[FFmpeg-devel] [PATCH v3] avformat/mp3dec: Fixes misdetection of zYLx.wav

[Limin Wang]

On Thu, Nov 07, 2019 at 01:16:32PM +0100, Hendrik Leppkes wrote:
> On Thu, Nov 7, 2019 at 12:38 PM <lance.lmwang at gmail.com> wrote:
> >
> > From: Limin Wang <lance.lmwang at gmail.com>
> >
> > Signed-off-by: Limin Wang <lance.lmwang at gmail.com>
> > ---
> >  libavformat/mp3dec.c | 5 +++++
> >  1 file changed, 5 insertions(+)
> >
> > diff --git a/libavformat/mp3dec.c b/libavformat/mp3dec.c
> > index 258f19174b..f15045dd6f 100644
> > --- a/libavformat/mp3dec.c
> > +++ b/libavformat/mp3dec.c
> > @@ -73,6 +73,7 @@ static int mp3_read_probe(const AVProbeData *p)
> >      int frames, ret;
> >      int framesizes, max_framesizes;
> >      uint32_t header;
> > +    uint32_t next;
> >      const uint8_t *buf, *buf0, *buf2, *end;
> >
> >      buf0 = p->buf;
> > @@ -93,6 +94,10 @@ static int mp3_read_probe(const AVProbeData *p)
> >              ret = avpriv_mpegaudio_decode_header(&h, header);
> >              if (ret != 0)
> >                  break;
> > +            next = AV_RB32(buf2 + 4);
> > +            /* detect invalid data after header */
> > +            if ((header & 0xfffe0000) == (next & 0xfffe0000))
> > +                break;
> >              buf2 += h.frame_size;
> >              framesizes += h.frame_size;
> >          }
> 
> This seems like one check designed to fix exactly one file. Thats
> really not the kind of checks that seem appropriate here.
> I'm not sure you can conclusively proof that the data after the header
> cannot have a certain bit-pattern, since audio data immediately starts
> here.

0xffe is the sync word for mp3(0xfffe is sync word for MPEG Version 1 layerI)
if the next sync word is near to the first sync word, that's means no frame
data between them, so we can consider it's invalid header. For the
current sample is such pattern, so I check the kind of case only.


> 
> - Hendrik
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".