[imms-users] Problem with rc8 marking songs as skipped that were not skipped
Wayne
installation went fairly smoothly (except for having to specify -
laudcore or something like that during the build) and imms plus
audacious appear to function together, I am noticing that imms is
registering songs as skipped that I did not skip. I've been using
imms for quite some time, and this fresh install on this new system is
the first time I've noticed this problem (and I tend to obsessively
watch what imms is doing. . . I find it entertaining).
I didn't see this issue mentioned in the archive, so I thought I'd
mention it in the hopes that it would get fixed in the next release
candidate, or in the final 3.1 release.
Thanks!
Wayne
--
You received this message because you are subscribed to the Google Groups "imms-users" group.
To post to this group, send email to imms-...@googlegroups.com.
To unsubscribe from this group, send email to imms-users+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/imms-users?hl=en.
Dave Schaefer
> I just downloaded and installed both audacious and imms,
> I am noticing that imms is registering songs as skipped that I did not skip.
Do you mean that imms is skipping some songs as they come up for play?
Or that you noticed a bunch of skips or rating changes in the database
all at once? Do you have a reproducible case where this happens for
certain songs or directories or anything?
> I've been using imms for quite some time, and this fresh install on this new system is
> the first time I've noticed this problem
I'm guessing you're using imms 3.1.0-rc8, is that right? Do you
remember what version you were using before? Was it on the same OS?
Was it also with Audacious, or with a different player?
Just seeing if we can narrow down the problem :)
Dave
Michael Grigoriev
Hey, here is Wayne's reply. Do you prefer to keep discussions like
this on-list, so everyone can see what's going on?
How would you feel if I wrote some unit tests for IMMS to help us
troubleshoot things like this? That could help us tell if this was a
problem only for Audacious, only for certain lengths of songs, etc.
What do you think?
Also, do you want to start tracking bugs and potential bugs in IMMS?
Do you want to use the google code page at
http://code.google.com/p/imms/issues/list ? Something else?
I'm going to get my linux machine back up and running in the next
week. I'll see if I can repro Wayne's problem and file a bug with
either IMMS or Audacious.
Dave
---------- Forwarded message ----------
From: Wayne Witzke <wayne....@gmail.com>
Date: Thu, May 27, 2010 at 1:27 PM
Subject: Re: [imms-users] Problem with rc8 marking songs as skipped
that were not skipped
To: Dave Schaefer <dave.s...@gmail.com>
Hello Dave,
What I mean exactly is that Audacious will play the song, and I will
see that imms will register the song in its log. The song will then
finish and then Audacious will move on to the next song, and even
though I did not tell Audacious "next" or jump to another song, imms
will tell me in the log that the song was skipped. So, the song plays
all the way through, and imms is registering the song as skipped.
I've actually done some digging since I posted this problem. This
issue is reproducible and it always seems to occur with the same
songs. The mechanism that Audacious was using to determine song
length, the "fast play-length calculation" is actually miscalculating
for a number of the songs in my collection. When the song plays
through, it finishes long before the calculated song length duration
has elapsed. This appears to be the case for every song that imms
incorrectly registers as skipped. These songs are all VBR songs, and
this incorrect song length calculation might even be a known problem
with the fast play-length calculation algorithm (though from the posts
I've read I'm not 100% certain about this). Unfortunately, when I
turn off fast play-length calculation, play-lengths are not calculated
at all and imms registers nothing as skipped, ever, and doesn't ever
seem to change the "after" rating.
I do not actually know how imms determines when a song is skipped. I
had originally thought that imms just captures "next", "previous" and
"jump" events. However, if this is not true, and imms is determining
song skippage by whether or not the song was able to play all the way
through, then this might explain entirely what is going on.
So, this may not be a problem with imms. I might *wish* that imms
worked independently of the song length, the fact that Audacious is
failing to calculate song lengths correctly, or at all when "fast
song-length calculation" is turned off, is actually a pretty big
problem in and of itself. When the play-length calculation is turned
on, I have now noticed that Audacious will cut off songs prematurely
occasionally, and with it turned off the entire mechanism for playing
arbitrary parts of songs is missing.
Unfortunately, I have no way to know whether the problem is with
Audacious, or with the MAD plug-in. I am tempted to say that it's
Audacious since I've used the MAD plug-in before without experiencing
this problem and the MAD project seems to be somewhat. . . dead.
