[FFmpeg-devel] [PATCH] ADTS AAC with ID3v2

Michael Niedermayer michaelni
Wed Jan 14 18:57:52 CET 2009


On Wed, Jan 14, 2009 at 10:17:33AM -0500, Alex Converse wrote:
> On Wed, Jan 14, 2009 at 10:10 AM, Michael Niedermayer <michaelni at gmx.at>wrote:
> 
> > On Wed, Jan 14, 2009 at 09:47:29AM -0500, Alex Converse wrote:
> > > On Wed, Jan 14, 2009 at 9:17 AM, Michael Niedermayer <michaelni at gmx.at
> > >wrote:
> > >
> > > > On Wed, Jan 14, 2009 at 12:23:00AM -0500, Alex Converse wrote:
> > > > > Hi,
> > > > >
> > > > > The attached patch adds support to the ADTS AAC probe to detect AAC
> > files
> > > > > with ID3v2 tags at a higher level than the MP3 probe and the MPC
> > probe.
> > > > It
> > > > > refactors some id3 detection from mp3.c into id3v2.c.
> > > > >
> > > > [...]
> > > >
> > > > >  #define ID3v1_TAG_SIZE 128
> > > > > -
> > > > >  #define ID3v1_GENRE_MAX 125
> > > > >
> > > > >  static const char * const id3v1_genre_str[ID3v1_GENRE_MAX + 1] = {
> > > >
> > > > cosmetic
> > > >
> > > > [...]
> > > > > @@ -487,7 +472,7 @@ static int mp3_read_header(AVFormatContext *s,
> > > > >      ret = get_buffer(s->pb, buf, ID3v2_HEADER_SIZE);
> > > > >      if (ret != ID3v2_HEADER_SIZE)
> > > > >          return -1;
> > > > > -    if (id3v2_match(buf)) {
> > > > > +    if (ff_id3v2_match(buf)) {
> > > > >          /* parse ID3v2 header */
> > > > >          len = ((buf[6] & 0x7f) << 21) |
> > > > >              ((buf[7] & 0x7f) << 14) |
> > > >
> > > > the split out of (ff_)id3v2_match is ok but should be a seperate patch
> > > > and commit
> > > >
> > > > [...]
> > > > >      }
> > > > > -    if   (first_frames>=3) return AVPROBE_SCORE_MAX/2+1;
> > > > > +    if   (first_frames>=3) return AVPROBE_SCORE_MAX/2+2;
> > > > >      else if(max_frames>500)return AVPROBE_SCORE_MAX/2;
> > > > >      else if(max_frames>=3) return AVPROBE_SCORE_MAX/4;
> > > > >      else if(max_frames>=1) return 1;
> > > >
> > > > why?
> > >
> > >
> > > Just having an ID3v2 tag gives a score of AVPROBE_SCORE_MAX/2+1 for mp3.
> >
> > That sounds like a bug ...
> >
> 
> The comment about mpeg-ps in mp3.c made me think it was a hackish design
> decision, not a bug.

well one could call it that way, still the existence of hackish code is no
argument to add more hackish code like the +1 -> +2

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Breaking DRM is a little like attempting to break through a door even
though the window is wide open and the only thing in the house is a bunch
of things you dont want and which you would get tomorrow for free anyway
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090114/64af723b/attachment.pgp>



More information about the ffmpeg-devel mailing list