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

547 lines
28 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: real, behind VOICECAT_HAS_NET (the `dev`/`release`/`server-release` presets
* vcpkg deps on; see docs/building.md). As of M3, control plane, voice, multi-stream, device
* enumeration, VAD/PTT, and stereo playback all work for real via core/src/core/client.cpp.
* The no-deps `skeleton` preset still links a stub vc_client that returns VC_ERR_NOT_IMPLEMENTED
* for everything below `connect`, purely to keep that skeleton build green. webrtc AEC/NS/AGC
* remains an inert passthrough regardless of preset (no Windows/MSVC port upstream
* docs/voice.md §8/§11, PROGRESS.md).
*/
#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,
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
/* M4: between TLS_HANDSHAKE and AUTHENTICATING — the handshake succeeded and the core is
* waiting for vc_confirm_server_identity() (see VC_EVENT_SERVER_IDENTITY below). Appended
* at the end (not inserted) to keep existing enum values stable additive-only ABI. */
VC_STATE_VERIFYING_IDENTITY = 5,
} 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,
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
/* Transmit unconditionally — no VAD gate. Added at the end to keep existing values stable. */
VC_INPUT_ALWAYS_ON = 2,
} 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 */
/* M4 additions — appended, not inserted, to keep existing enum values stable. */
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
VC_EVENT_JOIN_RESULT = 12, /* result (VC_OK/VC_ERR_*), channel_id, text = error on
failure. Reply to vc_join_channel(). */
VC_EVENT_SERVER_IDENTITY = 13, /* u32a = vc_tofu_status, text = hex-encoded TLS leaf-cert
SHA-256 fingerprint (the value being pinned see
vc_confirm_server_identity). Emitted once per connect
attempt, right after the TLS handshake succeeds. The
connection is held open until vc_confirm_server_identity()
is called. */
/* M5 additions — appended, not inserted. */
VC_EVENT_GENERIC_RESULT = 14, /* result, u32a = server error code, text = message. Reply
to vc_kick_user/vc_ban_user/vc_set_permission/
vc_move_user/vc_create_channel/vc_edit_channel/
vc_delete_channel/vc_create_account/vc_reset_password/
vc_delete_account. */
VC_EVENT_ACCOUNT_LIST = 15, /* Reply to vc_list_accounts. */
} vc_event_type;
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
/* TOFU server-identity classification (M4) — see VC_EVENT_SERVER_IDENTITY and
* vc_confirm_server_identity. Pins the TLS leaf certificate's own SHA-256 fingerprint
* (verifiable directly from the handshake), NOT the declared Ed25519
* server_identity_fingerprint from ServerHello the TLS cert and the server's Ed25519
* identity key are generated independently with no cryptographic binding between them today
* (docs/security.md §1.1), so pinning the self-declared value would be circular. The Ed25519
* fingerprint is still available for human-readable display via
* vc_get_server_identity_display(), it just isn't the value this gate accepts/rejects on. */
typedef enum vc_tofu_status {
VC_TOFU_FIRST_CONNECT = 0, /* no pin on file yet for this host:port */
VC_TOFU_MATCHED = 1, /* matches the previously pinned fingerprint */
VC_TOFU_MISMATCH = 2, /* DIFFERENT from the pinned fingerprint — possible MITM or a
legitimate server key rotation; warn loudly */
} vc_tofu_status;
/* ── 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;
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
/* M4, optional (added at the end — existing brace-initialized callers default this to
* NULL, no source change needed). Path to the TOFU pin file (see VC_EVENT_SERVER_IDENTITY/
* vc_confirm_server_identity). NULL = a built-in relative default
* ("./voicecat_tofu_pins.txt") so existing tests need no real persistence. A real app
* (e.g. the Windows client) should pass an explicit per-user path, e.g.