However, I also don't know if the MAD plug-in that I'm using with
Audacious has undergone modifications, and if it has I don't see a way
to actually register a problem with that plug-in on the main Audacious
site. Of course, I only have limited time and patience available to
me to work on this problem, so it could be that I'm just missing the
form or bug reporting mechanism. Even the bug reporting mechanism for
Audacious didn't appear to be entirely function the last time I
looked. That might have been temporary, though.
Anyway, at this point I'm probably going to say that this really isn't
something that is necessarily a bug with imms. If I were developing a
plug-in in for an application, I would make my list of required
function elements, design my plug-in with those functioning elements
in mind, and then if those elements broke it would be a problem with
the software package and not my plug-in. It might be nice if this
wasn't a requirement for imms to work, but of course I have no idea
what the technical requirements are for making various aspects of imms
function properly, and it may just not be possible to get imms to work
without a properly calculated play-length. Since, in the past,
play-length calculation has worked even if the "fast play-length
calculation" has not, I think this may just be a problem with
Audacious.
In case it's not, the answers to the rest of your questions are: Yes,
this is imms 3.1.0-rc8. I do not remember the version that I was
using before, but it was about 9 months ago that I found and installed
imms previously, and I know that I used the latest stable release, not
a release candidate. I was using the same operating system but it was
a different release (Ubuntu 9.04 at the time, Ubuntu 9.10 now). I was
using it with Audacious, but it was a different release of Audacious
(I do not remember the version I was using at the time, but it was
also installed around 9 months ago, and it was the latest stable
version, not installed via the Ubuntu repository). The version of
Audacious that I am using now is 2.3.
Thanks for getting back to me Dave.
Wayne
--Wayne Witzke
Aspiring Physicist
wayne....@gmail.com
WARNING: This e-mail message may have been typed on a tiny keyboard.
--
This is not an automated signature. I type this at the bottom of every message.
Dave Schaefer
http://code.google.com/p/imms/issues/detail?id=1 . I can move it
elsewhere if you guys want to use a different tracking system.
Dave Schaefer
I created a 5 second blank sound file in Audacity to test short song
times. Listening to it in IMMS and Audacious gives me this.
First listen:
[/media/CHARM/code/imms/test sound files/5 second blank.mp3]
[Rating: 71] [Last: 0h] [Unknown] [New] [Jumped] [After: 92]
Second listen:
[/media/CHARM/code/imms/test sound files/5 second blank.mp3]
[Rating: 92] [Last: 0h] [Unknown] [Jumped] [After: 100]
Third listen:
[/media/CHARM/code/imms/test sound files/5 second blank.mp3]
[Rating: 100] [Last: 0h] [Unknown] [After: 100]
So this looks like IMMS is not skipping the file but is treating it as
though I jumped away instead of letting it finish. I'm not sure why
it's not doing it once the rating is 100.
I grabbed a VBR music file from http://togeomusic.com/ (apparently
they have lots of stuff legally available under Creative Commons
licenses. Not a plug - I just searched and found them). Listening to
this file (duration ~8 minutes) I get
[/media/CHARM/music/01 Carassul-VBRlow.mp3]
[Rating: 71] [Last: 0h] [Unknown] [New] [Jumped] [After: 92]
--
[/media/CHARM/music/01 Carassul-VBRlow.mp3]
[Rating: 92] [Last: 0h] [Unknown] [After: 100]
This also looks like IMMS might be counting me as jumping away from
the song. Maybe this is two issues, one for short songs and another
for VBR files?
Dave
Artur R. Czechowski
> > IMMS marks some songs as skipped
> I created a 5 second blank sound file in Audacity to test short song
> times. Listening to it in IMMS and Audacious gives me this.
>
> First listen:
> [/media/CHARM/code/imms/test sound files/5 second blank.mp3]
> [Rating: 71] [Last: 0h] [Unknown] [New] [Jumped] [After: 92]
> So this looks like IMMS is not skipping the file but is treating it as
> though I jumped away instead of letting it finish. I'm not sure why
> it's not doing it once the rating is 100.
http://bugs.debian.org/427833
Regards
Artur
--
ciekawe zderzenie kulturowe: w pi�dziawicy in the middle of nowhere wy�ania si�
z bieli pani d�okejka na koniu, byrd�oczerzy i eskadra treser�w ps�w. Patrz�
na siebie wszyscy w milczeniu i ze zdziwnieniem, jak mo�na tak si� m�czy�
dla tak g�upiego hobby i odchodz� bez zrozumienia /emkaj/