mirror of
https://github.com/RGBCube/serenity
synced 2025-07-26 11:37:44 +00:00
LibVideo: Calculate Block timestamps for Matroska according to spec
Tracks have a timestamp scale value that should be present which scales each block's timestamp offset to allow video to be synced with audio. They should also contain a CodecDelay element and may also contain a TrackOffset that offsets the block timestamps.
This commit is contained in:
parent
ceb7632862
commit
a58bf7c3d1
4 changed files with 49 additions and 12 deletions
|
@ -114,6 +114,12 @@ public:
|
||||||
void set_language(FlyString const& language) { m_language = language; }
|
void set_language(FlyString const& language) { m_language = language; }
|
||||||
FlyString codec_id() const { return m_codec_id; }
|
FlyString codec_id() const { return m_codec_id; }
|
||||||
void set_codec_id(FlyString const& codec_id) { m_codec_id = codec_id; }
|
void set_codec_id(FlyString const& codec_id) { m_codec_id = codec_id; }
|
||||||
|
double timestamp_scale() const { return m_timestamp_scale; }
|
||||||
|
void set_timestamp_scale(double timestamp_scale) { m_timestamp_scale = timestamp_scale; }
|
||||||
|
u64 codec_delay() const { return m_codec_delay; }
|
||||||
|
void set_codec_delay(u64 codec_delay) { m_codec_delay = codec_delay; }
|
||||||
|
u64 timestamp_offset() const { return m_timestamp_offset; }
|
||||||
|
void set_timestamp_offset(u64 timestamp_offset) { m_timestamp_offset = timestamp_offset; }
|
||||||
Optional<VideoTrack> video_track() const
|
Optional<VideoTrack> video_track() const
|
||||||
{
|
{
|
||||||
if (track_type() != Video)
|
if (track_type() != Video)
|
||||||
|
@ -135,6 +141,9 @@ private:
|
||||||
TrackType m_track_type { Invalid };
|
TrackType m_track_type { Invalid };
|
||||||
FlyString m_language = "eng";
|
FlyString m_language = "eng";
|
||||||
FlyString m_codec_id;
|
FlyString m_codec_id;
|
||||||
|
double m_timestamp_scale { 1 };
|
||||||
|
u64 m_codec_delay { 0 };
|
||||||
|
u64 m_timestamp_offset { 0 };
|
||||||
|
|
||||||
union {
|
union {
|
||||||
VideoTrack m_video_track {};
|
VideoTrack m_video_track {};
|
||||||
|
|
|
@ -44,6 +44,8 @@ constexpr u32 TRACK_UID_ID = 0x73C5;
|
||||||
constexpr u32 TRACK_TYPE_ID = 0x83;
|
constexpr u32 TRACK_TYPE_ID = 0x83;
|
||||||
constexpr u32 TRACK_LANGUAGE_ID = 0x22B59C;
|
constexpr u32 TRACK_LANGUAGE_ID = 0x22B59C;
|
||||||
constexpr u32 TRACK_CODEC_ID = 0x86;
|
constexpr u32 TRACK_CODEC_ID = 0x86;
|
||||||
|
constexpr u32 TRACK_TIMESTAMP_SCALE_ID = 0x23314F;
|
||||||
|
constexpr u32 TRACK_OFFSET_ID = 0x537F;
|
||||||
constexpr u32 TRACK_VIDEO_ID = 0xE0;
|
constexpr u32 TRACK_VIDEO_ID = 0xE0;
|
||||||
constexpr u32 TRACK_AUDIO_ID = 0xE1;
|
constexpr u32 TRACK_AUDIO_ID = 0xE1;
|
||||||
|
|
||||||
|
@ -416,6 +418,14 @@ static DecoderErrorOr<TrackEntry> parse_track_entry(Streamer& streamer)
|
||||||
track_entry.set_codec_id(TRY_READ(streamer.read_string()));
|
track_entry.set_codec_id(TRY_READ(streamer.read_string()));
|
||||||
dbgln_if(MATROSKA_TRACE_DEBUG, "Read Track's CodecID attribute: {}", track_entry.codec_id());
|
