[FFmpeg-devel] Would like to work on lavf-based concatenation tool for GSoC qualification

Geza Kovacs gkovacs
Wed Apr 1 22:30:59 CEST 2009


Hello all,

I have attached my initial implementation of the lavf-based catenation
tool for GSoC qualification (this does not of course address the full
scope of the project description, it merely performs concatenation of
identical-format files). For now it is implemented as a standalone
application for simplicity (though I could also add it as an option in
ffmpeg's parse_options() if needed, I would simply need details on the
desired invocation syntax for the option).

The application currently supports both avi and mp4 input/output
containers, and appears to work with a variety of codecs (though I have
primarily tested with mpeg4 and huffyuv). According to ffplay there seem
to be issues with some files related to writing the header files,
presumably because I haven't fully populated the oformat struct, but the
files appear to play nevertheless; I will hopefully be able to track
this down before Friday.

I have attempted to follow the described coding and comments standards
as closely as possible, but if there's an issue please let me know.

Thanks for the feedback,
Geza

On Tue, 2009-03-31 at 10:55 -0700, Baptiste Coudurier wrote:
> Hi guys,
> 
> On 3/31/2009 7:26 AM, Robert Swain wrote:
> > On 31/3/09 15:00, Geza Kovacs wrote:
> >> I'd like to know, by "concatenation" I assume this means "without
> >> re-encoding"; therefore I assume I'm not supposed to handle situations
> >> when multiple files are encoded with different framerate settings or
> >> different codecs (or do I instead transcode them into an intermediate
> >> format in these situations?).
> > 
> > Concatenate means to link to together. I'm not sure what the plans are
> > for handling differences in frame rates or resolutions. Differing source
> > formats shouldn't matter for transcoding purposes as long as on entry to
> > the video encoder the resolution, frame rate and pixel format as all
> > appropriate and similarly the number of channels, sample rate and sample
> > format for the audio encoder, there shouldn't be any major problem.
> > 
> > I expect Baptiste will elaborate on this. I think he had plans to allow
> > specification of command line options per input so that one could affect
> > the different inputs to make them concatenate-able.
> 
> Here is the project description:
> 
> Playlist/Concatenation Support
> - primary goal: implement a playlist/concatenation interface to
>     transcode(FFmpeg) and play(FFplay) media
> - interface will use commandline switches.
> - interface must support every input format FFmpeg support
> - interface must work with different input stream parameters (different
>     formats, codecs, video resolution, audio sample rate, audio
>     channels, etc..)
> - interface must support track selection
> - interface must support existing playlist format files .m3u, .pls,
>     xpsf.
> 
> So according to this, different resolution must be handled. About
> different frame rates, this could be added later I think.
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ffmerge.c
Type: text/x-csrc
Size: 14993 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090401/37ac2ba0/attachment.c>



More information about the ffmpeg-devel mailing list