* "%AppData%\VoiceCat\tofu_pins.txt". */
const char* tofu_store_path;
} 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 */
int dred; /* bool — Deep REDundancy (Opus 1.6), off by default */
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
} vc_audio_config;
/* M5: permission bitset (mirrors protocol Permissions). */
typedef struct vc_permissions {
int can_create_temp_channel; /* bool */
int can_kick; /* bool */
int can_ban; /* bool */
int can_move_users; /* bool */
int can_admin_accounts; /* bool */
int is_admin; /* bool */
} vc_permissions;
/* M5: account entry (reply to vc_list_accounts / vc_get_account_list). */
typedef struct vc_account {
const char* username;
int is_admin; /* bool */
uint64_t created_at_unix_ms;
uint64_t last_login_unix_ms;
} vc_account;
typedef struct vc_account_list {
vc_account* items;
size_t count;
} vc_account_list;
/* M5: channel creation/edition descriptor. */
typedef struct vc_channel_info {
uint32_t id; /* 0 = new channel for create */
uint32_t parent_id; /* 0 = root */
const char* name;
const char* topic;
int password_protected; /* bool */
const char* password; /* nullable; ignored if password_protected == 0 */
uint32_t max_users; /* 0 = unlimited */
uint32_t sort_order;
/* Audio config — 0/NULL fields use server defaults. */
vc_audio_config audio;
} vc_channel_info;
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;
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
/* ── Channel / user / stream snapshots (M4 — for the channel-tree/user-list UI) ───────────
* Pull-based: re-call after VC_EVENT_CHANNEL_LIST / VC_EVENT_USER_JOINED / _LEFT / _UPDATED to
* refresh there is no push variant; those events just mean "go look". Same ownership
* contract as vc_device/vc_device_list above: core-allocated, caller frees with the matching
* vc_free_*, items' const char* fields are invalid after that call. */
typedef struct vc_channel {
uint32_t id;
uint32_t parent_id; /* 0 = root */
const char* name;
const char* topic;
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
int password_protected; /* bool */
uint32_t max_users; /* 0 = unlimited */
} vc_channel;
typedef struct vc_channel_list {
vc_channel* items;
size_t count;
} vc_channel_list;
typedef struct vc_user {
uint32_t id;
const char* nickname;
int is_guest; /* bool */
uint32_t channel_id;
int self_mic_muted; /* bool */
int self_deafened; /* bool */
int server_muted; /* bool */
int server_deafened; /* bool */
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
} vc_user;
typedef struct vc_user_list {
vc_user* items;
size_t count;
} vc_user_list;
/* Per-user stream summary — lighter than vc_audio_config; for the full effective Opus config
* of a specific (user_id, stream_id), use the existing vc_get_stream_audio_config. */
typedef struct vc_stream_summary {
uint32_t stream_id;
vc_stream_kind kind;
const char* label;
} vc_stream_summary;
typedef struct vc_stream_summary_list {
vc_stream_summary* items;
size_t count;
} vc_stream_summary_list;
/* Receive-side state the local listener has chosen for a specific remote stream — the
* counterpart to vc_set_remote_stream, so a UI can reopen its per-mix controls at the
* listener's actual current settings. All LOCAL (no protocol traffic) docs/voice.md §10.