dbgln_if(MATROSKA_TRACE_DEBUG, "Read Track's CodecID attribute: {}", track_entry.codec_id());
|
||||||
break;
|
break;
|
||||||
|
case TRACK_TIMESTAMP_SCALE_ID:
|
||||||
|
track_entry.set_timestamp_scale(TRY_READ(streamer.read_float()));
|
||||||
|
dbgln_if(MATROSKA_TRACE_DEBUG, "Read Track's TrackTimestampScale attribute: {}", track_entry.timestamp_scale());
|
||||||
|
break;
|
||||||
|
case TRACK_OFFSET_ID:
|
||||||
|
track_entry.set_timestamp_offset(TRY_READ(streamer.read_variable_size_signed_integer()));
|
||||||
|
dbgln_if(MATROSKA_TRACE_DEBUG, "Read Track's TrackOffset attribute: {}", track_entry.timestamp_offset());
|
||||||
|
break;
|
||||||
case TRACK_VIDEO_ID:
|
case TRACK_VIDEO_ID:
|
||||||
track_entry.set_video_track(TRY(parse_video_track_information(streamer)));
|
track_entry.set_video_track(TRY(parse_video_track_information(streamer)));
|
||||||
break;
|
break;
|
||||||
|
@ -520,7 +530,7 @@ static DecoderErrorOr<Cluster> parse_cluster(Streamer& streamer, u64 timestamp_s
|
||||||
return cluster;
|
return cluster;
|
||||||
}
|
}
|
||||||
|
|
||||||
static DecoderErrorOr<Block> parse_simple_block(Streamer& streamer, Time cluster_timestamp, u64 timestamp_scale)
|
static DecoderErrorOr<Block> parse_simple_block(Streamer& streamer, Time cluster_timestamp, u64 segment_timestamp_scale, TrackEntry track)
|
||||||
{
|
{
|
||||||
Block block;
|
Block block;
|
||||||
|
|
||||||
|
@ -529,7 +539,23 @@ static DecoderErrorOr<Block> parse_simple_block(Streamer& streamer, Time cluster
|
||||||
auto position_before_track_number = streamer.position();
|
auto position_before_track_number = streamer.position();
|
||||||
block.set_track_number(TRY_READ(streamer.read_variable_size_integer()));
|
block.set_track_number(TRY_READ(streamer.read_variable_size_integer()));
|
||||||
|
|
||||||
block.set_timestamp(cluster_timestamp + Time::from_nanoseconds(TRY_READ(streamer.read_i16()) * timestamp_scale));
|
// https://www.matroska.org/technical/notes.html
|
||||||
|
// Block Timestamps:
|
||||||
|
// The Block Element and SimpleBlock Element store their timestamps as signed integers,
|
||||||
|
// relative to the Cluster\Timestamp value of the Cluster they are stored in. To get the
|
||||||
|
// timestamp of a Block or SimpleBlock in nanoseconds you have to use the following formula:
|
||||||
|
// `( Cluster\Timestamp + ( block timestamp * TrackTimestampScale ) ) * TimestampScale`
|
||||||
|
//
|
||||||
|
// When a CodecDelay Element is set, its value MUST be substracted from each Block timestamp
|
||||||
|
// of that track. To get the timestamp in nanoseconds of the first frame in a Block or
|
||||||
|
// SimpleBlock, the formula becomes:
|
||||||
|
// `( ( Cluster\Timestamp + ( block timestamp * TrackTimestampScale ) ) * TimestampScale ) - CodecDelay`
|
||||||
|
Time timestamp_offset = Time::from_nanoseconds(static_cast<i64>(static_cast<double>(TRY_READ(streamer.read_i16()) * segment_timestamp_scale) * track.timestamp_scale()));
|
||||||
|
timestamp_offset -= Time::from_nanoseconds(static_cast<i64>(track.codec_delay()));
|
||||||
|
// This is only mentioned in the elements specification under TrackOffset.
