Files
voice-cat/core/include/voicecat.h

252 lines
10 KiB
C
Raw Normal View History

/*
* voicecat.h the C ABI for libvoicecat.
*
* This is the single boundary every front-end calls: Swift (macOS/iOS) and C# (Windows)
* both bind to this header, and the server links the same core. It is C-linkage and
* handle-based so it is stable and trivially bindable from any language.
*
* Design: docs/architecture.md §4. Everything here is async + event-driven calls return
* immediately and results/state changes arrive via the vc_callbacks.on_event callback.
*
* STATUS: M0 skeleton. Implementations live in core/src and currently return
* VC_ERR_NOT_IMPLEMENTED. The shapes below are the contract to build against.
*/
#ifndef VOICECAT_H
#define VOICECAT_H
#include <stddef.h>
#include <stdint.h>
#if defined(__cplusplus)
extern "C" {
#endif
/* ── Export macro ─────────────────────────────────────────────────────────── */
#if defined(VOICECAT_STATIC)
#define VC_API
#elif defined(_WIN32)
#if defined(VOICECAT_BUILDING)
#define VC_API __declspec(dllexport)
#else
#define VC_API __declspec(dllimport)
#endif
#else
#if defined(VOICECAT_BUILDING)
#define VC_API __attribute__((visibility("default")))
#else
#define VC_API
#endif
#endif
/* ── Version ──────────────────────────────────────────────────────────────── */
#define VOICECAT_VERSION_MAJOR 0
#define VOICECAT_VERSION_MINOR 0
#define VOICECAT_VERSION_PATCH 1
/* The control-protocol version this build speaks (docs/protocol.md §4). */
#define VOICECAT_PROTOCOL_VERSION 1
/* ── Result codes ─────────────────────────────────────────────────────────── */
typedef enum vc_result {
VC_OK = 0,
VC_ERR_NOT_IMPLEMENTED = 1, /* skeleton stub */
VC_ERR_INVALID_ARG = 2,
VC_ERR_NOT_CONNECTED = 3,
VC_ERR_ALREADY = 4,
VC_ERR_AUTH_FAILED = 5,
VC_ERR_PERMISSION_DENIED = 6,
VC_ERR_TIMEOUT = 7,
VC_ERR_IO = 8,
VC_ERR_PROTOCOL = 9,
VC_ERR_CRYPTO = 10,
VC_ERR_AUDIO = 11,
VC_ERR_INTERNAL = 12,
} vc_result;
typedef enum vc_log_level {
VC_LOG_TRACE = 0,
VC_LOG_DEBUG = 1,
VC_LOG_INFO = 2,
VC_LOG_WARN = 3,
VC_LOG_ERROR = 4,
VC_LOG_OFF = 5,
} vc_log_level;
typedef enum vc_connection_state {
VC_STATE_DISCONNECTED = 0,
VC_STATE_CONNECTING = 1,
VC_STATE_TLS_HANDSHAKE = 2,
VC_STATE_AUTHENTICATING = 3,
VC_STATE_CONNECTED = 4,
} vc_connection_state;
typedef enum vc_text_scope {
VC_TEXT_CHANNEL = 0,
VC_TEXT_PRIVATE = 1,
VC_TEXT_SERVER = 2,
} vc_text_scope;
typedef enum vc_device_kind {
VC_DEVICE_INPUT = 0,
VC_DEVICE_OUTPUT = 1,
} vc_device_kind;
typedef enum vc_stream_kind {
VC_STREAM_MIC = 0,
VC_STREAM_SCREEN_AUDIO = 1, /* system/desktop audio (docs/voice.md §9) */
VC_STREAM_AUX_DEVICE = 2,
} vc_stream_kind;
/* Send-side input gate (docs/voice.md §11). */
typedef enum vc_input_mode {
VC_INPUT_VOICE_ACTIVATION = 0,
VC_INPUT_PUSH_TO_TALK = 1,
} vc_input_mode;
typedef enum vc_event_type {
VC_EVENT_CONNECTION_STATE = 0, /* connection_state set */
VC_EVENT_AUTH_RESULT = 1, /* result set; user_id = self on success */
VC_EVENT_CHANNEL_LIST = 2, /* channel tree snapshot/delta available */
VC_EVENT_USER_JOINED = 3, /* user_id, channel_id, text = nickname */
VC_EVENT_USER_LEFT = 4, /* user_id */
VC_EVENT_USER_UPDATED = 5, /* user_id */
VC_EVENT_TEXT_MESSAGE = 6, /* text_scope, user_id (sender), channel_id, text */
VC_EVENT_STREAM_STARTED = 7, /* user_id, stream_id */
VC_EVENT_STREAM_STOPPED = 8, /* user_id, stream_id */
VC_EVENT_TALK_STATE = 9, /* user_id, stream_id, u32a = talking(0/1) */
VC_EVENT_ERROR = 10, /* result, text */
VC_EVENT_DISCONNECTED = 11, /* result, text = reason */
} vc_event_type;
/* ── Structs ──────────────────────────────────────────────────────────────── */
/*
* An event delivered to vc_callbacks.on_event. Pointer fields are owned by the core and
* valid ONLY for the duration of the callback copy what you need. Which fields are
* meaningful depends on `type` (see vc_event_type comments above).
