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 132746 - mp3parse/mpegaudio plugin broken (skips way too much)
mp3parse/mpegaudio plugin broken (skips way too much)
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: gst-plugins
git master
Other Linux
: Normal normal
: 0.7.4
Assigned To: Ronald Bultje
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2004-01-28 10:25 UTC by Tim-Philipp Müller
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Tim-Philipp Müller 2004-01-28 10:25:54 UTC
I've tried running an mp3 file through mp3parse like this in both 0.6 and 
0.7.x (cvs-head): 
 
gst-launch filesrc location=in.mp3 ! mp3parse ! filesink location=out.mp3 
 
(purpose was to clean mp3 from random rubbish and tags before processing 
it further). 
 
However, the output mp3parse produces contains noticable skips when 
played, so something is not working right. 
 
 
Source 1: 
(has clean mpeg audio frame sync at beginning): 
 
filesize   
4303306   mp3.in      (source, contains ID3v1 tag at the end) 
4302889   mp3.out-0.6 (sounds like source) 
4214403   mp3.out-0.7 (has at least 1 audible skip or error/second) 
 
 
Source 2: 
(first clean mpeg audio frame sync at byte 252): 
 
filesize 
3393919   mpeg.in        (source, has no ID3 tags) 
3393411   mpeg.out-0.6   (sounds like source) 
3320635   mpeg.out-0.7   (constant audible skips) 
 
 
So it looks like there has been a regression since 0.6.4 and 
0.7.x-cvs-HEAD. 
 
Cheers 
-Tim
Comment 1 Ronald Bultje 2004-01-28 16:14:06 UTC
I worked on that, I probably broke something.
Comment 2 Ronald Bultje 2004-01-28 16:22:41 UTC
We're sometimes parsing one byte too much, the header indicates a
certain length but the frame is really one byte shorter...

We probably committed a fix for this in 0.6.x but forgot to apply it
to HEAD, too.

Or I accidently removed it, dunno...

Working on it.
Comment 3 Ronald Bultje 2004-01-30 14:23:07 UTC
Fixed