* If (user_id, stream_id) is known but the listener has never called vc_set_remote_stream on
* it, the defaults are gain=1.0, muted=0, noise_reduction=0 (matching a fresh RemoteStream). */
typedef struct vc_remote_stream_state {
float gain; /* 0.0–… ; default 1.0 */
int muted; /* bool */
int noise_reduction; /* bool */
} vc_remote_stream_state;
/* 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 ─────────────────────────────────────────────────────────────── */
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
/* Result arrives as VC_EVENT_JOIN_RESULT, not a return value beyond "request queued". `password`
* is forwarded to the server's JoinChannelRequest.password for channels with
* vc_channel.password_protected set; NOTE (M4): no in-tree channel currently has a server-side
* password to check against channel creation/passwords are a future (M5+) feature, so this
* path is wired but not yet exercisable end-to-end. */
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);
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
/* VAD threshold: normalized RMS 0.01.0; default ~0.025. Takes effect immediately —
* recreates the VAD gate if a MIC stream is already active. No-op when mode != VOICE_ACTIVATION
* (value is remembered and applied if the mode switches back). */
VC_API vc_result vc_set_vad_threshold(vc_client* c, float threshold);
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);
feat(windows): UI overhaul -- toolbar, unified log, PM windows, channel counts, output volume - Voice actions (Join Voice, Share Screen Audio) moved to a ToolStrip toolbar and a new Voice menu in the menu bar; removed from the bottom voice panel - Activity log and chat log collapsed into a single RichTextBox (rtbLog); activity events appear in gray, chat messages in default color - Private messaging reworked: each conversation opens in its own modeless PrivateMessageForm instead of sharing the main chat log via a scope dropdown; cboScope removed; main compose bar always sends to the current channel - New "Messages -> New Private Message..." menu item (Ctrl+P) opens a UserPickerDialog listing all connected server users (not just the current channel) so you can PM anyone on the server - Channel tree now shows live user counts, e.g. "General (3)" -- counts sourced from the existing _users dictionary which already tracks all server users with channel IDs - Global output volume slider (TrackBar, 0-100, default 80) added to the right panel; wired to new vc_set_output_volume C ABI function that applies a master gain multiplier in the audio engine playback callback after mixing all streams - vc_set_output_volume added end-to-end: voicecat.h, audio_engine.h/.cpp, client.h/.cpp, voicecat.cpp, NativeMethods.cs, VoiceCatClient.cs - Documented Windows PowerShell ctest requirement in AGENTS.md and CLAUDE.md: MinGW binaries exit 0xc0000139 in Git Bash; always run ctest/.exe via PowerShell 22/22 ctest green (PowerShell); dotnet build 0 warnings. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-20 14:24:54 +02:00
/* Global playback volume applied after mixing all remote streams. gain 0.0 = silent,
* 1.0 = unity (default), >1.0 amplifies. Always LOCAL no protocol traffic. */
VC_API vc_result vc_set_output_volume(vc_client* c, float gain);
/* 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);
/* Reads back the receive-side state last set on (user_id, stream_id) via
* vc_set_remote_stream (or the defaults if never set). VC_ERR_INVALID_ARG if the user/stream
* isn't known. */
VC_API vc_result vc_get_remote_stream(vc_client* c, uint32_t user_id, uint32_t stream_id,
vc_remote_stream_state* out);
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);
feat(ios): audio overhaul, Join/Leave Voice, channel-id sync fix, stereo mic capture Three iOS client problems fixed plus a new core stereo-mic capture ABI: 1. Channel-id sync bug (mic button permanently dimmed): SessionState never synced currentChannelId from the self user's channelId on connect, so the mic button (gated on currentChannelId == 0) stayed dimmed. Added syncSelfChannel() (mirrors macOS MainWindowController.swift:461,491,522); called from init/.channelList/.userJoined/.userLeft/.userUpdated/.joinResult. Added applyServerMuteState() + serverMuted/serverDeafened to VoiceState. 2. Join/Leave Voice button: replaced icon-only mic toggle with explicit text button (parity with macOS). Mute/deafen disable when not in voice. 3. IOSAudioRouter.swift (new): full AVAudioSession routing layer — input port selection, built-in mic orientation/polar patterns, Bluetooth HFP/A2DP/Off modes, Standard/Raw mic processing, stereo capture, AirPlay, UserDefaults persistence. AudioSessionManager delegates to it. 4. Core stereo-mic capture (append-only ABI): vc_set_capture_channels() lets the core open the mic device in stereo (2-ch interleaved). LocalStream gains capture_channels; ensure_audio_running reads it; audio_engine.cpp capture_accum_ + on_capture updated to channel-aware accumulation. Test test_stereo_mic_capture (headless, L!=R stereo round-trip). Swift wrapper VoiceCatClient.setCaptureChannels. 5. Settings UI rework: AVAudioSession-derived input/output tree replaces miniaudio device picker. 6. iOS deployment target raised to 18.0 (Package.swift + project.pbxproj). swift-tools-version 6.0 with swiftLanguageModes .v5. Docs: tech-stack.md, architecture.md, voice.md, roadmap.md, building.md updated; stale 'vc_audio_suspend/resume deferred' claims corrected. Verified: ctest --preset dev 21/21 green; swift test 6/6 green; xcodebuild -target VoiceCatiOS -sdk iphonesimulator BUILD SUCCEEDED.
2026-06-19 13:17:52 +02:00
/* Set the capture channel count for a local MIC stream (1 = mono, 2 = stereo interleaved).
fix(ios): fix stereo mic + A2DP output silence Three coordinated fixes for the bug where enabling stereo mic capture caused all audio output (A2DP, speaker, wired) to go silent: 1. audio_engine.cpp — open playback before capture On iOS, starting the stereo capture AudioUnit can trigger an audio route reconfiguration that drops A2DP before the playback device has a chance to claim the route. Opening and starting the playback device first commits the output route (A2DP), so iOS is less likely to drop it when stereo capture activates afterward. 2. client.cpp — decouple set_capture_channels from engine restart Previously vc_set_capture_channels() stopped and restarted the engine immediately, which opened capture first (old ordering) and raced against the settling AVAudioSession route. Now it only stores the channel count; the caller (Swift via vc_audio_restart) controls when the engine restarts, after the route has settled. 3. IOSAudioRouter.swift — call audioRestart() after channel config selectCaptureChannels() and applyPreset() now call audioRestart() after applyConfiguration() + setCaptureChannels(). This is the vc_audio_restart() path that was added to the ABI in fdcc84f but never wired up in the Swift layer. The restart sees the stored channel count and reopens devices in the correct order (playback first, capture second). The doStartMicStream path is unaffected: setCaptureChannels is called before the server acknowledges the stream (engine not yet running), so ensure_audio_running() picks up capture_channels=2 directly when the stream is confirmed and opens with the right count from the start. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-19 17:39:23 +02:00
* Must be called after vc_stream_start. Stores the value; it takes effect on the next engine
* (re)start. Does NOT restart the engine itself the caller must follow up with
* vc_audio_restart() after AVAudioSession routing has settled (iOS) or after any platform
* audio-session reconfiguration. On iOS the Swift AVAudioSession routing layer enables stereo
* built-in mic capture by switching the built-in mic's data source to the .stereo polar
* pattern (setPreferredDataSource + setPreferredPolarPattern(.stereo) + setPreferredInput +
* setInputDataSource), calls this to record the desired channel count, and then calls
* vc_audio_restart() so the core reopens capture and playback against the new route.
* VC_ERR_INVALID_ARG if stream_id is unknown or channels is not 1 or 2. */
feat(ios): audio overhaul, Join/Leave Voice, channel-id sync fix, stereo mic capture Three iOS client problems fixed plus a new core stereo-mic capture ABI: 1. Channel-id sync bug (mic button permanently dimmed): SessionState never synced currentChannelId from the self user's channelId on connect, so the mic button (gated on currentChannelId == 0) stayed dimmed. Added syncSelfChannel() (mirrors macOS MainWindowController.swift:461,491,522); called from init/.channelList/.userJoined/.userLeft/.userUpdated/.joinResult. Added applyServerMuteState() + serverMuted/serverDeafened to VoiceState. 2. Join/Leave Voice button: replaced icon-only mic toggle with explicit text button (parity with macOS). Mute/deafen disable when not in voice. 3. IOSAudioRouter.swift (new): full AVAudioSession routing layer — input port selection, built-in mic orientation/polar patterns, Bluetooth HFP/A2DP/Off modes, Standard/Raw mic processing, stereo capture, AirPlay, UserDefaults persistence. AudioSessionManager delegates to it. 4. Core stereo-mic capture (append-only ABI): vc_set_capture_channels() lets the core open the mic device in stereo (2-ch interleaved). LocalStream gains capture_channels; ensure_audio_running reads it; audio_engine.cpp capture_accum_ + on_capture updated to channel-aware accumulation. Test test_stereo_mic_capture (headless, L!=R stereo round-trip). Swift wrapper VoiceCatClient.setCaptureChannels. 5. Settings UI rework: AVAudioSession-derived input/output tree replaces miniaudio device picker. 6. iOS deployment target raised to 18.0 (Package.swift + project.pbxproj). swift-tools-version 6.0 with swiftLanguageModes .v5. Docs: tech-stack.md, architecture.md, voice.md, roadmap.md, building.md updated; stale 'vc_audio_suspend/resume deferred' claims corrected. Verified: ctest --preset dev 21/21 green; swift test 6/6 green; xcodebuild -target VoiceCatiOS -sdk iphonesimulator BUILD SUCCEEDED.
