oak-team/oak-editor
3
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

jkl playback is functional

Browse Source
This commit is contained in:
itsmattkc
2019-01-23 18:57:52 +11:00
parent 0e3250fb59
commit 8462301caf
15 changed files with 162 additions and 63 deletions
Download Patch File Download Diff File Expand all files Collapse all files
playback/cacher.cpp
+52 -25
View File
@@ -86,11 +86,14 @@ void apply_audio_effects(Clip* c, double timecode_start, AVFrame* frame, int nb_
#define AUDIO_BUFFER_PADDING 2048
void cache_audio_worker(Clip* c, bool scrubbing, QVector<Clip*>& nests) {
void cache_audio_worker(Clip* c, bool scrubbing, QVector<Clip*>& nests, int playback_speed) {
long timeline_in = c->get_timeline_in_with_transition();
long timeline_out = c->get_timeline_out_with_transition();
long target_frame = c->audio_target_frame;
bool temp_reverse = (playback_speed < 0);
bool reverse_audio = (c->reverse != temp_reverse);
long frame_skip = 0;
double last_fr = c->sequence->frame_rate;
if (!nests.isEmpty()) {
@@ -113,6 +116,17 @@ void cache_audio_worker(Clip* c, bool scrubbing, QVector<Clip*>& nests) {
}
}
if (temp_reverse) {
long seq_end = sequence->getEndFrame();
timeline_in = seq_end - timeline_in;
timeline_out = seq_end - timeline_out;
target_frame = seq_end - target_frame;
long temp = timeline_in;
timeline_in = timeline_out;
timeline_out = temp;
}
while (true) {
AVFrame* frame;
int nb_bytes = INT_MAX;
@@ -147,7 +161,7 @@ void cache_audio_worker(Clip* c, bool scrubbing, QVector<Clip*>& nests) {
if (!c->reached_end) {
int loop = 0;
if (c->reverse && !c->audio_just_reset) {
if (reverse_audio && !c->audio_just_reset) {
avcodec_flush_buffers(c->codecCtx);
c->reached_end = false;
int64_t backtrack_seek = qMax(c->reverse_target - static_cast<int64_t>(av_q2d(av_inv_q(c->stream->time_base))), static_cast<int64_t>(0));
@@ -177,7 +191,7 @@ void cache_audio_worker(Clip* c, bool scrubbing, QVector<Clip*>& nests) {
dout << "reached EOF while reading";
#endif
// TODO revise usage of reached_end in audio
if (!c->reverse) {
if (!reverse_audio) {
c->reached_end = true;
} else {
}
@@ -198,11 +212,11 @@ void cache_audio_worker(Clip* c, bool scrubbing, QVector<Clip*>& nests) {
#ifdef AUDIOWARNINGS
dout << "reached EOF while pulling from filtergraph";
#endif
if (!c->reverse) break;
if (!reverse_audio) break;
}
}
if (c->reverse) {
if (reverse_audio) {
if (loop > 1) {
AVFrame* rev_frame = c->queue.at(1);
if (ret != AVERROR_EOF) {
@@ -313,7 +327,7 @@ void cache_audio_worker(Clip* c, bool scrubbing, QVector<Clip*>& nests) {
dout << "fsts:" << frame_sts << "tsts:" << target_sts << "nbs:" << nb_samples << "nbb:" << nb_bytes << "rev_targetToSec:" << (c->reverse_target * timebase);
dout << "fsi-calc:" << c->frame_sample_index;
#endif
if (c->reverse) c->frame_sample_index = nb_bytes - c->frame_sample_index;
if (reverse_audio) c->frame_sample_index = nb_bytes - c->frame_sample_index;
c->audio_just_reset = false;
}
@@ -343,7 +357,7 @@ void cache_audio_worker(Clip* c, bool scrubbing, QVector<Clip*>& nests) {
}
}
if (c->reverse) frame = c->queue.at(1);
if (reverse_audio) frame = c->queue.at(1);
#ifdef AUDIOWARNINGS
dout << "j" << c->frame_sample_index << nb_bytes;
@@ -368,20 +382,29 @@ void cache_audio_worker(Clip* c, bool scrubbing, QVector<Clip*>& nests) {
audio_write_lock.lock();
int audio_skip_resolution = 100; // FIXME magic number
int sample_skip = 4*qMax(0, qAbs(playback_speed)-1);
int sample_byte_size = av_get_bytes_per_sample(static_cast<AVSampleFormat>(frame->format));
while (c->frame_sample_index < nb_bytes
&& c->audio_buffer_write < audio_ibuffer_read+(audio_ibuffer_size>>1)
&& c->audio_buffer_write < buffer_timeline_out) {
int upper_byte_index = (c->audio_buffer_write+1)%audio_ibuffer_size;
int lower_byte_index = (c->audio_buffer_write)%audio_ibuffer_size;
qint16 old_sample = static_cast<qint16>((audio_ibuffer[upper_byte_index] & 0xFF) << 8 | (audio_ibuffer[lower_byte_index] & 0xFF));
