MayaFlux::IO::AudioEncodeContext Class Reference

RAII owner of one audio stream's encoder, resampler, and sample FIFO. More...

#include <AudioEncodeContext.hpp>

Collaboration diagram for MayaFlux::IO::AudioEncodeContext:

Public Member Functions
	AudioEncodeContext ()=default
	AudioEncodeContext (AudioEncodeContext &&)=delete
	AudioEncodeContext (const AudioEncodeContext &)=delete
void	close ()
	Release all owned resources.
bool	drain (FFmpegMuxContext &mux)
	Flush the FIFO remainder and encoder internal delay to the mux.
bool	encode_frames (std::span< const double > interleaved, uint32_t num_frames, FFmpegMuxContext &mux)
	Encode a block of interleaved double-precision PCM frames.
bool	is_valid () const
	True if codec, resampler, and FIFO are all ready.
const std::string &	last_error () const
bool	open (FFmpegMuxContext &mux, uint32_t sample_rate, uint32_t channels, AVCodecID codec_id)
	Open the encoder and register an audio stream in the mux context.
AudioEncodeContext &	operator= (AudioEncodeContext &&)=delete
AudioEncodeContext &	operator= (const AudioEncodeContext &)=delete
	~AudioEncodeContext ()

Public Attributes
AVCodecContext *	codec_context {}
	Owned; freed in destructor.
AVAudioFifo *	fifo {}
	Owned; freed in destructor.
uint32_t	m_channels {}
uint32_t	m_sample_rate {}
int	m_stream_index { -1 }
AVStream *	stream {}
	Owned by mux; pointer cached here.
SwrContext *	swr_context {}
	Owned; freed in destructor.

Private Member Functions
bool	drain_packets (FFmpegMuxContext &mux)
	Drain all packets currently available from the encoder into mux.
bool	send_fifo_frame (FFmpegMuxContext &mux, bool pad_to_frame_size)
	Pull one frame from the FIFO and send it to the encoder.
bool	setup_fifo ()
bool	setup_resampler ()

Private Attributes
std::string	m_last_error
int64_t	m_pts = 0

Detailed Description

RAII owner of one audio stream's encoder, resampler, and sample FIFO.

Encapsulates all audio-stream-specific FFmpeg objects on the write path:

• AVCodecContext for the selected or inferred encoder
• SwrContext converting AV_SAMPLE_FMT_DBL (engine canonical) to the encoder's native sample format
• AVAudioFifo absorbing arbitrary-size input chunks and emitting fixed-size frames required by block-aligned encoders (AAC: 1024, Vorbis: variable)
• AVStream registered inside the supplied FFmpegMuxContext

Does NOT own AVFormatContext — that belongs to FFmpegMuxContext.

Destruction order (enforced in destructor): fifo → swr_context → codec_context The associated FFmpegMuxContext must outlive this object.

Open sequence:

1. FFmpegMuxContext::open()
2. AudioEncodeContext::open() ← adds stream, configures codec+swr+fifo
3. FFmpegMuxContext::write_header()
4. encode_frames() loop
5. drain() ← flushes fifo remainder + encoder delay
6. FFmpegMuxContext::close() ← writes trailer

Definition at line 40 of file AudioEncodeContext.hpp.

---

The documentation for this class was generated from the following files:

• AudioEncodeContext.hpp
• AudioEncodeContext.cpp