On Mon, 27 Feb 2012 14:12:56 -0500, Justin Ruggles <[email protected]>
wrote:
> The time base is 1 / sample_rate, not 90000.
> Several more codecs also encode the sample count in the first 4 bytes of the
> chunk.
> ---
> libavformat/electronicarts.c | 25 +++++++++++++++++--------
> 1 files changed, 17 insertions(+), 8 deletions(-)
>
> diff --git a/libavformat/electronicarts.c b/libavformat/electronicarts.c
> index a10e645..d772f95 100644
> --- a/libavformat/electronicarts.c
> +++ b/libavformat/electronicarts.c
> @@ -493,22 +493,31 @@ static int ea_read_packet(AVFormatContext *s,
> num_samples = avio_rl32(pb);
> avio_skip(pb, 8);
> chunk_size -= 12;
> + } else if (ea->audio_codec == CODEC_ID_ADPCM_EA ||
> + ea->audio_codec == CODEC_ID_ADPCM_EA_R1 ||
> + ea->audio_codec == CODEC_ID_ADPCM_EA_R2 ||
> + ea->audio_codec == CODEC_ID_ADPCM_IMA_EA_EACS) {
> + num_samples = avio_rl32(pb);
> + avio_seek(pb, -4, SEEK_CUR);
> + } else if (ea->audio_codec == CODEC_ID_ADPCM_EA_R3) {
> + num_samples = avio_rb32(pb);
> + avio_seek(pb, -4, SEEK_CUR);
> + } else if (ea->audio_codec == CODEC_ID_ADPCM_IMA_EA_SEAD) {
> + num_samples = chunk_size * 2 / ea->num_channels;
> }
Might be better to put those after av_get_packet so you get rid of the
seeks.
> ret = av_get_packet(pb, pkt, chunk_size);
> if (ret < 0)
> return ret;
> pkt->stream_index = ea->audio_stream_index;
> - pkt->pts = 90000;
> - pkt->pts *= ea->audio_frame_counter;
> - pkt->pts /= ea->sample_rate;
> + pkt->pts = ea->audio_frame_counter;
>
> switch (ea->audio_codec) {
> case CODEC_ID_ADPCM_EA:
> - /* 2 samples/byte, 1 or 2 samples per frame depending
> - * on stereo; chunk also has 12-byte header */
> - ea->audio_frame_counter += ((chunk_size - 12) * 2) /
> - ea->num_channels;
> - break;
> + case CODEC_ID_ADPCM_EA_R1:
> + case CODEC_ID_ADPCM_EA_R2:
> + case CODEC_ID_ADPCM_EA_R3:
> + case CODEC_ID_ADPCM_IMA_EA_EACS:
> + case CODEC_ID_ADPCM_IMA_EA_SEAD:
> case CODEC_ID_PCM_S16LE_PLANAR:
> case CODEC_ID_MP3:
> ea->audio_frame_counter += num_samples;
> --
> 1.7.1
Otherwise looks ok.
--
Anton Khirnov
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
