[FFmpeg-devel] [PATCH] mov.c: factorize ff_mov_read_stsd_entries()

Ronald S. Bultje rsbultje
Sun Mar 15 14:07:51 CET 2009


On Sun, Mar 15, 2009 at 6:57 AM, Baptiste Coudurier
<baptiste.coudurier at gmail.com> wrote:
> Well this MSRLE file uses multiple stsd, don't waste your time.
> I finished with multi stsd support, and I already factorized the stsd
> parsing code. Works quite well.


+        sc->stsd_data[i].size = size;
+        AV_WB32(sc->stsd_data[i].data, size);
+        get_buffer(pb, sc->stsd_data[i].data+4, size-4);

If you're saving the size anyway, no need to prepend it to the data.
I'll cut it off myself from the SDP.

> Now we need a way to concatenate these tracks in one output track in
> ffmpeg.c :)

That sort of worries me. :-). I agree with this approach btw, since
you could get STSDs where the codecs are different etc., and all that
is legal (and probably editlists, etc.). I think this is right, but
I'm not sure how to "make it work".

But thanks! I'll redo my rtsp-x_qt.patch based on this work.


More information about the ffmpeg-devel mailing list