This repository has been archived on 2023-10-09. You can view files and clone it. You cannot open issues or pull requests or push a commit.
Files
blender-archive/extern/audaspace/plugins/ffmpeg/FFMPEGReader.cpp
Sebastian Parborg ded68fb102 VSE: Fix audaspace not reading ffmpeg files with start offset correctly
The duration and start time for audio strips were not correctly read in
audaspace.

Some video files have a "lead in" section of audio that plays before the
video starts playing back. Before this patch, we would play this lead in
audio at the same time as the video started and thus the audio would not
be in sync anymore.

Now the lead in audio is cut off and the duration should be correctly
calculated with this in mind.

If the audio starts after the video, the audio strip is shifted to
account for this, but it will also lead to cut off audio which might not
be wanted. However we don't have a simple way to solve this at this
point.

Differential Revision: http://developer.blender.org/D11917
2021-08-16 15:10:58 +02:00

509 lines
13 KiB
C++

/*******************************************************************************
* Copyright 2009-2016 Jörg Müller
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
******************************************************************************/
#include "FFMPEGReader.h"
#include "Exception.h"
#include <algorithm>
extern "C" {
#include <libavcodec/avcodec.h>
#include <libavformat/avio.h>
#include <libavutil/avutil.h>
}
AUD_NAMESPACE_BEGIN
#if LIBAVCODEC_VERSION_MAJOR < 58
#define FFMPEG_OLD_CODE
#endif
int FFMPEGReader::decode(AVPacket& packet, Buffer& buffer)
{
int buf_size = buffer.getSize();
int buf_pos = 0;
#ifdef FFMPEG_OLD_CODE
int got_frame;
int read_length;
uint8_t* orig_data = packet.data;
int orig_size = packet.size;
while(packet.size > 0)
{
got_frame = 0;
read_length = avcodec_decode_audio4(m_codecCtx, m_frame, &got_frame, &packet);
if(read_length < 0)
break;
if(got_frame)
{
int data_size = av_samples_get_buffer_size(nullptr, m_codecCtx->channels, m_frame->nb_samples, m_codecCtx->sample_fmt, 1);
if(buf_size - buf_pos < data_size)
{
buffer.resize(buf_size + data_size, true);
buf_size += data_size;
}
if(m_tointerleave)
{
int single_size = data_size / m_codecCtx->channels / m_frame->nb_samples;
for(int channel = 0; channel < m_codecCtx->channels; channel++)
{
for(int i = 0; i < m_frame->nb_samples; i++)
{
std::memcpy(((data_t*)buffer.getBuffer()) + buf_pos + ((m_codecCtx->channels * i) + channel) * single_size,
m_frame->data[channel] + i * single_size, single_size);
}
}
}
else
std::memcpy(((data_t*)buffer.getBuffer()) + buf_pos, m_frame->data[0], data_size);
buf_pos += data_size;
}
packet.size -= read_length;
packet.data += read_length;
}
packet.data = orig_data;
packet.size = orig_size;
#else
avcodec_send_packet(m_codecCtx, &packet);
while(true)
{
auto ret = avcodec_receive_frame(m_codecCtx, m_frame);
if(ret != 0)
break;
int data_size = av_samples_get_buffer_size(nullptr, m_codecCtx->channels, m_frame->nb_samples, m_codecCtx->sample_fmt, 1);
if(buf_size - buf_pos < data_size)
{
buffer.resize(buf_size + data_size, true);
buf_size += data_size;
}
if(m_tointerleave)
{
int single_size = data_size / m_codecCtx->channels / m_frame->nb_samples;
for(int channel = 0; channel < m_codecCtx->channels; channel++)
{
for(int i = 0; i < m_frame->nb_samples; i++)
