本文深入剖析ffmpeg中的av_read_frame()函数，讲解其在音频视频解码中的作用，如何读取和解析码流数据，以及与av_read_packet()和av_parser_parse2()的交互过程。
参考链接
av_read_frame()
- ffmpeg中的av_read_frame()的作用是读取码流中的音频若干帧或者视频一帧。
- 例如，解码视频的时候，每解码一个视频帧，需要先调用 av_read_frame()获得一帧视频的压缩数据，然后才能对该数据进行解码（例如H.264中一帧压缩数据通常对应一个NAL）
- 通过av_read_packet(***)，读取一个包，需要说明的是此函数必须是包含整数帧的，不存在半帧的情况，以ts流为例，是读取一个完整的PES包（一个完整pes包包含若干视频或音频es包），读取完毕后，通过av_parser_parse2(***)分析出视频一帧（或音频若干帧），返回，下次进入循环的时候，如果上次的数据没有完全取完，则st = s->cur_st;不会是NULL，即再此进入av_parser_parse2(***)流程，而不是下面的av_read_packet（**）流程，这样就保证了，如果读取一次包含了N帧视频数据（以视频为例），则调用av_read_frame（***）N次都不会去读数据，而是返回第一次读取的数据，直到全部解析完毕。
-
av_read_frame()的声明位于libavformat\avformat.h，如下所示
/**
* Return the next frame of a stream.
* This function returns what is stored in the file, and does not validate
* that what is there are valid frames for the decoder. It will split what is
* stored in the file into frames and return one for each call. It will not
* omit invalid data between valid frames so as to give the decoder the maximum
* information possible for decoding.
*
* On success, the returned packet is reference-counted (pkt->buf is set) and
* valid indefinitely. The packet must be freed with av_packet_unref() when
* it is no longer needed. For video, the packet contains exactly one frame.
* For audio, it contains an integer number of frames if each frame has
* a known fixed size (e.g. PCM or ADPCM data). If the audio frames have
* a variable size (e.g. MPEG audio), then it contains one frame.
*
* pkt->pts, pkt->dts and pkt->duration are always set to correct
* values in AVStream.time_base units (and guessed if the format cannot
* provide them). pkt->pts can be AV_NOPTS_VALUE if the video format
* has B-frames, so it is better to rely on pkt->dts if you do not
* decompress the payload.
*
* @return 0 if OK, < 0 on error or end of file. On error, pkt will be blank
* (as if it came from av_packet_alloc()).
*
* @note pkt will be initialized, so it may be uninitialized, but it must not
* contain data that needs to be freed.
*/
int av_read_frame(AVFormatContext *s, AVPacket *pkt);- av_read_frame()使用方法在注释中写得很详细，用中文简单描述一下它的两个参数：
- s：输入的AVFormatContext
- pkt：输出的AVPacket
- 如果返回0则说明读取正常。
函数调用结构图
- 函数调用结构图如下所示
- 可以从源代码中看出，av_read_frame()调用了read_frame_internal()。
int av_read_frame(AVFormatContext *s, AVPacket *pkt)
{
FFFormatContext *const si = ffformatcontext(s);
const int genpts = s->flags & AVFMT_FLAG_GENPTS;
int eof = 0;
int ret;
AVStream *st;
if (!genpts) {
ret = si->packet_buffer.head
? avpriv_packet_list_get(&si->packet_buffer, pkt)
: read_frame_internal(s, pkt);
if (ret < 0)
return ret;
goto return_packet;
}
for (;;) {
PacketListEntry *pktl = si->packet_buffer.head;
if (pktl) {
AVPacket *next_pkt = &pktl->pkt;
if (next_pkt->dts != AV_NOPTS_VALUE) {
int wrap_bits = s->streams[next_pkt->stream_index]->pts_wrap_bits;
// last dts seen for this stream. if any of packets following
// current one had no dts, we will set this to AV_NOPTS_VALUE.
int64_t last_dts = next_pkt->dts;
av_assert2(wrap_bits <= 64);
while (pktl && next_pkt->pts == AV_NOPTS_VALUE) {
if (pktl->pkt.stream_index == next_pkt->stream_index &&
av_compare_mod(next_pkt->dts, pktl->pkt.dts, 2ULL << (wrap_bits - 1)) < 0) {
if (av_compare_mod(pktl->pkt.pts, pktl->pkt.dts, 2ULL << (wrap_bits - 1))) {
// not B-frame
next_pkt->pts = pktl->pkt.dts;
}
if (last_dts != AV_NOPTS_VALUE) {
// Once last dts was set to AV_NOPTS_VALUE, we don't change it.
last_dts = pktl->pkt.dts;
}
}
pktl = pktl->next;
}
if (eof && next_pkt->pts == AV_NOPTS_VALUE && last_dts != AV_NOPTS_VALUE) {
// Fixing the last reference frame had none pts issue (For MXF etc).
// We only do this when
// 1. eof.
// 2. we are not able to resolve a pts value for current packet.
