[FFmpeg-trac] #3723(undetermined:new): aac copy issue with -bsf:a aac_adtstoasc when container is not flv, m4a, mov or mp4

FFmpeg trac at avcodec.org
Wed Jun 18 09:08:59 CEST 2014

#3723: aac copy issue with -bsf:a aac_adtstoasc when container is not flv, m4a,
mov or mp4
             Reporter:  feelart      |                    Owner:
                 Type:  defect       |                   Status:  new
             Priority:  normal       |                Component:
              Version:  git-master   |  undetermined
             Keywords:               |               Resolution:
             Blocking:               |               Blocked By:
Analyzed by developer:  0            |  Reproduced by developer:  0

Comment (by feelart):

 >It is possible that this specific issue - avi
 It is not at all specific to avi, it is also for ac3, eac3, wav, wmv, etc.
 It applies to ALL CONTAINERS to the exception of: m4a, flv mp4, mov.

 > if invalid aac stream are written - gets fixed but please understand
 that generally, ffmpeg (and the libraries) cannot guess what you want.
 What do you mean? To my knowledge I do no see any guess in the command
 line, only a specific instruction to perform.
 Furthermore, if you instruct

 ffmpeg -i aac_test.aac -c:a copy aac_complain.flv

 Than ffmpeg does complain and stop execution.

 > If you specify a bitstream filter on the command line, you should expect
 that it is used and if you don't specify a bitstream filter, it won't be
 Normal, I do expect so, but one also expects that if one instructs
 something wrong, than at least a warning is raised. In the present case it
 is not.
 Try for instance:

 ffmpeg -i someOpusFile.opus -c:a copy someOpusFile.mp4
 It won't work, but you wouldn't be "happy" if a "dummy" corrupted file was

 >There is a feature request for automatically inserting bitstream filters
 but this is not trivial and it is not what you request in your ticket
 The minimun request is to raise execution warning, if not a stop (stop is
 definitely desired for batch processing).
 As there is already some IF clause to test for [m4a, flv mp4, mov], all I
 request is some ELSE clause.

 >Please understand that it is not the fault of the FFmpeg developers that
 different methods of writing aac into containers are defined.
 I do understand.
 All I try is to help making ffmpeg better and reliable, by I say that
 internally ffmpeg ought to handle whether or not to add a bitstream
 filter, instead of each user having to deal with it and every developer to
 write some switch statement - probably more by trial and error, than by
 aac specification knowledge.

Ticket URL: <https://trac.ffmpeg.org/ticket/3723#comment:6>
FFmpeg <https://ffmpeg.org>
FFmpeg issue tracker

More information about the FFmpeg-trac mailing list