|
||||||
|
// https://www.matroska.org/technical/elements.html
|
||||||
|
timestamp_offset += Time::from_nanoseconds(static_cast<i64>(track.timestamp_offset()));
|
||||||
|
block.set_timestamp(cluster_timestamp + timestamp_offset);
|
||||||
|
|
||||||
auto flags = TRY_READ(streamer.read_octet());
|
auto flags = TRY_READ(streamer.read_octet());
|
||||||
block.set_only_keyframes((flags & (1u << 7u)) != 0);
|
block.set_only_keyframes((flags & (1u << 7u)) != 0);
|
||||||
|
@ -596,7 +622,7 @@ DecoderErrorOr<SampleIterator> Reader::create_sample_iterator(u64 track_number)
|
||||||
auto position = optional_position.value() - get_element_id_size(CLUSTER_ELEMENT_ID) - m_segment_contents_position;
|
auto position = optional_position.value() - get_element_id_size(CLUSTER_ELEMENT_ID) - m_segment_contents_position;
|
||||||
|
|
||||||
dbgln_if(MATROSKA_DEBUG, "Creating sample iterator starting at {} relative to segment at {}", position, m_segment_contents_position);
|
dbgln_if(MATROSKA_DEBUG, "Creating sample iterator starting at {} relative to segment at {}", position, m_segment_contents_position);
|
||||||
return SampleIterator(this->m_mapped_file, segment_view, track_number, position, TRY(segment_information()).timestamp_scale());
|
return SampleIterator(this->m_mapped_file, segment_view, TRY(track_for_track_number(track_number)), TRY(segment_information()).timestamp_scale(), position);
|
||||||
}
|
}
|
||||||
|
|
||||||
static DecoderErrorOr<bool> find_keyframe_before_timestamp(SampleIterator& iterator, Time const& timestamp)
|
static DecoderErrorOr<bool> find_keyframe_before_timestamp(SampleIterator& iterator, Time const& timestamp)
|
||||||
|
@ -643,7 +669,7 @@ DecoderErrorOr<void> Reader::seek_to_random_access_point(SampleIterator& iterato
|
||||||
// FIXME: This could cache the keyframes it finds. Is it worth doing? Probably not, most files will have Cues :^)
|
// FIXME: This could cache the keyframes it finds. Is it worth doing? Probably not, most files will have Cues :^)
|
||||||
if (timestamp < iterator.last_timestamp() || iterator.last_timestamp().is_negative()) {
|
if (timestamp < iterator.last_timestamp() || iterator.last_timestamp().is_negative()) {
|
||||||
// If the timestamp is before the iterator's current position, then we need to start from the beginning of the Segment.
|
// If the timestamp is before the iterator's current position, then we need to start from the beginning of the Segment.
|
||||||
iterator = TRY(create_sample_iterator(iterator.m_track_id));
|
iterator = TRY(create_sample_iterator(iterator.m_track.track_number()));
|
||||||
if (!TRY(find_keyframe_before_timestamp(iterator, timestamp)))
|
if (!TRY(find_keyframe_before_timestamp(iterator, timestamp)))
|
||||||
return DecoderError::corrupted("No random access points found"sv);
|
return DecoderError::corrupted("No random access points found"sv);
|
||||||
|
|
||||||
|
@ -678,11 +704,11 @@ DecoderErrorOr<Block> SampleIterator::next_block()
|
||||||
|
|
||||||
if (element_id == CLUSTER_ELEMENT_ID) {
|
if (element_id == CLUSTER_ELEMENT_ID) {
|
||||||
dbgln_if(MATROSKA_DEBUG, " Iterator is parsing new cluster.");
|
dbgln_if(MATROSKA_DEBUG, " Iterator is parsing new cluster.");
|
||||||
m_current_cluster = TRY(parse_cluster(streamer, m_timestamp_scale));
|
m_current_cluster = TRY(parse_cluster(streamer, m_segment_timestamp_scale));
|
||||||
} else if (element_id == SIMPLE_BLOCK_ID) {
|