// 3. the packets for this stream at the end of the files had valid dts.
next_pkt->pts = last_dts + next_pkt->duration;
}
pktl = si->packet_buffer.head;
}
/* read packet from packet buffer, if there is data */
st = s->streams[next_pkt->stream_index];
if (!(next_pkt->pts == AV_NOPTS_VALUE && st->discard < AVDISCARD_ALL &&
next_pkt->dts != AV_NOPTS_VALUE && !eof)) {
ret = avpriv_packet_list_get(&si->packet_buffer, pkt);
goto return_packet;
}
}
ret = read_frame_internal(s, pkt);
if (ret < 0) {
if (pktl && ret != AVERROR(EAGAIN)) {
eof = 1;
continue;
} else
return ret;
}
ret = avpriv_packet_list_put(&si->packet_buffer,
pkt, NULL, 0);
if (ret < 0) {
av_packet_unref(pkt);
return ret;
}
}
return_packet:
st = s->streams[pkt->stream_index];
if ((s->iformat->flags & AVFMT_GENERIC_INDEX) && pkt->flags & AV_PKT_FLAG_KEY) {
ff_reduce_index(s, st->index);
av_add_index_entry(st, pkt->pos, pkt->dts, 0, 0, AVINDEX_KEYFRAME);
}
if (is_relative(pkt->dts))
pkt->dts -= RELATIVE_TS_BASE;
if (is_relative(pkt->pts))
pkt->pts -= RELATIVE_TS_BASE;
return ret;
}read_frame_internal
- read_frame_internal()代码比较长，这里只简单看一下它前面的部分。它前面部分有2步是十分关键的：
- （1）调用了ff_read_packet()从相应的AVInputFormat读取数据。
- （2）如果媒体频流需要使用AVCodecParser，则调用parse_packet()解析相应的AVPacket。
- 下面我们分成分别看一下ff_read_packet()和parse_packet()的源代码。
static int read_frame_internal(AVFormatContext *s, AVPacket *pkt)
{
FFFormatContext *const si = ffformatcontext(s);
int ret, got_packet = 0;
AVDictionary *metadata = NULL;
while (!got_packet && !si->parse_queue.head) {
AVStream *st;
FFStream *sti;
/* read next packet */
ret = ff_read_packet(s, pkt);
if (ret < 0) {
if (ret == AVERROR(EAGAIN))
return ret;
/* flush the parsers */
for (unsigned i = 0; i < s->nb_streams; i++) {
AVStream *const st = s->streams[i];
FFStream *const sti = ffstream(st);
if (sti->parser && sti->need_parsing)
parse_packet(s, pkt, st->index, 1);
}
/* all remaining packets are now in parse_queue =>
* really terminate parsing */
break;
}
ret = 0;
st = s->streams[pkt->stream_index];
sti = ffstream(st);
st->event_flags |= AVSTREAM_EVENT_FLAG_NEW_PACKETS;
/* update context if required */
if (sti->need_context_update) {
if (avcodec_is_open(sti->avctx)) {
av_log(s, AV_LOG_DEBUG, "Demuxer context update while decoder is open, closing and trying to re-open\n");
avcodec_close(sti->avctx);
sti->info->found_decoder = 0;
}
/* close parser, because it depends on the codec */
if (sti->parser && sti->avctx->codec_id != st->codecpar->codec_id) {
av_parser_close(sti->parser);
sti->parser = NULL;
}
ret = avcodec_parameters_to_context(sti->avctx, st->codecpar);
if (ret < 0) {
av_packet_unref(pkt);
return ret;
}
sti->need_context_update = 0;
}
if (pkt->pts != AV_NOPTS_VALUE &&
pkt->dts != AV_NOPTS_VALUE &&
pkt->pts < pkt->dts) {
av_log(s, AV_LOG_WARNING,
"Invalid timestamps stream=%d, pts=%s, dts=%s, size=%d\n",
pkt->stream_index,
av_ts2str(pkt->pts),
av_ts2str(pkt->dts),
pkt->size);
}
if (s->debug & FF_FDEBUG_TS)
av_log(s, AV_LOG_DEBUG,
"ff_read_packet stream=%d, pts=%s, dts=%s, size=%d, duration=%"PRId64", flags=%d\n",
pkt->stream_index,
av_ts2str(pkt->pts),
av_ts2str(pkt->dts),
pkt->size, pkt->duration, pkt->flags);
if (sti->need_parsing && !sti->parser && !(s->flags & AVFMT_FLAG_NOPARSE)) {
sti->parser = av_parser_init(st->codecpar->codec_id);
if (!sti->parser) {
av_log(s, AV_LOG_VERBOSE, "parser not found for codec "
"%s, packets or times may be invalid.\n",
avcodec_get_name(st->codecpar->codec_id));
/* no parser available: just output the raw packets */
sti->need_parsing = AVSTREAM_PARSE_NONE;
} else if (sti->need_parsing == AVSTREAM_PARSE_HEADERS)
sti->parser->flags |= PARSER_FLAG_COMPLETE_FRAMES;
else if (sti->need_parsing == AVSTREAM_PARSE_FULL_ONCE)
sti->parser->flags |= PARSER_FLAG_ONCE;
else if (sti->need_parsing == AVSTREAM_PARSE_FULL_RAW)
sti->parser->flags |= PARSER_FLAG_USE_CODEC_TS;
}
if (!sti->need_parsing || !sti->parser) {
/* no parsing needed: we just output the packet as is */
compute_pkt_fields(s, st, NULL, pkt, AV_NOPTS_VALUE, AV_NOPTS_VALUE);
if ((s->iformat->flags & AVFMT_GENERIC_INDEX) &&
(pkt->flags & AV_PKT_FLAG_KEY) && pkt->dts != AV_NOPTS_VALUE) {
ff_reduce_index(s, st->index);
av_add_index_entry(st, pkt->pos, pkt->dts,
0, 0, AVINDEX_KEYFRAME);
}
got_packet = 1;
} else if (st->discard < AVDISCARD_ALL) {
if ((ret = parse_packet(s, pkt, pkt->stream_index, 0)) < 0)
return ret;
st->codecpar->sample_rate = sti->avctx->sample_rate;
st->codecpar->bit_rate = sti->avctx->bit_rate;
#if FF_API_OLD_CHANNEL_LAYOUT
FF_DISABLE_DEPRECATION_WARNINGS
st->codecpar->channels = sti->avctx->ch_layout.nb_channels;
st->codecpar->channel_layout = sti->avctx->ch_layout.order == AV_CHANNEL_ORDER_NATIVE ?
