On Thu, Oct 22, 2015 at 11:48 PM, Thierry Foucu <tfo...@gmail.com> wrote: > On Thu, Oct 22, 2015 at 2:29 PM, Hendrik Leppkes <h.lepp...@gmail.com> > wrote: > >> On Thu, Oct 22, 2015 at 11:15 PM, Thierry Foucu <tfo...@gmail.com> wrote: >> > On Thu, Oct 22, 2015 at 2:08 PM, Hendrik Leppkes <h.lepp...@gmail.com> >> > wrote: >> > >> >> On Thu, Oct 22, 2015 at 9:19 PM, Thierry Foucu <tfo...@gmail.com> >> wrote: >> >> > --- >> >> > libavformat/riff.c | 1 + >> >> > 1 file changed, 1 insertion(+) >> >> > >> >> > diff --git a/libavformat/riff.c b/libavformat/riff.c >> >> > index a9197e7..7de8936 100644 >> >> > --- a/libavformat/riff.c >> >> > +++ b/libavformat/riff.c >> >> > @@ -487,5 +487,6 @@ const AVCodecGuid ff_codec_wav_guids[] = { >> >> > { AV_CODEC_ID_ATRAC3P, { 0xBF, 0xAA, 0x23, 0xE9, 0x58, 0xCB, >> 0x71, >> >> 0x44, 0xA1, 0x19, 0xFF, 0xFA, 0x01, 0xE4, 0xCE, 0x62 } }, >> >> > { AV_CODEC_ID_EAC3, { 0xAF, 0x87, 0xFB, 0xA7, 0x02, 0x2D, >> 0xFB, >> >> 0x42, 0xA4, 0xD4, 0x05, 0xCD, 0x93, 0x84, 0x3B, 0xDD } }, >> >> > { AV_CODEC_ID_MP2, { 0x2B, 0x80, 0x6D, 0xE0, 0x46, 0xDB, >> 0xCF, >> >> 0x11, 0xB4, 0xD1, 0x00, 0x80, 0x5F, 0x6C, 0xBB, 0xEA } }, >> >> > + { AV_CODEC_ID_PCM_S24LE,{ 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, >> 0x00, >> >> 0x00, 0x00, 0x00, 0x10, 0x00, 0x80, 0x00, 0x00, 0xAA } }, >> >> > { AV_CODEC_ID_NONE } >> >> > }; >> >> > -- >> >> >> >> This is the generic PCM GUID, isn't it? It could be any PCM bitdepth, >> >> so the patch does not appear correct to me. >> >> >> > >> > Should i then look for the generic PCM GUID (to decide this is a PCM >> codec) >> > and base on the bits_per_coded_sample, selects the PCM codec? >> > >> > >> >> From what I can tell such files should already be handled by the >> generic WAVEFORMATEXTENSIBLE handler. >> Can you share a file which doesn't work as expected? >> >> > No pb. just uploaded to the ftp a file name > raw_video_pcm_s24le.avi >
That tool sure is writing weird broken files. The usual base GUID is {xxxxxxxx-0000-0010-8000-00AA00389B71} This thing writes {xxxxxxxx-0000-0000-0000-1000800000aa} and then has 8 bytes of extradata which contain the remainder of the GUID (and 4 zeros) It seems like they screwed something up and shifted the proper GUID by 4 bytes after the expected field. Does any software play this PCM track? - Hendrik _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel