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 335515 - [id3tag] xingmux and id3mux don't work together
[id3tag] xingmux and id3mux don't work together
Status: RESOLVED WONTFIX
Product: GStreamer
Classification: Platform
Component: gst-plugins-ugly
git master
Other All
: Normal normal
: NONE
Assigned To: GStreamer Maintainers
GStreamer Maintainers
: 368420 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2006-03-22 13:35 UTC by Alex Lancaster
Modified: 2006-10-31 19:22 UTC
See Also:
GNOME target: ---
GNOME version: 2.9/2.10


Attachments
hexdump of in-progress rip (before xingmux writes headers) (4.37 KB, text/plain)
2006-03-23 09:29 UTC, Alex Lancaster
Details
hexdump after rip is complete (xing headers written) (4.37 KB, text/plain)
2006-03-23 09:30 UTC, Alex Lancaster
Details

Description Alex Lancaster 2006-03-22 13:35:06 UTC
Please describe the problem:
When ripping a CD, use the following pipeline:

lame <options> ! xingmux ! id3mux

no ID3 tags are written.  However:

lame <options> ! xingmux ! tagid3v2mux

works fine

Steps to reproduce:
1. setup an audio property: "audio/x-raw-int,rate=44100,channels=2 ! lame
name=enc vbr=4 preset=1001 ! xingmux ! id3mux "
2. use rhythmbox with new ripping patch at #322268 (sound-juicer should also
work, but untested
3. rip a CD


Actual results:
tracks have xingheaders but no ID3 information

Expected results:
ID3 info be written

Does this happen every time?
Yes

Other information:
This pipeline works:

"audio/x-raw-int,rate=44100,channels=2 ! lame name=enc vbr=4 preset=1001 !
xingmux ! tagid3v2mux"
Comment 1 Alex Lancaster 2006-03-22 13:36:51 UTC
Using the "xingmux ! id3mux", when the rip is in progress it appears that the tags are being written to the file (checking using hexdump) and "file new.mp3" reports it as having ID3 tags, but after it has been written, the tags are removed (maybe by the xingmux element?)
Comment 2 Christophe Fergeau 2006-03-23 09:20:20 UTC
It might be id3mux putting xingmux at the beginning of the stream (where it just added its ID3v2 tag) rather than at the beginning of the mp3 data (ie right after the ID3v2 tag) when xingmux tries to seek to 0 upon receiving EOS.
Comment 3 Alex Lancaster 2006-03-23 09:29:46 UTC
Created attachment 61831 [details]
hexdump of in-progress rip (before xingmux writes headers)

hexdump -Cv of file before end of rip using "xingmux ! id3mux"
Comment 4 Alex Lancaster 2006-03-23 09:30:58 UTC
Created attachment 61832 [details]
hexdump after rip is complete (xing headers written)
Comment 5 Tim-Philipp Müller 2006-03-24 22:34:05 UTC
If tagid3v2mux works and id3mux doesn't, then this is most likely an id3mux bug. I suspect it probably doesn't adjust newsegment event offsets properly or something similar.

For the time being I prefer getting id3v2tagmux in a state where it can be moved into -good (and then ultimately replace id3mux) rather than trying to fix up the rather messy id3mux codebase.

Comment 6 Christophe Fergeau 2006-03-24 23:26:04 UTC
Oops, sorry for the wrong reassignments from -ugly to -good, I mixed id3mux (from -bad) and id3demux (from -good) ;)
Comment 7 Tim-Philipp Müller 2006-05-04 08:50:51 UTC
What's the status on this?

Has this been resolved by your other fixes to id3mux by any chance, Alex?
Comment 8 Alex Lancaster 2006-09-24 01:15:13 UTC
Now that the id3v2mux is in -good and in the released version and id3mux is deprecated in favour of the taglib-based retagger, I'm closing this as a WONTFIX.

I haven't tested whether my fixes to id3mux fixed it yet, but given that id3mux is broken in so many other ways, seems sensible to be close this bug.
Comment 9 Tim-Philipp Müller 2006-10-31 19:22:05 UTC
*** Bug 368420 has been marked as a duplicate of this bug. ***