sti->avctx->ch_layout.u.mask : 0;
FF_ENABLE_DEPRECATION_WARNINGS
#endif
ret = av_channel_layout_copy(&st->codecpar->ch_layout, &sti->avctx->ch_layout);
if (ret < 0)
return ret;
st->codecpar->codec_id = sti->avctx->codec_id;
} else {
/* free packet */
av_packet_unref(pkt);
}
if (pkt->flags & AV_PKT_FLAG_KEY)
sti->skip_to_keyframe = 0;
if (sti->skip_to_keyframe) {
av_packet_unref(pkt);
got_packet = 0;
}
}
if (!got_packet && si->parse_queue.head)
ret = avpriv_packet_list_get(&si->parse_queue, pkt);
if (ret >= 0) {
AVStream *const st = s->streams[pkt->stream_index];
FFStream *const sti = ffstream(st);
int discard_padding = 0;
if (sti->first_discard_sample && pkt->pts != AV_NOPTS_VALUE) {
int64_t pts = pkt->pts - (is_relative(pkt->pts) ? RELATIVE_TS_BASE : 0);
int64_t sample = ts_to_samples(st, pts);
int64_t duration = ts_to_samples(st, pkt->duration);
int64_t end_sample = sample + duration;
if (duration > 0 && end_sample >= sti->first_discard_sample &&
sample < sti->last_discard_sample)
discard_padding = FFMIN(end_sample - sti->first_discard_sample, duration);
}
if (sti->start_skip_samples && (pkt->pts == 0 || pkt->pts == RELATIVE_TS_BASE))
sti->skip_samples = sti->start_skip_samples;
sti->skip_samples = FFMAX(0, sti->skip_samples);
if (sti->skip_samples || discard_padding) {
uint8_t *p = av_packet_new_side_data(pkt, AV_PKT_DATA_SKIP_SAMPLES, 10);
if (p) {
AV_WL32(p, sti->skip_samples);
AV_WL32(p + 4, discard_padding);
av_log(s, AV_LOG_DEBUG, "demuxer injecting skip %u / discard %u\n",
(unsigned)sti->skip_samples, (unsigned)discard_padding);
}
sti->skip_samples = 0;
}
if (sti->inject_global_side_data) {
for (int i = 0; i < st->nb_side_data; i++) {
const AVPacketSideData *const src_sd = &st->side_data[i];
uint8_t *dst_data;
if (av_packet_get_side_data(pkt, src_sd->type, NULL))
continue;
dst_data = av_packet_new_side_data(pkt, src_sd->type, src_sd->size);
if (!dst_data) {
av_log(s, AV_LOG_WARNING, "Could not inject global side data\n");
continue;
}
memcpy(dst_data, src_sd->data, src_sd->size);
}
sti->inject_global_side_data = 0;
}
}
if (!si->metafree) {
int metaret = av_opt_get_dict_val(s, "metadata", AV_OPT_SEARCH_CHILDREN, &metadata);
if (metadata) {
s->event_flags |= AVFMT_EVENT_FLAG_METADATA_UPDATED;
av_dict_copy(&s->metadata, metadata, 0);
av_dict_free(&metadata);
av_opt_set_dict_val(s, "metadata", NULL, AV_OPT_SEARCH_CHILDREN);
}
si->metafree = metaret == AVERROR_OPTION_NOT_FOUND;
}
if (s->debug & FF_FDEBUG_TS)
av_log(s, AV_LOG_DEBUG,
"read_frame_internal stream=%d, pts=%s, dts=%s, "
"size=%d, duration=%"PRId64", flags=%d\n",
pkt->stream_index,
av_ts2str(pkt->pts),
av_ts2str(pkt->dts),
pkt->size, pkt->duration, pkt->flags);
/* A demuxer might have returned EOF because of an IO error, let's
* propagate this back to the user. */
if (ret == AVERROR_EOF && s->pb && s->pb->error < 0 && s->pb->error != AVERROR(EAGAIN))
ret = s->pb->error;
return ret;
}
ff_read_packet()
- ff_read_packet()的代码比较长，如下所示。
- ff_read_packet()中最关键的地方就是调用了AVInputFormat的read_packet()方法。
int ff_read_packet(AVFormatContext *s, AVPacket *pkt)
{
FFFormatContext *const si = ffformatcontext(s);
int err;
#if FF_API_INIT_PACKET
FF_DISABLE_DEPRECATION_WARNINGS
pkt->data = NULL;
pkt->size = 0;
av_init_packet(pkt);
FF_ENABLE_DEPRECATION_WARNINGS
#else
av_packet_unref(pkt);
#endif
for (;;) {
PacketListEntry *pktl = si->raw_packet_buffer.head;
AVStream *st;
FFStream *sti;
const AVPacket *pkt1;
if (pktl) {
AVStream *const st = s->streams[pktl->pkt.stream_index];
if (si->raw_packet_buffer_size >= s->probesize)
if ((err = probe_codec(s, st, NULL)) < 0)
return err;
if (ffstream(st)->request_probe <= 0) {
avpriv_packet_list_get(&si->raw_packet_buffer, pkt);
si->raw_packet_buffer_size -= pkt->size;
return 0;
}
}
err = s->iformat->read_packet(s, pkt);
if (err < 0) {
av_packet_unref(pkt);
/* Some demuxers return FFERROR_REDO when they consume
data and discard it (ignored streams, junk, extradata).