*/
typedef struct vc_event {
vc_event_type type;
vc_connection_state connection_state;
int32_t result; /* vc_result */
uint32_t user_id;
uint32_t channel_id;
uint32_t stream_id;
vc_text_scope text_scope;
uint32_t u32a; /* generic small payload, meaning per event type */
const char* text;
uint64_t timestamp_unix_ms;
} vc_event;
typedef struct vc_callbacks {
/* State changes, messages, presence. Called on the core's event thread. */
void (*on_event)(void* user, const vc_event* ev);
/* Throttled level meter (RMS 0..1) for a local or remote stream; may be NULL. */
void (*on_level)(void* user, uint32_t stream_id, float rms);
void* user;
} vc_callbacks;
typedef struct vc_config {
const char* client_name; /* e.g. "VoiceCat-macOS" */
const char* client_version; /* e.g. "0.0.1" */
vc_log_level log_level;
} vc_config;
typedef struct vc_stream_desc {
vc_stream_kind kind;
const char* device_id; /* NULL = default device for this kind */
const char* label; /* human label, e.g. "Microphone" */
} vc_stream_desc;
feat(M3): multi-stream & per-channel tuning Implements docs/roadmap.md M3: multiple concurrent streams per user (MIC + SCREEN_AUDIO + AUX_DEVICE), independent per-stream receiver gain/mute/noise- reduction, talk indicators, and enforced per-channel Opus configurability (mono/stereo, bitrate, frame size, FEC/DTX, application). Bugs fixed along the way (found while implementing, not pre-existing scope): - Server hard-coded stream_id=1 for every announce, so a second stream from the same user silently overwrote the first in SessionRegistry::set_user_stream. Now a per-session counter (ConnSession::next_stream_id_); handle_stream_stop validates against announced_stream_ids_ before clearing. - Client dropped mode/dtx/complexity/application from effective_audio even for the single M2 stream -- only sample_rate/bitrate_bps/frame_ms/fec were ever applied to OpusParams. Fixed on both the send (handle_stream_announce_result) and receive (sync_remote_streams) paths via a shared opus_params_from_audio_config() helper. - OpusEncoder always used OPUS_APPLICATION_VOIP; added OpusParams::application and wired it through. - on_playback's per-stream decode passed the wrong frame_size to opus_decode (total samples instead of samples-per-channel), which would have overflowed the decode buffer for any stereo stream. - teardown_voice() raced when called concurrently from run_io()'s own cleanup and from disconnect() on a different thread -- both could see udp_thread_/talk_timer_thread_ as joinable() at once and race to join() the same std::thread (intermittent std::system_error under ctest). Fixed with a teardown_mu_ guard instead of carrying the flake forward. New: - Per-channel AudioConfig: SessionRegistry now seeds Lobby (mono/24kbps/VOIP/ FEC+DTX) and a new "Music Room" channel (stereo/128kbps/AUDIO/no DTX); handle_stream_announce enforces the channel's config, clamping (not overriding) bitrate_bps to its ceiling. - core/src/core/client.h/.cpp: local-stream state is now a std::unordered_map<int, LocalStream> keyed by vc_stream_kind, with request_id-correlated announce/result handling (request_id already round-tripped on the wire; just wasn't read before). on_capture_frame is kind-aware and upmixes mono capture to stereo when a stream's config calls for it. set_self_mute's mic_muted now only gates the MIC kind. NS is wired through set_remote_stream. New run_talk_timer() thread emits VC_EVENT_TALK_STATE from both remote and local edge detection. - core/src/audio/audio_engine.h/.cpp: kind-keyed injection taps (inject_capture), stereo-to-mono downmix at the decode/mix boundary, RemoteStream gains recv_ns (lazy ApmProcessor) + noise_reduction_enabled and last_voice_ms/talking; new set_stream_noise_reduction() and poll_talk_transitions(). - core/src/session/session.h/.cpp: Stream now carries the full AudioConfig, not just sample_rate/frame_ms. - New additive C ABI (core/include/voicecat.h): vc_audio_config + vc_get_stream_audio_config (effective Opus config for any stream you own or a peer's); vc_test_inject_capture (test-only synthetic PCM injection, clearly marked, mirrors AudioEngine::inject_capture). - tests/test_m3_multistream.cpp: the M3 exit criterion through the real ABI (mirrors test_voice_client_abi.cpp's approach, not raw sockets) -- two concurrent local streams, independent gain/mute/NS control, per-channel config divergence via vc_get_stream_audio_config, talk indicators. Explicitly out of scope for this pass (tracked in PROGRESS.md, not silently dropped): VAD/PTT input gate + device enumeration; real WASAPI loopback capture for SCREEN_AUDIO (synthetic injection only); true stereo playback output (AudioEngine's mixer/output device stays mono -- Opus itself is fully stereo-correct on the wire). ctest --test-dir build/m1-dev: 11/11 green, verified across 3 consecutive full-suite runs plus 8 standalone runs of the new test. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-16 14:12:37 +02:00
/* The effective Opus configuration in use for a stream — for a stream you own, this is
* StreamAnnounceResult.effective_audio (channel-enforced, docs/voice.md §3); for a remote
* stream, it's the peer's broadcast StreamInfo.audio. See vc_get_stream_audio_config. */
typedef struct vc_audio_config {
uint32_t codec; /* 0 = OPUS */
uint32_t mode; /* 0 = mono, 1 = stereo */
uint32_t sample_rate;
uint32_t bitrate_bps;
uint32_t frame_ms;
uint32_t application; /* 0 = VOIP, 1 = AUDIO, 2 = LOWDELAY */
int fec; /* bool */
uint32_t expected_packet_loss; /* % 0..100 */
int dtx; /* bool */
uint32_t complexity; /* 0..10 */
} vc_audio_config;
typedef struct vc_device {
const char* id;
const char* name;
int is_default; /* bool */
} vc_device;
typedef struct vc_device_list {
vc_device* items;
size_t count;
} vc_device_list;
/* Opaque client handle. */
typedef struct vc_client vc_client;
/* ── Lifecycle ────────────────────────────────────────────────────────────── */
VC_API const char* vc_version_string(void);
VC_API const char* vc_result_string(vc_result code);
VC_API vc_client* vc_client_create(const vc_config* cfg, vc_callbacks cb);
VC_API void vc_client_destroy(vc_client* c);
/* ── Connection & auth (async; results via on_event) ──────────────────────── */
VC_API vc_result vc_connect(vc_client* c, const char* host, uint16_t port);
VC_API vc_result vc_disconnect(vc_client* c);
VC_API vc_result vc_authenticate_guest(vc_client* c, const char* nickname);
VC_API vc_result vc_authenticate_user(vc_client* c, const char* username,
const char* password);
/* ── Channels ─────────────────────────────────────────────────────────────── */
VC_API vc_result vc_join_channel(vc_client* c, uint32_t channel_id,
const char* password /* nullable */);
VC_API vc_result vc_leave_channel(vc_client* c);
/* ── Local media streams (mic / screen audio / aux) ───────────────────────── */
VC_API vc_result vc_stream_start(vc_client* c, const vc_stream_desc* desc,
uint32_t* out_stream_id);
VC_API vc_result vc_stream_stop(vc_client* c, uint32_t stream_id);
VC_API vc_result vc_set_input_device(vc_client* c, uint32_t stream_id,
const char* device_id);
/* Send-side: input gate mode + PTT key state, and self mute/deafen. */
VC_API vc_result vc_set_input_mode(vc_client* c, vc_input_mode mode);
VC_API vc_result vc_set_push_to_talk(vc_client* c, int active /* bool */);
VC_API vc_result vc_set_self_mute(vc_client* c, int mic_muted, int deafened);
/* Receive-side, per remote stream, all LOCAL (no protocol traffic) — docs/voice.md §10:
* gain (0..) , mute, and listener-chosen noise reduction on a specific user's stream. */
VC_API vc_result vc_set_remote_stream(vc_client* c, uint32_t user_id, uint32_t stream_id,
float gain, int muted, int noise_reduction);
feat(M3): multi-stream & per-channel tuning Implements docs/roadmap.md M3: multiple concurrent streams per user (MIC + SCREEN_AUDIO + AUX_DEVICE), independent per-stream receiver gain/mute/noise- reduction, talk indicators, and enforced per-channel Opus configurability (mono/stereo, bitrate, frame size, FEC/DTX, application). Bugs fixed along the way (found while implementing, not pre-existing scope): - Server hard-coded stream_id=1 for every announce, so a second stream from the same user silently overwrote the first in SessionRegistry::set_user_stream. Now a per-session counter (ConnSession::next_stream_id_); handle_stream_stop validates against announced_stream_ids_ before clearing. - Client dropped mode/dtx/complexity/application from effective_audio even for the single M2 stream -- only sample_rate/bitrate_bps/frame_ms/fec were ever applied to