qint16 new_sample = static_cast<qint16>((frame->data[0][c->frame_sample_index+1] & 0xFF) << 8 | (frame->data[0][c->frame_sample_index] & 0xFF));
qint16 mixed_sample = mix_audio_sample(old_sample, new_sample);
for (int i=0;i<frame->channels;i++) {
int upper_byte_index = (c->audio_buffer_write+1)%audio_ibuffer_size;
int lower_byte_index = (c->audio_buffer_write)%audio_ibuffer_size;
qint16 old_sample = static_cast<qint16>((audio_ibuffer[upper_byte_index] & 0xFF) << 8 | (audio_ibuffer[lower_byte_index] & 0xFF));
qint16 new_sample = static_cast<qint16>((frame->data[0][c->frame_sample_index+1] & 0xFF) << 8 | (frame->data[0][c->frame_sample_index] & 0xFF));
qint16 mixed_sample = mix_audio_sample(old_sample, new_sample);
audio_ibuffer[upper_byte_index] = quint8((mixed_sample >> 8) & 0xFF);
audio_ibuffer[lower_byte_index] = quint8(mixed_sample & 0xFF);
audio_ibuffer[upper_byte_index] = quint8((mixed_sample >> 8) & 0xFF);
audio_ibuffer[lower_byte_index] = quint8(mixed_sample & 0xFF);
c->audio_buffer_write+=2;
c->frame_sample_index+=2;
c->audio_buffer_write+=sample_byte_size;
c->frame_sample_index+=sample_byte_size;
}
c->frame_sample_index += sample_skip;
if (c->audio_reset) break;
}
@@ -398,7 +421,7 @@ void cache_audio_worker(Clip* c, bool scrubbing, QVector<Clip*>& nests) {
if (audio_thread != nullptr) audio_thread->notifyReceiver();
}
if (c->frame_sample_index == nb_bytes) {
if (c->frame_sample_index >= nb_bytes) {
c->frame_sample_index = -1;
} else {
// assume we have no more data to send
@@ -544,7 +567,7 @@ void cache_video_worker(Clip* c, long playhead) {
}
}
void reset_cache(Clip* c, long target_frame) {
void reset_cache(Clip* c, long target_frame, int playback_speed) {
// if we seek to a whole other place in the timeline, we'll need to reset the cache with new values
if (c->media == nullptr) {
if (c->track >= 0) {
@@ -607,9 +630,12 @@ void reset_cache(Clip* c, long target_frame) {
// seek (target_frame represents timeline timecode in frames, not clip timecode)
// bool reverse = (c->reverse != temp_reverse);
int64_t timestamp = seconds_to_timestamp(c, playhead_to_clip_seconds(c, target_frame));
if (c->reverse) {
bool temp_reverse = (playback_speed < 0);
if (c->reverse != temp_reverse) {
c->reverse_target = timestamp;
timestamp -= av_q2d(av_inv_q(c->stream->time_base));
#ifdef AUDIOWARNINGS
@@ -790,7 +816,8 @@ void open_clip_worker(Clip* clip) {
// set up cache
clip->queue.append(av_frame_alloc());
if (clip->reverse) {
// if (clip->reverse) {
if (true) {
AVFrame* reverse_frame = av_frame_alloc();
reverse_frame->format = sample_format;
@@ -886,22 +913,22 @@ void open_clip_worker(Clip* clip) {
qInfo() << "Clip opened on track" << clip->track;
}
void cache_clip_worker(Clip* clip, long playhead, bool reset, bool scrubbing, QVector<Clip*> nests) {
void cache_clip_worker(Clip* clip, long playhead, bool reset, bool scrubbing, QVector<Clip*> nests, int playback_speed) {
if (reset) {
// note: for video, playhead is in "internal clip" frames - for audio, it's the timeline playhead
reset_cache(clip, playhead);
reset_cache(clip, playhead, playback_speed);
clip->audio_reset = false;
}
if (clip->media == nullptr) {
if (clip->track >= 0) {
cache_audio_worker(clip, scrubbing, nests);
cache_audio_worker(clip, scrubbing, nests, playback_speed);
}
} else if (clip->media->get_type() == MEDIA_TYPE_FOOTAGE) {
if (clip->stream->codecpar->codec_type == AVMEDIA_TYPE_VIDEO) {
cache_video_worker(clip, playhead);
} else if (clip->stream->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
cache_audio_worker(clip, scrubbing, nests);
cache_audio_worker(clip, scrubbing, nests, playback_speed);
}
}
}
@@ -945,7 +972,7 @@ void Cacher::run() {
break;
} else {
while (true) {
cache_clip_worker(clip, playhead, reset, scrubbing, nests);
cache_clip_worker(clip, playhead, reset, scrubbing, nests, playback_speed);
if (clip->multithreaded && clip->cacher->interrupt && clip->track < 0) {
clip->cacher->interrupt = false;
} else {