We must re-call the demuxer to get the real packet. */
if (err == FFERROR_REDO)
continue;
if (!pktl || err == AVERROR(EAGAIN))
return err;
for (unsigned i = 0; i < s->nb_streams; i++) {
AVStream *const st = s->streams[i];
FFStream *const sti = ffstream(st);
if (sti->probe_packets || sti->request_probe > 0)
if ((err = probe_codec(s, st, NULL)) < 0)
return err;
av_assert0(sti->request_probe <= 0);
}
continue;
}
err = av_packet_make_refcounted(pkt);
if (err < 0) {
av_packet_unref(pkt);
return err;
}
if (pkt->flags & AV_PKT_FLAG_CORRUPT) {
av_log(s, AV_LOG_WARNING,
"Packet corrupt (stream = %d, dts = %s)",
pkt->stream_index, av_ts2str(pkt->dts));
if (s->flags & AVFMT_FLAG_DISCARD_CORRUPT) {
av_log(s, AV_LOG_WARNING, ", dropping it.\n");
av_packet_unref(pkt);
continue;
}
av_log(s, AV_LOG_WARNING, ".\n");
}
av_assert0(pkt->stream_index < (unsigned)s->nb_streams &&
"Invalid stream index.\n");
st = s->streams[pkt->stream_index];
sti = ffstream(st);
if (update_wrap_reference(s, st, pkt->stream_index, pkt) && sti->pts_wrap_behavior == AV_PTS_WRAP_SUB_OFFSET) {
// correct first time stamps to negative values
if (!is_relative(sti->first_dts))
sti->first_dts = wrap_timestamp(st, sti->first_dts);
if (!is_relative(st->start_time))
st->start_time = wrap_timestamp(st, st->start_time);
if (!is_relative(sti->cur_dts))
sti->cur_dts = wrap_timestamp(st, sti->cur_dts);
}
pkt->dts = wrap_timestamp(st, pkt->dts);
pkt->pts = wrap_timestamp(st, pkt->pts);
force_codec_ids(s, st);
/* TODO: audio: time filter; video: frame reordering (pts != dts) */
if (s->use_wallclock_as_timestamps)
pkt->dts = pkt->pts = av_rescale_q(av_gettime(), AV_TIME_BASE_Q, st->time_base);
if (!pktl && sti->request_probe <= 0)
return 0;
err = avpriv_packet_list_put(&si->raw_packet_buffer,
pkt, NULL, 0);
if (err < 0) {
av_packet_unref(pkt);
return err;
}
pkt1 = &si->raw_packet_buffer.tail->pkt;
si->raw_packet_buffer_size += pkt1->size;
if ((err = probe_codec(s, st, pkt1)) < 0)
return err;
}
}
- AVInputFormat的read_packet()是一个函数指针，指向当前的AVInputFormat的读取数据的函数。
- 在这里我们以FLV封装格式对应的AVInputFormat为例，看看read_packet()的实现函数是什么样子的。
- FLV封装格式对应的AVInputFormat的定义位于libavformat\flvdec.c，如下所示。
const AVInputFormat ff_flv_demuxer = {
.name = "flv",
.long_name = NULL_IF_CONFIG_SMALL("FLV (Flash Video)"),
.priv_data_size = sizeof(FLVContext),
.read_probe = flv_probe,
.read_header = flv_read_header,
.read_packet = flv_read_packet,
.read_seek = flv_read_seek,
.read_close = flv_read_close,
.extensions = "flv",
.priv_class = &flv_kux_class,
};
- 从ff_flv_demuxer的定义可以看出，read_packet()对应的是flv_read_packet()函数。
- 在看flv_read_packet()函数之前，我们先回顾一下FLV封装格式的结构，如下图所示。
- 从图中可以看出，FLV文件体部分是由一个一个的Tag连接起来的（中间间隔着Previous Tag Size）。
- 每个Tag包含了Tag Header和Tag Data两个部分。
- Tag Data根据Tag的Type不同而不同：可以分为音频Tag Data，视频Tag Data以及Script Tag Data。
- 下面简述一下音频Tag Data和视频Tag Data。
Audio Tag Data
- Audio Tag在官方标准中定义如下
- Audio Tag开始的第1个字节包含了音频数据的参数信息，从第2个字节开始为音频流数据。
- 第1个字节的前4位的数值表示了音频数据格式：
- 0 = Linear PCM, platform endian
- 1 = ADPCM
- 2 = MP3
- 3 = Linear PCM, little endian
- 4 = Nellymoser 16-kHz mono
- 5 = Nellymoser 8-kHz mono
- 6 = Nellymoser
- 7 = G.711 A-law logarithmic PCM
- 8 = G.711 mu-law logarithmic PCM
- 9 = reserved
- 10 = AAC
- 14 = MP3 8-Khz
- 15 = Device-specific sound
- 第1个字节的第5-6位的数值表示采样率：0 = 5.5kHz，1 = 11KHz，2 = 22 kHz，3 = 44 kHz。
- 第1个字节的第7位表示采样精度：0 = 8bits，1 = 16bits。
- 第1个字节的第8位表示音频类型：0 = sndMono，1 = sndStereo。
- 其中，当音频编码为AAC的时候，第一个字节后面存储的是AACAUDIODATA，格式如下所示。
