Summary: | audiocd kio does not show correct number of songs | ||
---|---|---|---|
Product: | [Unmaintained] kio | Reporter: | Beat Wolf <asraniel> |
Component: | audiocd | Assignee: | icefox |
Status: | RESOLVED WORKSFORME | ||
Severity: | normal | CC: | aacid, arthur.breitman_kde |
Priority: | NOR | ||
Version: | unspecified | ||
Target Milestone: | --- | ||
Platform: | Ubuntu | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | ||
Sentry Crash Report: | |||
Attachments: | Fixes problems of kio_audiocd hanging in endless loop due to failure of reading number of tracks. |
Description
Beat Wolf
2006-04-24 15:21:11 UTC
This sounds like the problem we were having with audiocd on FreeBSD where the CD wasn't getting opened (because of a problem with libcdparanoia IIRC), and then the "number of tracks" variable was used uninitialized. You can see if you're having a similar problem by just reloading the same CD with audiocd:/ - does the same (incorrect) number of tracks appear each time? If not, can you see any relationship between the real number of tracks and the number shown? Sorry: "if not" in the previous comment should have been "if so" Any luck trying the suggestion in comment #1? Feedback timeout. Please reopen if you have the answers to questions in the comments Created attachment 17799 [details]
Fixes problems of kio_audiocd hanging in endless loop due to failure of reading number of tracks.
Hi! This looks very similar to the problem, I have just debugged.
In fact, cdparanoia-IIIalpha9.8-550 (from SuSE 10.0) does not return any device
name in drive->ioctl_device_name (see
kdemultimedia/kioslave/audiocd/audiocd.cpp). Therefore, the device cannot be
opened and the number of tracks is not initialised. Then kio_audiocd hangs in a
(virtually) endless loop, eating up lots of memory while filling a very long
list of track names...
I have to admit, that I am very surprised by the for me incomprehensible mix of
using libwm (inside kscd) and libcdda_paranoia... paranoia already gives the
right count in the drive structure referred above... so?
Anyway, I could fix the problem by the attached patch which falls back to
drive->cdda_device_name in case drive->ioctl_device_name is a null pointer or
zero length.
I don't know how cdparanoia behaves on other drives, but with my drive I
definitely need this. cdparanoia identifies my drive as follows:
Found an accessible SCSI CDROM drive.
Looking at revision of the SG interface in use...
SG interface version 3.5.33; OK.
CDROM model sensed sensed: MATSHITA DVD-RAM UJ-822S 1.61
Other SG drives with that version of cdparanoia most likely all have this
problem.
Cheers,
Tilman
I didn't receive any feedback on the patch yet. Is anybody reading this? I'd very much appreciate my fix to make it into the code. Shall I submit it somewhere else? You might want to send the patch to kde-core-devel@kde.org if it gets no attention here. *** Bug 119819 has been marked as a duplicate of this bug. *** Do you still have this problem with KDE >= 4.4.0? Closing because of lack of answer to my question |