diff --git a/libavcodec/libx264.c b/libavcodec/libx264.c index 39a7d7db8e..d23ced1c0d 100644 --- a/libavcodec/libx264.c +++ b/libavcodec/libx264.c @@ -72,6 +72,7 @@ typedef struct X264Context { int direct_pred; int slice_max_size; char *stats; + int nal_hrd } X264Context; static void X264_log(void *p, int level, const char *fmt, va_list args) @@ -481,6 +482,10 @@ static av_cold int X264_init(AVCodecContext *avctx) default: break; } + + if (x4->nal_hrd >= 0) + x4->params.i_nal_hrd = x4->nal_hrd; + if (x4->profile) if (x264_param_apply_profile(&x4->params, x4->profile) < 0) { int i; @@ -643,6 +648,11 @@ static const AVOption options[] = { { "auto", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = X264_DIRECT_PRED_AUTO }, 0, 0, VE, "direct-pred" }, { "slice-max-size","Limit the size of each slice in bytes", OFFSET(slice_max_size),AV_OPT_TYPE_INT, { .i64 = -1 }, -1, INT_MAX, VE }, { "stats", "Filename for 2 pass stats", OFFSET(stats), AV_OPT_TYPE_STRING, { 0 }, 0, 0, VE }, + { "nal-hrd", "Signal HRD information (requires vbv-bufsize; " + "cbr not allowed in .mp4)", OFFSET(nal_hrd), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, INT_MAX, VE, "nal-hrd" }, + { "none", NULL, 0, AV_OPT_TYPE_CONST, {.i64 = X264_NAL_HRD_NONE}, INT_MIN, INT_MAX, VE, "nal-hrd" }, + { "vbr", NULL, 0, AV_OPT_TYPE_CONST, {.i64 = X264_NAL_HRD_VBR}, INT_MIN, INT_MAX, VE, "nal-hrd" }, + { "cbr", NULL, 0, AV_OPT_TYPE_CONST, {.i64 = X264_NAL_HRD_CBR}, INT_MIN, INT_MAX, VE, "nal-hrd" }, { NULL }, }; diff --git a/libavformat/isom.h b/libavformat/isom.h index 6565bdcc17..875c3a9fdf 100644 --- a/libavformat/isom.h +++ b/libavformat/isom.h @@ -87,6 +87,11 @@ typedef struct { unsigned flags; } MOVTrackExt; +typedef struct { + unsigned int count; + unsigned int index; +} MOVSbgp; + typedef struct MOVStreamContext { AVIOContext *pb; int ffindex; ///< AVStream index @@ -136,6 +141,8 @@ typedef struct MOVStreamContext { uint32_t tmcd_flags; ///< tmcd track flags int64_t track_end; ///< used for dts generation in fragmented movie files int start_pad; ///< amount of samples to skip due to enc-dec delay + unsigned int rap_group_count; + MOVSbgp *rap_group; } MOVStreamContext; typedef struct MOVContext { diff --git a/libavformat/matroskadec.c b/libavformat/matroskadec.c index eb34ef4ee9..f4d7071352 100644 --- a/libavformat/matroskadec.c +++ b/libavformat/matroskadec.c @@ -2146,7 +2146,7 @@ static int matroska_parse_frame(MatroskaDemuxContext *matroska, static int matroska_parse_block(MatroskaDemuxContext *matroska, uint8_t *data, int size, int64_t pos, uint64_t cluster_time, - uint64_t duration, int is_keyframe, + uint64_t block_duration, int is_keyframe, int64_t cluster_pos) { uint64_t timecode = AV_NOPTS_VALUE; @@ -2175,7 +2175,7 @@ static int matroska_parse_block(MatroskaDemuxContext *matroska, uint8_t *data, st = track->stream; if (st->discard >= AVDISCARD_ALL) return res; - av_assert1(duration != AV_NOPTS_VALUE); + av_assert1(block_duration != AV_NOPTS_VALUE); block_time = AV_RB16(data); data += 2; @@ -2211,14 +2211,15 @@ static int matroska_parse_block(MatroskaDemuxContext *matroska, uint8_t *data, if (res) goto end; - if (!duration) - duration = track->default_duration * laces / matroska->time_scale; + if (!block_duration) + block_duration = track->default_duration * laces / matroska->time_scale; if (cluster_time != (uint64_t)-1 && (block_time >= 0 || cluster_time >= -block_time)) - track->end_timecode = FFMAX(track->end_timecode, timecode+duration); + track->end_timecode = + FFMAX(track->end_timecode, timecode + block_duration); for (n = 0; n < laces; n++) { - int64_t lace_duration = duration*(n+1) / laces - duration*n / laces; + int64_t lace_duration = block_duration*(n+1) / laces - block_duration*n / laces; if (lace_size[n] > size) { av_log(matroska->ctx, AV_LOG_ERROR, "Invalid packet size\n"); @@ -2232,7 +2233,7 @@ static int matroska_parse_block(MatroskaDemuxContext *matroska, uint8_t *data, st->codec->block_align && track->audio.sub_packet_size) { res = matroska_parse_rm_audio(matroska, track, st, data, size, - timecode, duration, pos); + timecode, lace_duration, pos); if (res) goto end; diff --git a/libavformat/mov.c b/libavformat/mov.c index c937174ba3..84565c8696 100644 --- a/libavformat/mov.c +++ b/libavformat/mov.c @@ -1879,6 +1879,46 @@ static int mov_read_ctts(MOVContext *c, AVIOContext *pb, MOVAtom atom) return 0; } +static int mov_read_sbgp(MOVContext *c, AVIOContext *pb, MOVAtom atom) +{ + AVStream *st; + MOVStreamContext *sc; + unsigned int i, entries; + uint8_t version; + uint32_t grouping_type; + + if (c->fc->nb_streams < 1) + return 0; + st = c->fc->streams[c->fc->nb_streams-1]; + sc = st->priv_data; + + version = avio_r8(pb); /* version */ + avio_rb24(pb); /* flags */ + grouping_type = avio_rl32(pb); + if (grouping_type != MKTAG( 'r','a','p',' ')) + return 0; /* only support 'rap ' grouping */ + if (version == 1) + avio_rb32(pb); /* grouping_type_parameter */ + + entries = avio_rb32(pb); + if (!entries) + return 0; + if (entries >= UINT_MAX / sizeof(*sc->rap_group)) + return AVERROR_INVALIDDATA; + sc->rap_group = av_malloc(entries * sizeof(*sc->rap_group)); + if (!sc->rap_group) + return AVERROR(ENOMEM); + + for (i = 0; i < entries && !pb->eof_reached; i++) { + sc->rap_group[i].count = avio_rb32(pb); /* sample_count */ + sc->rap_group[i].index = avio_rb32(pb); /* group_description_index */ + } + + sc->rap_group_count = i; + + return pb->eof_reached ? AVERROR_EOF : 0; +} + static void mov_build_index(MOVContext *mov, AVStream *st) { MOVStreamContext *sc = st->priv_data; @@ -1914,6 +1954,9 @@ static void mov_build_index(MOVContext *mov, AVStream *st) unsigned int stts_sample = 0; unsigned int sample_size; unsigned int distance = 0; + unsigned int rap_group_index = 0; + unsigned int rap_group_sample = 0; + int rap_group_present = sc->rap_group_count && sc->rap_group; int key_off = (sc->keyframe_count && sc->keyframes[0] > 0) || (sc->stps_data && sc->stps_data[0] > 0); current_dts -= sc->dts_shift; @@ -1949,6 +1992,14 @@ static void mov_build_index(MOVContext *mov, AVStream *st) if (stps_index + 1 < sc->stps_count) stps_index++; } + if (rap_group_present && rap_group_index < sc->rap_group_count) { + if (sc->rap_group[rap_group_index].index > 0) + keyframe = 1; + if (++rap_group_sample == sc->rap_group[rap_group_index].count) { + rap_group_sample = 0; + rap_group_index++; + } + } if (keyframe) distance = 0; sample_size = sc->alt_sample_size > 0 ? sc->alt_sample_size : sc->sample_sizes[current_sample]; @@ -2204,6 +2255,7 @@ static int mov_read_trak(MOVContext *c, AVIOContext *pb, MOVAtom atom) av_freep(&sc->keyframes); av_freep(&sc->stts_data); av_freep(&sc->stps_data); + av_freep(&sc->rap_group); return 0; } @@ -2704,6 +2756,7 @@ static const MOVParseTableEntry mov_default_parse_table[] = { { MKTAG('c','m','o','v'), mov_read_cmov }, { MKTAG('c','h','a','n'), mov_read_chan }, /* channel layout */ { MKTAG('d','v','c','1'), mov_read_dvc1 }, +{ MKTAG('s','b','g','p'), mov_read_sbgp }, { 0, NULL } }; diff --git a/libavformat/mp3dec.c b/libavformat/mp3dec.c index 5bd0e9648b..db01cd2081 100644 --- a/libavformat/mp3dec.c +++ b/libavformat/mp3dec.c @@ -29,6 +29,12 @@ #include "id3v1.h" #include "libavcodec/mpegaudiodecheader.h" +#define XING_FLAG_FRAMES 0x01 +#define XING_FLAG_SIZE 0x02 +#define XING_FLAG_TOC 0x04 + +#define XING_TOC_COUNT 100 + typedef struct { int64_t filesize; int start_pad; @@ -79,6 +85,26 @@ static int mp3_read_probe(AVProbeData *p) //mpegps_mp3_unrecognized_format.mpg has max_frames=3 } +static void read_xing_toc(AVFormatContext *s, int64_t filesize, int64_t duration) +{ + int i; + + if (!filesize && + !(filesize = avio_size(s->pb))) { + av_log(s, AV_LOG_WARNING, "Cannot determine file size, skipping TOC table.\n"); + return; + } + + for (i = 0; i < XING_TOC_COUNT; i++) { + uint8_t b = avio_r8(s->pb); + + av_add_index_entry(s->streams[0], + av_rescale(b, filesize, 256), + av_rescale(i, duration, XING_TOC_COUNT), + 0, 0, AVINDEX_KEYFRAME); + } +} + /** * Try to find Xing/Info/VBRI tags and compute duration from info therein */ @@ -101,17 +127,20 @@ static int mp3_parse_vbr_tags(AVFormatContext *s, AVStream *st, int64_t base) if(c.layer != 3) return -1; + spf = c.lsf ? 576 : 1152; /* Samples per frame, layer 3 */ + /* Check for Xing / Info tag */ avio_skip(s->pb, xing_offtbl[c.lsf == 1][c.nb_channels == 1]); v = avio_rb32(s->pb); if(v == MKBETAG('X', 'i', 'n', 'g') || v == MKBETAG('I', 'n', 'f', 'o')) { v = avio_rb32(s->pb); - if(v & 0x1) + if(v & XING_FLAG_FRAMES) frames = avio_rb32(s->pb); - if(v & 0x2) + if(v & XING_FLAG_SIZE) size = avio_rb32(s->pb); - if(v & 4) - avio_skip(s->pb, 100); + if (v & XING_FLAG_TOC && frames) + read_xing_toc(s, size, av_rescale_q(frames, (AVRational){spf, c.sample_rate}, + st->time_base)); if(v & 8) avio_skip(s->pb, 4); @@ -145,7 +174,6 @@ static int mp3_parse_vbr_tags(AVFormatContext *s, AVStream *st, int64_t base) /* Skip the vbr tag frame */ avio_seek(s->pb, base + vbrtag_size, SEEK_SET); - spf = c.lsf ? 576 : 1152; /* Samples per frame, layer 3 */ if(frames) st->duration = av_rescale_q(frames, (AVRational){spf, c.sample_rate}, st->time_base); @@ -222,14 +250,36 @@ static int mp3_read_packet(AVFormatContext *s, AVPacket *pkt) return ret; } -static int read_seek(AVFormatContext *s, int stream_index, int64_t timestamp, int flags) +static int mp3_seek(AVFormatContext *s, int stream_index, int64_t timestamp, + int flags) { MP3Context *mp3 = s->priv_data; - AVStream *st = s->streams[stream_index]; + AVIndexEntry *ie; + AVStream *st = s->streams[0]; + int64_t ret = av_index_search_timestamp(st, timestamp, flags); + uint32_t header = 0; - st->skip_samples = timestamp <= 0 ? mp3->start_pad + 528 + 1 : 0; + if (ret < 0) + return ret; - return -1; + ie = &st->index_entries[ret]; + ret = avio_seek(s->pb, ie->pos, SEEK_SET); + if (ret < 0) + return ret; + + while (!s->pb->eof_reached) { + header = (header << 8) + avio_r8(s->pb); + if (ff_mpa_check_header(header) >= 0) { + ff_update_cur_dts(s, st, ie->timestamp); + ret = avio_seek(s->pb, -4, SEEK_CUR); + + st->skip_samples = ie->timestamp <= 0 ? mp3->start_pad + 528 + 1 : 0; + + return (ret >= 0) ? 0 : ret; + } + } + + return AVERROR_EOF; } AVInputFormat ff_mp3_demuxer = { @@ -239,7 +289,7 @@ AVInputFormat ff_mp3_demuxer = { .read_probe = mp3_read_probe, .read_header = mp3_read_header, .read_packet = mp3_read_packet, - .read_seek = read_seek, + .read_seek = mp3_seek, .flags = AVFMT_GENERIC_INDEX, .extensions = "mp2,mp3,m2a", /* XXX: use probe */ };