Video Tag Data
- Video Tag在官方标准中的定义如下。
- Video Tag也用开始的第1个字节包含视频数据的参数信息，从第2个字节为视频流数据。 第1个字节的前4位的数值表示帧类型（FrameType）：
- 1: keyframe (for AVC, a seekableframe)（关键帧）
- 2: inter frame (for AVC, a nonseekableframe)
- 3: disposable inter frame (H.263only)
- 4: generated keyframe (reservedfor server use only)
- 5: video info/command frame
- 第1个字节的后4位的数值表示视频编码ID（CodecID）：
- 1: JPEG (currently unused)
- 2: Sorenson H.263
- 3: Screen video
- 4: On2 VP6
- 5: On2 VP6 with alpha channel
- 6: Screen video version 2
- 7: AVC 其中，当音频编码为AVC（H.264）的时候，第一个字节后面存储的是AVCVIDEOPACKET，格式如下所示。
flv_read_packet()
- flv_read_packet()的定义位于libavformat\flvdec.c，如下所示。
static int flv_read_packet(AVFormatContext *s, AVPacket *pkt)
{
FLVContext *flv = s->priv_data;
int ret, i, size, flags;
enum FlvTagType type;
int stream_type=-1;
int64_t next, pos, meta_pos;
int64_t dts, pts = AV_NOPTS_VALUE;
int av_uninit(channels);
int av_uninit(sample_rate);
AVStream *st = NULL;
int last = -1;
int orig_size;
retry:
/* pkt size is repeated at end. skip it */
pos = avio_tell(s->pb);
type = (avio_r8(s->pb) & 0x1F);
orig_size =
size = avio_rb24(s->pb);
flv->sum_flv_tag_size += size + 11;
dts = avio_rb24(s->pb);
dts |= (unsigned)avio_r8(s->pb) << 24;
av_log(s, AV_LOG_TRACE, "type:%d, size:%d, last:%d, dts:%"PRId64" pos:%"PRId64"\n", type, size, last, dts, avio_tell(s->pb));
if (avio_feof(s->pb))
return AVERROR_EOF;
avio_skip(s->pb, 3); /* stream id, always 0 */
flags = 0;
if (flv->validate_next < flv->validate_count) {
int64_t validate_pos = flv->validate_index[flv->validate_next].pos;
if (pos == validate_pos) {
if (FFABS(dts - flv->validate_index[flv->validate_next].dts) <=
VALIDATE_INDEX_TS_THRESH) {
flv->validate_next++;
} else {
clear_index_entries(s, validate_pos);
flv->validate_count = 0;
}
} else if (pos > validate_pos) {
clear_index_entries(s, validate_pos);
flv->validate_count = 0;
}
}
if (size == 0) {
ret = FFERROR_REDO;
goto leave;
}
next = size + avio_tell(s->pb);
if (type == FLV_TAG_TYPE_AUDIO) {
stream_type = FLV_STREAM_TYPE_AUDIO;
flags = avio_r8(s->pb);
size--;
} else if (type == FLV_TAG_TYPE_VIDEO) {
stream_type = FLV_STREAM_TYPE_VIDEO;
flags = avio_r8(s->pb);
size--;
if ((flags & FLV_VIDEO_FRAMETYPE_MASK) == FLV_FRAME_VIDEO_INFO_CMD)
goto skip;
} else if (type == FLV_TAG_TYPE_META) {
stream_type=FLV_STREAM_TYPE_SUBTITLE;
if (size > 13 + 1 + 4) { // Header-type metadata stuff
int type;
meta_pos = avio_tell(s->pb);
type = flv_read_metabody(s, next);
if (type == 0 && dts == 0 || type < 0) {
if (type < 0 && flv->validate_count &&
flv->validate_index[0].pos > next &&
flv->validate_index[0].pos - 4 < next) {
av_log(s, AV_LOG_WARNING, "Adjusting next position due to index mismatch\n");
next = flv->validate_index[0].pos - 4;
}
goto skip;
} else if (type == TYPE_ONTEXTDATA) {
avpriv_request_sample(s, "OnTextData packet");
return flv_data_packet(s, pkt, dts, next);
} else if (type == TYPE_ONCAPTION) {
return flv_data_packet(s, pkt, dts, next);
} else if (type == TYPE_UNKNOWN) {
stream_type = FLV_STREAM_TYPE_DATA;
}
avio_seek(s->pb, meta_pos, SEEK_SET);
}
} else {
av_log(s, AV_LOG_DEBUG,
"Skipping flv packet: type %d, size %d, flags %d.\n",
type, size, flags);
skip:
if (avio_seek(s->pb, next, SEEK_SET) != next) {
// This can happen if flv_read_metabody above read past
// next, on a non-seekable input, and the preceding data has
// been flushed out from the IO buffer.