OpusParams. Fixed on both the send (handle_stream_announce_result) and receive (sync_remote_streams) paths via a shared opus_params_from_audio_config() helper. - OpusEncoder always used OPUS_APPLICATION_VOIP; added OpusParams::application and wired it through. - on_playback's per-stream decode passed the wrong frame_size to opus_decode (total samples instead of samples-per-channel), which would have overflowed the decode buffer for any stereo stream. - teardown_voice() raced when called concurrently from run_io()'s own cleanup and from disconnect() on a different thread -- both could see udp_thread_/talk_timer_thread_ as joinable() at once and race to join() the same std::thread (intermittent std::system_error under ctest). Fixed with a teardown_mu_ guard instead of carrying the flake forward. New: - Per-channel AudioConfig: SessionRegistry now seeds Lobby (mono/24kbps/VOIP/ FEC+DTX) and a new "Music Room" channel (stereo/128kbps/AUDIO/no DTX); handle_stream_announce enforces the channel's config, clamping (not overriding) bitrate_bps to its ceiling. - core/src/core/client.h/.cpp: local-stream state is now a std::unordered_map<int, LocalStream> keyed by vc_stream_kind, with request_id-correlated announce/result handling (request_id already round-tripped on the wire; just wasn't read before). on_capture_frame is kind-aware and upmixes mono capture to stereo when a stream's config calls for it. set_self_mute's mic_muted now only gates the MIC kind. NS is wired through set_remote_stream. New run_talk_timer() thread emits VC_EVENT_TALK_STATE from both remote and local edge detection. - core/src/audio/audio_engine.h/.cpp: kind-keyed injection taps (inject_capture), stereo-to-mono downmix at the decode/mix boundary, RemoteStream gains recv_ns (lazy ApmProcessor) + noise_reduction_enabled and last_voice_ms/talking; new set_stream_noise_reduction() and poll_talk_transitions(). - core/src/session/session.h/.cpp: Stream now carries the full AudioConfig, not just sample_rate/frame_ms. - New additive C ABI (core/include/voicecat.h): vc_audio_config + vc_get_stream_audio_config (effective Opus config for any stream you own or a peer's); vc_test_inject_capture (test-only synthetic PCM injection, clearly marked, mirrors AudioEngine::inject_capture). - tests/test_m3_multistream.cpp: the M3 exit criterion through the real ABI (mirrors test_voice_client_abi.cpp's approach, not raw sockets) -- two concurrent local streams, independent gain/mute/NS control, per-channel config divergence via vc_get_stream_audio_config, talk indicators. Explicitly out of scope for this pass (tracked in PROGRESS.md, not silently dropped): VAD/PTT input gate + device enumeration; real WASAPI loopback capture for SCREEN_AUDIO (synthetic injection only); true stereo playback output (AudioEngine's mixer/output device stays mono -- Opus itself is fully stereo-correct on the wire). ctest --test-dir build/m1-dev: 11/11 green, verified across 3 consecutive full-suite runs plus 8 standalone runs of the new test. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-16 14:12:37 +02:00
/* Effective Opus config in use for (user_id, stream_id) — your own stream or a peer's.
* VC_ERR_INVALID_ARG if the user/stream isn't known. */
VC_API vc_result vc_get_stream_audio_config(vc_client* c, uint32_t user_id, uint32_t stream_id,
vc_audio_config* out);
/* TEST-ONLY — not for production use. Bypasses the real capture device, injecting raw PCM
* directly into the named local stream's encode pipeline (see AudioEngine::inject_capture).
* Exists so automated tests can drive the real vc_client/ABI path end-to-end without a
* microphone. `stream_id` is the id returned by vc_stream_start. */
VC_API vc_result vc_test_inject_capture(vc_client* c, uint32_t stream_id, const int16_t* pcm,
size_t samples);
/* ── Text ─────────────────────────────────────────────────────────────────── */
VC_API vc_result vc_send_text(vc_client* c, vc_text_scope scope, uint32_t target_id,
const char* utf8);
/* ── Device enumeration (for UI pickers) ──────────────────────────────────── */
VC_API vc_result vc_list_devices(vc_client* c, vc_device_kind kind, vc_device_list* out);
VC_API void vc_free_device_list(vc_device_list* list);
#if defined(__cplusplus)
} /* extern "C" */
#endif
#endif /* VOICECAT_H */