{
std::memcpy(((data_t*)buffer.getBuffer()) + buf_pos + ((m_codecCtx->channels * i) + channel) * single_size,
m_frame->data[channel] + i * single_size, single_size);
}
}
}
else
std::memcpy(((data_t*)buffer.getBuffer()) + buf_pos, m_frame->data[0], data_size);
buf_pos += data_size;
}
#endif
return buf_pos;
}
void FFMPEGReader::init()
{
m_position = 0;
m_start_offset = 0.0f;
m_pkgbuf_left = 0;
m_st_time = 0;
m_duration = 0;
if(avformat_find_stream_info(m_formatCtx, nullptr) < 0)
AUD_THROW(FileException, "File couldn't be read, ffmpeg couldn't find the stream info.");
// find audio stream and codec
m_stream = -1;
double dur_sec = 0;
for(unsigned int i = 0; i < m_formatCtx->nb_streams; i++)
{
#ifdef FFMPEG_OLD_CODE
if(m_formatCtx->streams[i]->codec->codec_type == AVMEDIA_TYPE_AUDIO)
#else
if(m_formatCtx->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_AUDIO)
#endif
{
AVStream *audio_stream = m_formatCtx->streams[i];
double audio_timebase = av_q2d(audio_stream->time_base);
if (audio_stream->start_time != AV_NOPTS_VALUE)
{
m_st_time = audio_stream->start_time;
}
int64_t ctx_start_time = 0;
if (m_formatCtx->start_time != AV_NOPTS_VALUE) {
ctx_start_time = m_formatCtx->start_time;
}
m_start_offset = m_st_time * audio_timebase - (double)ctx_start_time / AV_TIME_BASE;
if(audio_stream->duration != AV_NOPTS_VALUE)
{
dur_sec = audio_stream->duration * audio_timebase;
}
else
{
/* If the audio starts after the stream start time, subract this from the total duration. */
dur_sec = (double)m_formatCtx->duration / AV_TIME_BASE - m_start_offset;
}
m_stream=i;
break;
}
}
if(m_stream == -1)
AUD_THROW(FileException, "File couldn't be read, no audio stream found by ffmpeg.");
// get a decoder and open it
#ifndef FFMPEG_OLD_CODE
AVCodec* aCodec = avcodec_find_decoder(m_formatCtx->streams[m_stream]->codecpar->codec_id);
if(!aCodec)
AUD_THROW(FileException, "File couldn't be read, no decoder found with ffmpeg.");
#endif
m_frame = av_frame_alloc();
if(!m_frame)
AUD_THROW(FileException, "File couldn't be read, ffmpeg frame couldn't be allocated.");
#ifdef FFMPEG_OLD_CODE
m_codecCtx = m_formatCtx->streams[m_stream]->codec;
AVCodec* aCodec = avcodec_find_decoder(m_codecCtx->codec_id);
#else
m_codecCtx = avcodec_alloc_context3(aCodec);
#endif
if(!m_codecCtx)
AUD_THROW(FileException, "File couldn't be read, ffmpeg context couldn't be allocated.");
#ifndef FFMPEG_OLD_CODE
if(avcodec_parameters_to_context(m_codecCtx, m_formatCtx->streams[m_stream]->codecpar) < 0)
AUD_THROW(FileException, "File couldn't be read, ffmpeg decoder parameters couldn't be copied to decoder context.");
#endif
if(avcodec_open2(m_codecCtx, aCodec, nullptr) < 0)
AUD_THROW(FileException, "File couldn't be read, ffmpeg codec couldn't be opened.");
m_specs.channels = (Channels) m_codecCtx->channels;
m_tointerleave = av_sample_fmt_is_planar(m_codecCtx->sample_fmt);
switch(av_get_packed_sample_fmt(m_codecCtx->sample_fmt))
{
case AV_SAMPLE_FMT_U8:
m_convert = convert_u8_float;
m_specs.format = FORMAT_U8;
break;
case AV_SAMPLE_FMT_S16:
m_convert = convert_s16_float;
m_specs.format = FORMAT_S16;
break;
case AV_SAMPLE_FMT_S32:
m_convert = convert_s32_float;
m_specs.format = FORMAT_S32;
break;
case AV_SAMPLE_FMT_FLT:
m_convert = convert_copy<float>;
m_specs.format = FORMAT_FLOAT32;
break;
case AV_SAMPLE_FMT_DBL:
m_convert = convert_double_float;
m_specs.format = FORMAT_FLOAT64;
break;
default:
AUD_THROW(FileException, "File couldn't be read, ffmpeg sample format unknown.");
}
m_specs.rate = (SampleRate) m_codecCtx->sample_rate;
m_duration = lround(dur_sec * m_codecCtx->sample_rate);
}
FFMPEGReader::FFMPEGReader(std::string filename) :
m_pkgbuf(),
m_formatCtx(nullptr),
m_codecCtx(nullptr),
m_frame(nullptr),
m_aviocontext(nullptr),
m_membuf(nullptr)
{
// open file
if(avformat_open_input(&m_formatCtx, filename.c_str(), nullptr, nullptr)!=0)
AUD_THROW(FileException, "File couldn't be opened with ffmpeg.");
try
{
init();
}
catch(Exception&)
{
avformat_close_input(&m_formatCtx);
throw;
}
}
FFMPEGReader::FFMPEGReader(std::shared_ptr<Buffer> buffer) :
m_pkgbuf(),
m_codecCtx(nullptr),
m_frame(nullptr),
m_membuffer(buffer),
m_membufferpos(0)
{
m_membuf = reinterpret_cast<data_t*>(av_malloc(AV_INPUT_BUFFER_MIN_SIZE + AV_INPUT_BUFFER_PADDING_SIZE));
m_aviocontext = avio_alloc_context(m_membuf, AV_INPUT_BUFFER_MIN_SIZE, 0, this, read_packet, nullptr, seek_packet);
if(!m_aviocontext)
{
av_free(m_aviocontext);
AUD_THROW(FileException, "Buffer reading context couldn't be created with ffmpeg.");
}
m_formatCtx = avformat_alloc_context();
m_formatCtx->pb = m_aviocontext;
if(avformat_open_input(&m_formatCtx, "", nullptr, nullptr)!=0)
{
av_free(m_aviocontext);
AUD_THROW(FileException, "Buffer couldn't be read with ffmpeg.");
}
try
{
init();
}
catch(Exception&)
{
avformat_close_input(&m_formatCtx);
av_free(m_aviocontext);
throw;
}
}
FFMPEGReader::~FFMPEGReader()
{
if(m_frame)
av_frame_free(&m_frame);
#ifdef FFMPEG_OLD_CODE
avcodec_close(m_codecCtx);
#else
if(m_codecCtx)
avcodec_free_context(&m_codecCtx);
#endif
avformat_close_input(&m_formatCtx);
}
int FFMPEGReader::read_packet(void* opaque, uint8_t* buf, int buf_size)
{
FFMPEGReader* reader = reinterpret_cast<FFMPEGReader*>(opaque);
int size = std::min(buf_size, reader->m_membuffer->getSize() - reader->m_membufferpos);
if(size < 0)
return -1;
std::memcpy(buf, ((data_t*)reader->m_membuffer->getBuffer()) + reader->m_membufferpos, size);
reader->m_membufferpos += size;
return size;
}
int64_t FFMPEGReader::seek_packet(void* opaque, int64_t offset, int whence)
{
FFMPEGReader* reader = reinterpret_cast<FFMPEGReader*>(opaque);
switch(whence)
{
case SEEK_SET:
reader->m_membufferpos = 0;
break;
case SEEK_END:
reader->m_membufferpos = reader->m_membuffer->getSize();
break;
case AVSEEK_SIZE:
return reader->m_membuffer->getSize();
}
int64_t position = reader->m_membufferpos + offset;
if(position > reader->m_membuffer->getSize())
position = reader->m_membuffer->getSize();
reader->m_membufferpos = int(position);
return position;
}
bool FFMPEGReader::isSeekable() const
{
return true;
}
void FFMPEGReader::seek(int position)
{
if(position >= 0)
{
double pts_time_base =
av_q2d(m_formatCtx->streams[m_stream]->time_base);
uint64_t seek_pts = (((uint64_t)position) / ((uint64_t)m_specs.rate)) / pts_time_base;
if(m_st_time != AV_NOPTS_VALUE) {
seek_pts += m_st_time;
}
// a value < 0 tells us that seeking failed
if(av_seek_frame(m_formatCtx, m_stream, seek_pts,
AVSEEK_FLAG_BACKWARD | AVSEEK_FLAG_ANY) >= 0)
{
avcodec_flush_buffers(m_codecCtx);
m_position = position;
AVPacket packet;
bool search = true;
while(search && av_read_frame(m_formatCtx, &packet) >= 0)
{
// is it a frame from the audio stream?