av_log(s, AV_LOG_ERROR, "Unable to seek to the next packet\n");
return AVERROR_INVALIDDATA;
}
ret = FFERROR_REDO;
goto leave;
}
/* skip empty data packets */
if (!size) {
ret = FFERROR_REDO;
goto leave;
}
/* now find stream */
for (i = 0; i < s->nb_streams; i++) {
st = s->streams[i];
if (stream_type == FLV_STREAM_TYPE_AUDIO) {
if (st->codecpar->codec_type == AVMEDIA_TYPE_AUDIO &&
(s->audio_codec_id || flv_same_audio_codec(st->codecpar, flags)))
break;
} else if (stream_type == FLV_STREAM_TYPE_VIDEO) {
if (st->codecpar->codec_type == AVMEDIA_TYPE_VIDEO &&
(s->video_codec_id || flv_same_video_codec(st->codecpar, flags)))
break;
} else if (stream_type == FLV_STREAM_TYPE_SUBTITLE) {
if (st->codecpar->codec_type == AVMEDIA_TYPE_SUBTITLE)
break;
} else if (stream_type == FLV_STREAM_TYPE_DATA) {
if (st->codecpar->codec_type == AVMEDIA_TYPE_DATA)
break;
}
}
if (i == s->nb_streams) {
static const enum AVMediaType stream_types[] = {AVMEDIA_TYPE_VIDEO, AVMEDIA_TYPE_AUDIO, AVMEDIA_TYPE_SUBTITLE, AVMEDIA_TYPE_DATA};
st = create_stream(s, stream_types[stream_type]);
if (!st)
return AVERROR(ENOMEM);
}
av_log(s, AV_LOG_TRACE, "%d %X %d \n", stream_type, flags, st->discard);
if (flv->time_pos <= pos) {
dts += flv->time_offset;
}
if ((s->pb->seekable & AVIO_SEEKABLE_NORMAL) &&
((flags & FLV_VIDEO_FRAMETYPE_MASK) == FLV_FRAME_KEY ||
stream_type == FLV_STREAM_TYPE_AUDIO))
av_add_index_entry(st, pos, dts, size, 0, AVINDEX_KEYFRAME);
if ((st->discard >= AVDISCARD_NONKEY && !((flags & FLV_VIDEO_FRAMETYPE_MASK) == FLV_FRAME_KEY || stream_type == FLV_STREAM_TYPE_AUDIO)) ||
(st->discard >= AVDISCARD_BIDIR && ((flags & FLV_VIDEO_FRAMETYPE_MASK) == FLV_FRAME_DISP_INTER && stream_type == FLV_STREAM_TYPE_VIDEO)) ||
st->discard >= AVDISCARD_ALL) {
avio_seek(s->pb, next, SEEK_SET);
ret = FFERROR_REDO;
goto leave;
}
// if not streamed and no duration from metadata then seek to end to find
// the duration from the timestamps
if ((s->pb->seekable & AVIO_SEEKABLE_NORMAL) &&
(!s->duration || s->duration == AV_NOPTS_VALUE) &&
!flv->searched_for_end) {
int size;
const int64_t pos = avio_tell(s->pb);
// Read the last 4 bytes of the file, this should be the size of the
// previous FLV tag. Use the timestamp of its payload as duration.
int64_t fsize = avio_size(s->pb);
retry_duration:
avio_seek(s->pb, fsize - 4, SEEK_SET);
size = avio_rb32(s->pb);
if (size > 0 && size < fsize) {
// Seek to the start of the last FLV tag at position (fsize - 4 - size)
// but skip the byte indicating the type.
