7 Tom kalmar 2005-06-19 22:18:30 UTC
the Problem i have with this behavior is that i expect to navigate within the
current song not to skip the current song.
It's sometimes hard do navigate to the last seconds of a song without skipping
to the next. so in my eys this is a bug  at least unexpected behaviour
Comment 8 John Daiker 2006-07-26 16:44:04 UTC
I would argue that this behaviour is expected.  If the position slider is moved 100% to the right, then the current track should stop playing, and the next track should begin.  The slider should also reset itself (since a new song is playing).

If the slider is not moved 100% to the right (IE only 80% or something), then the same song should continue playing from that point forward.  If this is the  current behaviour, then I have no problem with it.

What behaviour exists in the latest CVS? (I'm at work, so I can't test, sorry)
Comment 9 Tom kalmar 2006-08-14 20:38:39 UTC
In 0.9.5 the behaviour is even stranger.

Song A duration 1:00
Song B duration 3:00

skipping Song A with the positionslider results in only beeing able to move the slider to position 1:00 wich is at 100% 

I also think that the behaviour of the slider should be consistent on booth ends. Navigating for AND back or dont navigate.

I realized at this moment that i actualy meant right not left ... OMG ...
 
Comment 10 James "Doc" Livingston 2006-08-15 06:16:51 UTC
(In reply to comment #9)
> In 0.9.5 the behaviour is even stranger.
> 
> Song A duration 1:00
> Song B duration 3:00
> 
> skipping Song A with the positionslider results in only beeing able to move the
> slider to position 1:00 wich is at 100% 

This sounds like a bug, not updating the slider correctly if the song changed when it's being dragged. I'll take a look.


> I also think that the behaviour of the slider should be consistent on booth
> ends. Navigating for AND back or dont navigate.

When you drag the slider it seeks - so you drag it all the way to the right it seeks to the end of the song, and because the song ends shortly after it changes to the next one.

What do you suggest it does instead, if you seek towards the end of the song?
Comment 11 Tom kalmar 2006-08-15 11:18:16 UTC
(In reply to comment #10)
> 
> When you drag the slider it seeks - so you drag it all the way to the right it
> seeks to the end of the song, and because the song ends shortly after it
> changes to the next one.
> 
> What do you suggest it does instead, if you seek towards the end of the song?
> 

It shouldn't skip unless i release the mouse i think. Stoping/pausing playback if the slider is dragged to 100% is an option. On the left side the playbackposition is permanently resetet to 0 so you hear the first few millis of the song over and over again until you release the mouse. Could  similar behavoiur be implemented for the right side?
Playing the last frame/second/milliseconds and not skipping to next song? The positionslider should IMHO navigate the current song only.

The simplest solution would be to update the position after releasing the mouse and not while moving the slider.
Comment 12 Edgar Luna 2006-08-28 20:00:48 UTC
Maybe a plugin that plays the first N time or the last N time of the actual song, could that solve what you want?

I really think that the actual is the expected behaviour.
Comment 13 Laszlo 2006-11-25 13:35:15 UTC
I had a similar problem to these, but not quite the same from what I can tell by the previous comments. No problems with moving slider 100% left. Goes to beginning of song but doesn't skip back. 

But when I move the slider 100% right, it will sometimes skip a second or even third song as played if I don't let go of my click very quickly. I would understand it skipping to the song immediately after (b/c it's still playing, sliding to the end will make it play through the end of the track), but the slider should be reset at 0% for each track to prevent multiple skip-throughs.

Running version 0.9.6.
Comment 14 John Bryant 2007-01-06 07:40:03 UTC
I think Rhythmbox shouldn't change which song it's playing as long as the user is holding the slider.  It's a valid assumption that they're looking for something in that particular song.  Maybe it's odd, but I feel like holding the slider is like holding on to the current song.  If that song changes, I feel like it's yanked away.

Still, the slider must be held to the right for about 4 seconds before the next song appears.  Maybe this is a track-skipping feature.  Even so, when this occurs the "length" of the slider no longer matches that of the song.
Comment 15 Jonathan Matthew 2008-05-20 07:45:03 UTC
(In reply to comment #14)
> I think Rhythmbox shouldn't change which song it's playing as long as the user
> is holding the slider.  It's a valid assumption that they're looking for
> something in that particular song.  Maybe it's odd, but I feel like holding the
> slider is like holding on to the current song.  If that song changes, I feel
> like it's yanked away.

I agree.  I think it would be sensible to defer handling of end-of-stream events while the slider is being dragged.  If the stream is still at its end when the user releases the slider, then we should go to the next track as normal.  I don't think the slider is a good interface for skipping tracks, and we already have something else that does that pretty well.
Comment 16 Jonathan Matthew 2008-05-20 07:46:55 UTC
Created attachment 111204 [details] [review]
fix various bits of slider behaviour

This also cleans up the code a bit and fixes a few other problems (bug 496816, for instance).
Comment 17 Paul Drain 2008-05-24 14:46:17 UTC
This patch seems to also fix the slider going nuts if you scroll toward the end of the song playing in your library and switch to your iPod (or click any of the other areas, like podcasting).

Nice, fixes a bugbear i've had with RB for a fair while now.
Comment 18 Paul Drain 2008-05-26 03:29:04 UTC
Was part of this committed as part of 5706?

When trying to patch against today's SVN (5708) I get:

patching file shell/rb-shell-player.c
Reversed (or previously applied) patch detected!  Assume -R? [n]
Apply anyway? [n]
Skipping patch.
6 out of 6 hunks ignored -- saving rejects to file shell/rb-shell-player.c.rej
patching file widgets/rb-header.c

Should I nuke it from my patch set, or should the parts of rb-header.c be committed?
Comment 19 Jonathan Matthew 2008-05-26 03:44:07 UTC
Looks like I did accidentally commit the changes from shell/rb-shell-player.c.  Oops.  I'll back out the bits I didn't mean to commit later.
Comment 20 Jonathan Matthew 2008-05-26 10:52:34 UTC
OK, I've reverted the extra changes and this patch appears to apply correctly again.  Thanks for pointing that out.
Comment 21 Jonathan Matthew 2008-06-28 04:31:17 UTC
committed.