if(packet.stream_index == m_stream)
{
// decode the package
m_pkgbuf_left = decode(packet, m_pkgbuf);
search = false;
// check position
if(packet.pts != AV_NOPTS_VALUE)
{
// calculate real position, and read to frame!
m_position = (packet.pts - m_st_time) * pts_time_base * m_specs.rate;
if(m_position < position)
{
// read until we're at the right position
int length = AUD_DEFAULT_BUFFER_SIZE;
Buffer buffer(length * AUD_SAMPLE_SIZE(m_specs));
bool eos;
for(int len = position - m_position; len > 0; len -= AUD_DEFAULT_BUFFER_SIZE)
{
if(len < AUD_DEFAULT_BUFFER_SIZE)
length = len;
read(length, eos, buffer.getBuffer());
}
}
}
}
av_packet_unref(&packet);
}
}
else
{
fprintf(stderr, "seeking failed!\n");
// Seeking failed, do nothing.
}
}
}
int FFMPEGReader::getLength() const
{
// return approximated remaning size
return m_duration - m_position;
}
int FFMPEGReader::getPosition() const
{
return m_position;
}
double FFMPEGReader::getStartOffset() const
{
return m_start_offset;
}
Specs FFMPEGReader::getSpecs() const
{
return m_specs.specs;
}
void FFMPEGReader::read(int& length, bool& eos, sample_t* buffer)
{
// read packages and decode them
AVPacket packet = {};
int data_size = 0;
int pkgbuf_pos;
int left = length;
int sample_size = AUD_DEVICE_SAMPLE_SIZE(m_specs);
sample_t* buf = buffer;
pkgbuf_pos = m_pkgbuf_left;
m_pkgbuf_left = 0;
// there may still be data in the buffer from the last call
if(pkgbuf_pos > 0)
{
data_size = std::min(pkgbuf_pos, left * sample_size);
m_convert((data_t*) buf, (data_t*) m_pkgbuf.getBuffer(), data_size / AUD_FORMAT_SIZE(m_specs.format));
buf += data_size / AUD_FORMAT_SIZE(m_specs.format);
left -= data_size / sample_size;
}
// for each frame read as long as there isn't enough data already
while((left > 0) && (av_read_frame(m_formatCtx, &packet) >= 0))
{
// is it a frame from the audio stream?
if(packet.stream_index == m_stream)
{
// decode the package
pkgbuf_pos = decode(packet, m_pkgbuf);
if (packet.pts >= m_st_time) {
// copy to output buffer
data_size = std::min(pkgbuf_pos, left * sample_size);
m_convert((data_t*) buf, (data_t*) m_pkgbuf.getBuffer(), data_size / AUD_FORMAT_SIZE(m_specs.format));
buf += data_size / AUD_FORMAT_SIZE(m_specs.format);
left -= data_size / sample_size;
}
}
av_packet_unref(&packet);
}
// read more data than necessary?
if(pkgbuf_pos > data_size)
{
m_pkgbuf_left = pkgbuf_pos-data_size;
memmove(m_pkgbuf.getBuffer(),
((data_t*)m_pkgbuf.getBuffer())+data_size,
pkgbuf_pos-data_size);
}
if((eos = (left > 0)))
length -= left;
m_position += length;
}
AUD_NAMESPACE_END