avio_seek(s->pb, fsize - 3 - size, SEEK_SET);
if (size == avio_rb24(s->pb) + 11) {
uint32_t ts = avio_rb24(s->pb);
ts |= (unsigned)avio_r8(s->pb) << 24;
if (ts)
s->duration = ts * (int64_t)AV_TIME_BASE / 1000;
else if (fsize >= 8 && fsize - 8 >= size) {
fsize -= size+4;
goto retry_duration;
}
}
}
avio_seek(s->pb, pos, SEEK_SET);
flv->searched_for_end = 1;
}
if (stream_type == FLV_STREAM_TYPE_AUDIO) {
int bits_per_coded_sample;
channels = (flags & FLV_AUDIO_CHANNEL_MASK) == FLV_STEREO ? 2 : 1;
sample_rate = 44100 << ((flags & FLV_AUDIO_SAMPLERATE_MASK) >>
FLV_AUDIO_SAMPLERATE_OFFSET) >> 3;
bits_per_coded_sample = (flags & FLV_AUDIO_SAMPLESIZE_MASK) ? 16 : 8;
if (!av_channel_layout_check(&st->codecpar->ch_layout) ||
!st->codecpar->sample_rate ||
!st->codecpar->bits_per_coded_sample) {
av_channel_layout_default(&st->codecpar->ch_layout, channels);
st->codecpar->sample_rate = sample_rate;
st->codecpar->bits_per_coded_sample = bits_per_coded_sample;
}
if (!st->codecpar->codec_id) {
flv_set_audio_codec(s, st, st->codecpar,
flags & FLV_AUDIO_CODECID_MASK);
flv->last_sample_rate =
sample_rate = st->codecpar->sample_rate;
flv->last_channels =
channels = st->codecpar->ch_layout.nb_channels;
} else {
AVCodecParameters *par = avcodec_parameters_alloc();
if (!par) {
ret = AVERROR(ENOMEM);
goto leave;
}
par->sample_rate = sample_rate;
par->bits_per_coded_sample = bits_per_coded_sample;
flv_set_audio_codec(s, st, par, flags & FLV_AUDIO_CODECID_MASK);
sample_rate = par->sample_rate;
avcodec_parameters_free(&par);
}
} else if (stream_type == FLV_STREAM_TYPE_VIDEO) {
int ret = flv_set_video_codec(s, st, flags & FLV_VIDEO_CODECID_MASK, 1);
if (ret < 0)
return ret;
size -= ret;
} else if (stream_type == FLV_STREAM_TYPE_SUBTITLE) {
st->codecpar->codec_id = AV_CODEC_ID_TEXT;
} else if (stream_type == FLV_STREAM_TYPE_DATA) {
st->codecpar->codec_id = AV_CODEC_ID_NONE; // Opaque AMF data
}
if (st->codecpar->codec_id == AV_CODEC_ID_AAC ||
st->codecpar->codec_id == AV_CODEC_ID_H264 ||
st->codecpar->codec_id == AV_CODEC_ID_MPEG4) {
int type = avio_r8(s->pb);
size--;
if (size < 0) {
ret = AVERROR_INVALIDDATA;
goto leave;
}
if (st->codecpar->codec_id == AV_CODEC_ID_H264 || st->codecpar->codec_id == AV_CODEC_ID_MPEG4) {
// sign extension
int32_t cts = (avio_rb24(s->pb) + 0xff800000) ^ 0xff800000;
pts = av_sat_add64(dts, cts);
if (cts < 0) { // dts might be wrong
if (!flv->wrong_dts)
av_log(s, AV_LOG_WARNING,
"Negative cts, previous timestamps might be wrong.\n");
flv->wrong_dts = 1;
} else if (FFABS(dts - pts) > 1000*60*15) {
av_log(s, AV_LOG_WARNING,
"invalid timestamps %"PRId64" %"PRId64"\n", dts, pts);
dts = pts = AV_NOPTS_VALUE;
}
}
if (type == 0 && (!st->codecpar->extradata || st->codecpar->codec_id == AV_CODEC_ID_AAC ||
st->codecpar->codec_id == AV_CODEC_ID_H264)) {
AVDictionaryEntry *t;
if (st->codecpar->extradata) {
if ((ret = flv_queue_extradata(flv, s->pb, stream_type, size)) < 0)
return ret;
ret = FFERROR_REDO;
goto leave;
}
if ((ret = flv_get_extradata(s, st, size)) < 0)
return ret;
/* Workaround for buggy Omnia A/XE encoder */
t = av_dict_get(s->metadata, "Encoder", NULL, 0);
if (st->codecpar->codec_id == AV_CODEC_ID_AAC && t && !strcmp(t->value, "Omnia A/XE"))
st->codecpar->extradata_size = 2;
ret = FFERROR_REDO;
goto leave;
}
}
/* skip empty data packets */
if (!size) {
ret = FFERROR_REDO;
goto leave;
}
ret = av_get_packet(s->pb, pkt, size);
if (ret < 0)
return ret;
pkt->dts = dts;
pkt->pts = pts == AV_NOPTS_VALUE ? dts : pts;
pkt->stream_index = st->index;
pkt->pos = pos;
if (flv->new_extradata[stream_type]) {
int ret = av_packet_add_side_data(pkt, AV_PKT_DATA_NEW_EXTRADATA,
flv->new_extradata[stream_type],
flv->new_extradata_size[stream_type]);
if (ret >= 0) {
flv->new_extradata[stream_type] = NULL;
flv->new_extradata_size[stream_type] = 0;
}
}
if (stream_type == FLV_STREAM_TYPE_AUDIO &&
(sample_rate != flv->last_sample_rate ||
channels != flv->last_channels)) {
flv->last_sample_rate = sample_rate;
flv->last_channels = channels;
ff_add_param_change(pkt, channels, 0, sample_rate, 0, 0);
}
if (stream_type == FLV_STREAM_TYPE_AUDIO ||
(flags & FLV_VIDEO_FRAMETYPE_MASK) == FLV_FRAME_KEY ||
stream_type == FLV_STREAM_TYPE_SUBTITLE ||
stream_type == FLV_STREAM_TYPE_DATA)
pkt->flags |= AV_PKT_FLAG_KEY;
leave:
last = avio_rb32(s->pb);
if (!flv->trust_datasize) {
if (last != orig_size + 11 && last != orig_size + 10 &&
!avio_feof(s->pb) &&
(last != orig_size || !last) && last != flv->sum_flv_tag_size &&
!flv->broken_sizes) {
av_log(s, AV_LOG_ERROR, "Packet mismatch %d %d %d\n", last, orig_size + 11, flv->sum_flv_tag_size);
avio_seek(s->pb, pos + 1, SEEK_SET);
ret = resync(s);
av_packet_unref(pkt);
if (ret >= 0) {
goto retry;
}
}
}
if (ret >= 0)
flv->last_ts = pkt->dts;
return ret;
}
- flv_read_packet()的代码比较长，但是逻辑比较简单。它的主要功能就是根据FLV文件格式的规范，逐层解析Tag以及TagData，获取Tag以及TagData中的信息。比较关键的地方已经写上了注释，不再详细叙述。
parse_packet()
- parse_packet()给需要AVCodecParser的媒体流提供解析AVPacket的功能。它的代码如下所示：
- 从代码中可以看出，最终调用了相应AVCodecParser的av_parser_parse2()函数，解析出来AVPacket。此后根据解析的信息还进行了一系列的赋值工作，不再详细叙述。
/**
* Parse a packet, add all split parts to parse_queue.
