diff options
Diffstat (limited to 'libavcodec/mpegaudiodec.c')
-rw-r--r-- | libavcodec/mpegaudiodec.c | 89 |
1 files changed, 57 insertions, 32 deletions
diff --git a/libavcodec/mpegaudiodec.c b/libavcodec/mpegaudiodec.c index 7dded97cf4..5cc3c74339 100644 --- a/libavcodec/mpegaudiodec.c +++ b/libavcodec/mpegaudiodec.c @@ -79,6 +79,7 @@ typedef struct MPADecodeContext { int err_recognition; AVCodecContext* avctx; MPADSPContext mpadsp; + AVFrame frame; } MPADecodeContext; #if CONFIG_FLOAT @@ -479,6 +480,10 @@ static av_cold int decode_init(AVCodecContext * avctx) if (avctx->codec_id == CODEC_ID_MP3ADU) s->adu_mode = 1; + + avcodec_get_frame_defaults(&s->frame); + avctx->coded_frame = &s->frame; + return 0; } @@ -1581,7 +1586,7 @@ static int mp_decode_layer3(MPADecodeContext *s) static int mp_decode_frame(MPADecodeContext *s, OUT_INT *samples, const uint8_t *buf, int buf_size) { - int i, nb_frames, ch; + int i, nb_frames, ch, ret; OUT_INT *samples_ptr; init_get_bits(&s->gb, buf + HEADER_SIZE, (buf_size - HEADER_SIZE) * 8); @@ -1629,8 +1634,16 @@ static int mp_decode_frame(MPADecodeContext *s, OUT_INT *samples, assert(i <= buf_size - HEADER_SIZE && i >= 0); memcpy(s->last_buf + s->last_buf_size, s->gb.buffer + buf_size - HEADER_SIZE - i, i); s->last_buf_size += i; + } - break; + /* get output buffer */ + if (!samples) { + s->frame.nb_samples = s->avctx->frame_size; + if ((ret = s->avctx->get_buffer(s->avctx, &s->frame)) < 0) { + av_log(s->avctx, AV_LOG_ERROR, "get_buffer() failed\n"); + return ret; + } + samples = (OUT_INT *)s->frame.data[0]; } /* apply the synthesis filter */ @@ -1650,7 +1663,7 @@ static int mp_decode_frame(MPADecodeContext *s, OUT_INT *samples, return nb_frames * 32 * sizeof(OUT_INT) * s->nb_channels; } -static int decode_frame(AVCodecContext * avctx, void *data, int *data_size, +static int decode_frame(AVCodecContext * avctx, void *data, int *got_frame_ptr, AVPacket *avpkt) { const uint8_t *buf = avpkt->data; @@ -1658,7 +1671,6 @@ static int decode_frame(AVCodecContext * avctx, void *data, int *data_size, MPADecodeContext *s = avctx->priv_data; uint32_t header; int out_size; - OUT_INT *out_samples = data; if (buf_size < HEADER_SIZE) return AVERROR_INVALIDDATA; @@ -1681,10 +1693,6 @@ static int decode_frame(AVCodecContext * avctx, void *data, int *data_size, avctx->bit_rate = s->bit_rate; avctx->sub_id = s->layer; - if (*data_size < avctx->frame_size * avctx->channels * sizeof(OUT_INT)) - return AVERROR(EINVAL); - *data_size = 0; - if (s->frame_size <= 0 || s->frame_size > buf_size) { av_log(avctx, AV_LOG_ERROR, "incomplete frame\n"); return AVERROR_INVALIDDATA; @@ -1693,9 +1701,10 @@ static int decode_frame(AVCodecContext * avctx, void *data, int *data_size, buf_size= s->frame_size; } - out_size = mp_decode_frame(s, out_samples, buf, buf_size); + out_size = mp_decode_frame(s, NULL, buf, buf_size); if (out_size >= 0) { - *data_size = out_size; + *got_frame_ptr = 1; + *(AVFrame *)data = s->frame; avctx->sample_rate = s->sample_rate; //FIXME maybe move the other codec info stuff from above here too } else { @@ -1704,6 +1713,7 @@ static int decode_frame(AVCodecContext * avctx, void *data, int *data_size, If there is more data in the packet, just consume the bad frame instead of returning an error, which would discard the whole packet. */ + *got_frame_ptr = 0; if (buf_size == avpkt->size) return out_size; } @@ -1719,15 +1729,14 @@ static void flush(AVCodecContext *avctx) } #if CONFIG_MP3ADU_DECODER || CONFIG_MP3ADUFLOAT_DECODER -static int decode_frame_adu(AVCodecContext *avctx, void *data, int *data_size, - AVPacket *avpkt) +static int decode_frame_adu(AVCodecContext *avctx, void *data, + int *got_frame_ptr, AVPacket *avpkt) { const uint8_t *buf = avpkt->data; int buf_size = avpkt->size; MPADecodeContext *s = avctx->priv_data; uint32_t header; int len, out_size; - OUT_INT *out_samples = data; len = buf_size; @@ -1757,9 +1766,6 @@ static int decode_frame_adu(AVCodecContext *avctx, void *data, int *data_size, avctx->bit_rate = s->bit_rate; avctx->sub_id = s->layer; - if (*data_size < avctx->frame_size * avctx->channels * sizeof(OUT_INT)) - return AVERROR(EINVAL); - s->frame_size = len; #if FF_API_PARSE_FRAME @@ -1767,9 +1773,11 @@ static int decode_frame_adu(AVCodecContext *avctx, void *data, int *data_size, out_size = buf_size; else #endif - out_size = mp_decode_frame(s, out_samples, buf, buf_size); + out_size = mp_decode_frame(s, NULL, buf, buf_size); + + *got_frame_ptr = 1; + *(AVFrame *)data = s->frame; - *data_size = out_size; return buf_size; } #endif /* CONFIG_MP3ADU_DECODER || CONFIG_MP3ADUFLOAT_DECODER */ @@ -1780,6 +1788,7 @@ static int decode_frame_adu(AVCodecContext *avctx, void *data, int *data_size, * Context for MP3On4 decoder */ typedef struct MP3On4DecodeContext { + AVFrame *frame; int frames; ///< number of mp3 frames per block (number of mp3 decoder instances) int syncword; ///< syncword patch const uint8_t *coff; ///< channel offsets in output buffer @@ -1843,7 +1852,8 @@ static int decode_init_mp3on4(AVCodecContext * avctx) return AVERROR_INVALIDDATA; } - avpriv_mpeg4audio_get_config(&cfg, avctx->extradata, avctx->extradata_size); + avpriv_mpeg4audio_get_config(&cfg, avctx->extradata, + avctx->extradata_size * 8, 1); if (!cfg.chan_config || cfg.chan_config > 7) { av_log(avctx, AV_LOG_ERROR, "Invalid channel config number.\n"); return AVERROR_INVALIDDATA; @@ -1870,6 +1880,7 @@ static int decode_init_mp3on4(AVCodecContext * avctx) // Put decoder context in place to make init_decode() happy avctx->priv_data = s->mp3decctx[0]; decode_init(avctx); + s->frame = avctx->coded_frame; // Restore mp3on4 context pointer avctx->priv_data = s; s->mp3decctx[0]->adu_mode = 1; // Set adu mode @@ -1914,9 +1925,8 @@ static void flush_mp3on4(AVCodecContext *avctx) } -static int decode_frame_mp3on4(AVCodecContext * avctx, - void *data, int *data_size, - AVPacket *avpkt) +static int decode_frame_mp3on4(AVCodecContext *avctx, void *data, + int *got_frame_ptr, AVPacket *avpkt) { const uint8_t *buf = avpkt->data; int buf_size = avpkt->size; @@ -1924,14 +1934,17 @@ static int decode_frame_mp3on4(AVCodecContext * avctx, MPADecodeContext *m; int fsize, len = buf_size, out_size = 0; uint32_t header; - OUT_INT *out_samples = data; + OUT_INT *out_samples; OUT_INT *outptr, *bp; - int fr, j, n, ch; + int fr, j, n, ch, ret; - if (*data_size < MPA_FRAME_SIZE * avctx->channels * sizeof(OUT_INT)) { - av_log(avctx, AV_LOG_ERROR, "output buffer is too small\n"); - return AVERROR(EINVAL); + /* get output buffer */ + s->frame->nb_samples = MPA_FRAME_SIZE; + if ((ret = avctx->get_buffer(avctx, s->frame)) < 0) { + av_log(avctx, AV_LOG_ERROR, "get_buffer() failed\n"); + return ret; } + out_samples = (OUT_INT *)s->frame->data[0]; // Discard too short frames if (buf_size < HEADER_SIZE) @@ -1990,7 +2003,10 @@ static int decode_frame_mp3on4(AVCodecContext * avctx, /* update codec info */ avctx->sample_rate = s->mp3decctx[0]->sample_rate; - *data_size = out_size; + s->frame->nb_samples = out_size / (avctx->channels * sizeof(OUT_INT)); + *got_frame_ptr = 1; + *(AVFrame *)data = *s->frame; + return buf_size; } #endif /* CONFIG_MP3ON4_DECODER || CONFIG_MP3ON4FLOAT_DECODER */ @@ -2005,7 +2021,9 @@ AVCodec ff_mp1_decoder = { .init = decode_init, .decode = decode_frame, #if FF_API_PARSE_FRAME - .capabilities = CODEC_CAP_PARSE_ONLY, + .capabilities = CODEC_CAP_PARSE_ONLY | CODEC_CAP_DR1, +#else + .capabilities = CODEC_CAP_DR1, #endif .flush = flush, .long_name = NULL_IF_CONFIG_SMALL("MP1 (MPEG audio layer 1)"), @@ -2020,7 +2038,9 @@ AVCodec ff_mp2_decoder = { .init = decode_init, .decode = decode_frame, #if FF_API_PARSE_FRAME - .capabilities = CODEC_CAP_PARSE_ONLY, + .capabilities = CODEC_CAP_PARSE_ONLY | CODEC_CAP_DR1, +#else + .capabilities = CODEC_CAP_DR1, #endif .flush = flush, .long_name = NULL_IF_CONFIG_SMALL("MP2 (MPEG audio layer 2)"), @@ -2035,7 +2055,9 @@ AVCodec ff_mp3_decoder = { .init = decode_init, .decode = decode_frame, #if FF_API_PARSE_FRAME - .capabilities = CODEC_CAP_PARSE_ONLY, + .capabilities = CODEC_CAP_PARSE_ONLY | CODEC_CAP_DR1, +#else + .capabilities = CODEC_CAP_DR1, #endif .flush = flush, .long_name = NULL_IF_CONFIG_SMALL("MP3 (MPEG audio layer 3)"), @@ -2050,7 +2072,9 @@ AVCodec ff_mp3adu_decoder = { .init = decode_init, .decode = decode_frame_adu, #if FF_API_PARSE_FRAME - .capabilities = CODEC_CAP_PARSE_ONLY, + .capabilities = CODEC_CAP_PARSE_ONLY | CODEC_CAP_DR1, +#else + .capabilities = CODEC_CAP_DR1, #endif .flush = flush, .long_name = NULL_IF_CONFIG_SMALL("ADU (Application Data Unit) MP3 (MPEG audio layer 3)"), @@ -2065,6 +2089,7 @@ AVCodec ff_mp3on4_decoder = { .init = decode_init_mp3on4, .close = decode_close_mp3on4, .decode = decode_frame_mp3on4, + .capabilities = CODEC_CAP_DR1, .flush = flush_mp3on4, .long_name = NULL_IF_CONFIG_SMALL("MP3onMP4"), }; |