GNOME Bugzilla – Bug 783100
asfdemux: Seek to end of the file is failing with error msg
Last modified: 2017-06-07 04:51:34 UTC
I am using gst-plugins-ugly 1.12 and while trying to seek to the end of an asf file , the operation is getting failed with some error messages thrown by asfdemux:- GST_EVENT_EOS: (_("This stream contains no data.")), ("got eos and didn't receive a complete header object")); wrong_type: "This doesn't seem to be an ASF file")) sorry for not posting proper logs however if i revert the changes in the defination of the function "gst_asf_demux_check_header" of earlier version than commit 97294eb8bbed1b9dad7d3f2c52dd69eb1812cc06 , it is working fine.
Thanks for the bug report. Could you make such a file available somewhere by any chance?
(In reply to Tim-Philipp Müller from comment #1) > Thanks for the bug report. > > Could you make such a file available somewhere by any chance? Hi Tim , Is there any server where can i upload my file? Any specific which is used for file sharing on this community??Thanks
Created attachment 352627 [details] test content added test content to verify this bug.
https://www.dropbox.com/s/859atxzzeki19ni/ASF_VC-1%28MP%40LL%29_WMA%28Version%202%29_308k_384_1h53mn_176X114_23.976fps_CZ12.2012.asf?dl=0 added another test content to verify this bug. For the first content the issue can only be seen when playing from filesrc , and not from httpsrc But for the second content the issue can be seen while playing from both of the sources.Thanks
Dear Members, Any update on this thread?? Thanks
(In reply to Tim-Philipp Müller from comment #1) > Thanks for the bug report. > > Could you make such a file available somewhere by any chance? Hi Tim , I've made the file available , is there any update on this issue??Thanks
I can't seem to reproduce the issue. Can you provide an exact sequence of actions to reproduce the issue ?
(In reply to Edward Hervey from comment #7) > I can't seem to reproduce the issue. > > Can you provide an exact sequence of actions to reproduce the issue ? Hi Edward , you can either seek to end or play file normally till the end , i mean to say in any way try to play the file till the length the issue can be reproduced , make sure you use the second file.Thanks
I still can't reproduce it. Can you provide a gst debug log ? GST_DEBUG_FILE=/tmp/log GST_DEBUG=2,*asf*:6 <yourapplication> Compress the resulting /tmp/log file and attach it here.
(In reply to Edward Hervey from comment #10) > I still can't reproduce it. > > Can you provide a gst debug log ? > > GST_DEBUG_FILE=/tmp/log GST_DEBUG=2,*asf*:6 <yourapplication> > > Compress the resulting /tmp/log file and attach it here. I'll share the logs, meanwhile can you share the cmd line pipeline or anything that you used to reproduce this issue , so i can cross-verify this issue.I am able to reproduce this issue 100/100 times. So it's quite confusing why you are not able to reproduce it.Thanks
I can reproduce this with the second ASF file (md5sum=254411b8...) in push mode: gst-launch-1.0 filesrc location= 783100-ASF_VC-1\(MP@LL\)_WMA\(Version\ 2\)_308k_384_1h53mn_176X114_23.976fps_CZ12.2012.asf ! queue ! asfdemux ! fakesink ... ERROR: from element /GstPipeline:pipeline0/GstASFDemux:asfdemux0: The stream is of a different type than handled by this element. Additional debug info: gstasfdemux.c(1015): gst_asf_demux_chain_headers (): /GstPipeline:pipeline0/GstASFDemux:asfdemux0: This doesn't seem to be an ASF file
Marking as blocker, since it appears to be a regressions.
This should fix it. Please reopen if not. commit 42f8e22f173250c2ebe336ef019e01abcdc06915 Author: Thiago Santos <thiagossantos@gmail.com> Date: Tue Jun 6 21:48:19 2017 -0700 asfdemux: fix checking for header objects The if 'and' condition was mixed up with the last parameter of the function and would lead to wrong results Merged to 1.12: commit 45e54dd7a2a1e0d066d09b8f9b613a4427bc9fdc Author: Thiago Santos <thiagossantos@gmail.com> Date: Tue Jun 6 21:48:19 2017 -0700 asfdemux: fix checking for header objects The if 'and' condition was mixed up with the last parameter of the function and would lead to wrong results