*
* @param pkt Packet to parse; must not be NULL.
* @param flush Indicates whether to flush. If set, pkt must be blank.
*/
static int parse_packet(AVFormatContext *s, AVPacket *pkt,
int stream_index, int flush)
{
FFFormatContext *const si = ffformatcontext(s);
AVPacket *out_pkt = si->parse_pkt;
AVStream *st = s->streams[stream_index];
FFStream *const sti = ffstream(st);
const uint8_t *data = pkt->data;
int size = pkt->size;
int ret = 0, got_output = flush;
if (!size && !flush && sti->parser->flags & PARSER_FLAG_COMPLETE_FRAMES) {
// preserve 0-size sync packets
compute_pkt_fields(s, st, sti->parser, pkt, AV_NOPTS_VALUE, AV_NOPTS_VALUE);
}
while (size > 0 || (flush && got_output)) {
int64_t next_pts = pkt->pts;
int64_t next_dts = pkt->dts;
int len;
len = av_parser_parse2(sti->parser, sti->avctx,
&out_pkt->data, &out_pkt->size, data, size,
pkt->pts, pkt->dts, pkt->pos);
pkt->pts = pkt->dts = AV_NOPTS_VALUE;
pkt->pos = -1;
/* increment read pointer */
av_assert1(data || !len);
data = len ? data + len : data;
size -= len;
got_output = !!out_pkt->size;
if (!out_pkt->size)
continue;
if (pkt->buf && out_pkt->data == pkt->data) {
/* reference pkt->buf only when out_pkt->data is guaranteed to point
* to data in it and not in the parser's internal buffer. */
/* XXX: Ensure this is the case with all parsers when sti->parser->flags
* is PARSER_FLAG_COMPLETE_FRAMES and check for that instead? */
out_pkt->buf = av_buffer_ref(pkt->buf);
if (!out_pkt->buf) {
ret = AVERROR(ENOMEM);
goto fail;
}
} else {
ret = av_packet_make_refcounted(out_pkt);
if (ret < 0)
goto fail;
}
if (pkt->side_data) {
out_pkt->side_data = pkt->side_data;
out_pkt->side_data_elems = pkt->side_data_elems;
pkt->side_data = NULL;
pkt->side_data_elems = 0;
}
/* set the duration */
out_pkt->duration = (sti->parser->flags & PARSER_FLAG_COMPLETE_FRAMES) ? pkt->duration : 0;
if (st->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
if (sti->avctx->sample_rate > 0) {
out_pkt->duration =
av_rescale_q_rnd(sti->parser->duration,
(AVRational) { 1, sti->avctx->sample_rate },
st->time_base,
AV_ROUND_DOWN);
}
}
out_pkt->stream_index = st->index;
out_pkt->pts = sti->parser->pts;
out_pkt->dts = sti->parser->dts;
out_pkt->pos = sti->parser->pos;
out_pkt->flags |= pkt->flags & (AV_PKT_FLAG_DISCARD | AV_PKT_FLAG_CORRUPT);
if (sti->need_parsing == AVSTREAM_PARSE_FULL_RAW)
out_pkt->pos = sti->parser->frame_offset;
if (sti->parser->key_frame == 1 ||
(sti->parser->key_frame == -1 &&
sti->parser->pict_type == AV_PICTURE_TYPE_I))
out_pkt->flags |= AV_PKT_FLAG_KEY;
if (sti->parser->key_frame == -1 && sti->parser->pict_type ==AV_PICTURE_TYPE_NONE && (pkt->flags&AV_PKT_FLAG_KEY))
out_pkt->flags |= AV_PKT_FLAG_KEY;
compute_pkt_fields(s, st, sti->parser, out_pkt, next_dts, next_pts);
ret = avpriv_packet_list_put(&si->parse_queue,
out_pkt, NULL, 0);
if (ret < 0)
goto fail;
}
/* end of the stream => close and free the parser */
if (flush) {
av_parser_close(sti->parser);
sti->parser = NULL;
}
fail:
if (ret < 0)
av_packet_unref(out_pkt);
av_packet_unref(pkt);
return ret;
}
扫一扫
998
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