} else if (element_id == SIMPLE_BLOCK_ID) {
|
||||||
dbgln_if(MATROSKA_TRACE_DEBUG, " Iterator is parsing new block.");
|
dbgln_if(MATROSKA_TRACE_DEBUG, " Iterator is parsing new block.");
|
||||||
auto candidate_block = TRY(parse_simple_block(streamer, m_current_cluster->timestamp(), m_timestamp_scale));
|
auto candidate_block = TRY(parse_simple_block(streamer, m_current_cluster->timestamp(), m_segment_timestamp_scale, m_track));
|
||||||
if (candidate_block.track_number() == m_track_id)
|
if (candidate_block.track_number() == m_track.track_number())
|
||||||
block = move(candidate_block);
|
block = move(candidate_block);
|
||||||
} else {
|
} else {
|
||||||
dbgln_if(MATROSKA_TRACE_DEBUG, " Iterator is skipping unknown element with ID {:#010x}.", element_id);
|
dbgln_if(MATROSKA_TRACE_DEBUG, " Iterator is skipping unknown element with ID {:#010x}.", element_id);
|
||||||
|
|
|
@ -80,12 +80,12 @@ public:
|
||||||
private:
|
private:
|
||||||
friend class Reader;
|
friend class Reader;
|
||||||
|
|
||||||
SampleIterator(RefPtr<Core::MappedFile> file, ReadonlyBytes data, u64 track_id, size_t position, u64 timestamp_scale)
|
SampleIterator(RefPtr<Core::MappedFile> file, ReadonlyBytes data, TrackEntry track, u64 timestamp_scale, size_t position)
|
||||||
: m_file(move(file))
|
: m_file(move(file))
|
||||||
, m_data(data)
|
, m_data(data)
|
||||||
, m_track_id(track_id)
|
, m_track(move(track))
|
||||||
|
, m_segment_timestamp_scale(timestamp_scale)
|
||||||
, m_position(position)
|
, m_position(position)
|
||||||
, m_timestamp_scale(timestamp_scale)
|
|
||||||
{
|
{
|
||||||
}
|
}
|
||||||
|
|
||||||
|
@ -93,12 +93,12 @@ private:
|
||||||
|
|
||||||
RefPtr<Core::MappedFile> m_file;
|
RefPtr<Core::MappedFile> m_file;
|
||||||
ReadonlyBytes m_data;
|
ReadonlyBytes m_data;
|
||||||
u64 m_track_id;
|
TrackEntry m_track;
|
||||||
|
u64 m_segment_timestamp_scale { 0 };
|
||||||
|
|
||||||
// Must always point to an element ID or the end of the stream.
|
// Must always point to an element ID or the end of the stream.
|
||||||
size_t m_position { 0 };
|
size_t m_position { 0 };
|
||||||
|
|
||||||
u64 m_timestamp_scale { 0 };
|
|
||||||
Time m_last_timestamp { Time::min() };
|
Time m_last_timestamp { Time::min() };
|
||||||
|
|
||||||
Optional<Cluster> m_current_cluster;
|
Optional<Cluster> m_current_cluster;
|
||||||
|
|
|
@ -36,6 +36,8 @@ ErrorOr<int> serenity_main(Main::Arguments)
|
||||||
outln("\tTrack has TrackType {}", static_cast<u8>(track_entry.track_type()));
|
outln("\tTrack has TrackType {}", static_cast<u8>(track_entry.track_type()));
|
||||||
outln("\tTrack has Language \"{}\"", track_entry.language().characters());
|
outln("\tTrack has Language \"{}\"", track_entry.language().characters());
|
||||||
outln("\tTrack has CodecID \"{}\"", track_entry.codec_id().characters());
|
outln("\tTrack has CodecID \"{}\"", track_entry.codec_id().characters());
|
||||||
|
outln("\tTrack has TrackTimestampScale {}", track_entry.timestamp_scale());
|
||||||
|
outln("\tTrack has CodecDelay {}", track_entry.codec_delay());
|
||||||
|
|
||||||
if (track_entry.track_type() == Video::Matroska::TrackEntry::TrackType::Video) {
|
if (track_entry.track_type() == Video::Matroska::TrackEntry::TrackType::Video) {
|
||||||
auto const video_track = track_entry.video_track().value();
|
auto const video_track = track_entry.video_track().value();
|
||||||
|
|
Loading…
Add table
Add a link
Reference in a new issue