2026-06-19 13:17:52 +02:00
VC_API vc_result vc_set_capture_channels(vc_client* c, uint32_t stream_id, uint32_t channels);
feat: external PCM feed/tap API (vc_stream_feed_pcm + vc_set_pcm_sink) Promotes vc_test_inject_capture (mono-only, TEST-ONLY) to a public, stereo-capable production API and adds a symmetric PCM tap on the receive side. Enables ReplayKit (iOS), ScreenCaptureKit (macOS), bots, soundboards, and custom clients — all without a hardware audio device. Core C++: - voicecat.h: new vc_stream_feed_pcm, vc_pcm_sink_cb typedef, vc_set_pcm_sink; vc_test_inject_capture kept as deprecated alias - audio_engine: stereo-aware inject_capture (channels param + ring reset on channel-count change); atomic pcm_sink_ fired per decoded frame in on_playback; RemoteStream carries user_id/stream_id for RT-safe sink metadata; init_recv_stream takes user_id+stream_id - client.cpp: stream_feed_pcm / set_pcm_sink implementations; sync_remote_streams passes user_id/stream_id to init_recv_stream - voicecat.cpp: trampolines + channels=1/2 validation Tests: test_external_pcm (headless, 3 sub-tests: mono round-trip, stereo feed L≠R, sink metadata+disable). ctest 23/23. Swift: feedPcm / setPcmSink in VoiceCatClient.swift + 4 XCTest smoke tests (ExternalPcmTests.swift). C#: StreamFeedPcm / SetPcmSink in VoiceCatClient.cs + NativeMethods.cs (vc_stream_feed_pcm unsafe P/Invoke, VcPcmSinkCallback delegate, vc_set_pcm_sink via nint) + 4 xUnit smoke tests (ExternalPcmTests.cs). Docs: architecture.md §4 new subsection, voice.md §9 updated (macOS/iOS now reference vc_stream_feed_pcm), protocol.md §8 explicit no-protocol-change note, roadmap.md M5 entry. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-20 17:52:09 +02:00
/* ── External PCM feed/tap ─────────────────────────────────────────────────── */
/* External PCM feed — production-grade API for driving a local stream's encode pipeline
* with caller-supplied PCM instead of (or in addition to) a hardware capture device. The
* stream must already be started (vc_stream_start). The core frames, encodes (Opus), seals
* (AEAD), and sends (UDP) the provided samples exactly as it would mic/loopback audio.
*
* samples_per_channel : samples per channel (e.g. 960 for 20 ms @ 48 kHz).
* channels : 1 (mono) or 2 (stereo interleaved L/R). VC_ERR_INVALID_ARG otherwise.
*
* Use cases: ReplayKit Broadcast Extension (iOS), ScreenCaptureKit (macOS), bots (TTS /
* music / relay), soundboards, DAW integration. Works for any stream kind (MIC /
* SCREEN_AUDIO / AUX_DEVICE). Thread-safe; may be called from any thread.
*
* Replaces vc_test_inject_capture (deprecated alias, see below). */
VC_API vc_result vc_stream_feed_pcm(vc_client* c, uint32_t stream_id,
const int16_t* pcm, size_t samples_per_channel,
uint32_t channels);
/* External PCM tap — receive decoded remote audio as int16 PCM per stream, before it is
* summed into the hardware mix. The callback fires on the audio playback thread once per
* decoded Opus frame (typically every 20 ms) for each active remote stream:
*
* cb(user, user_id, stream_id, pcm, samples_per_channel, channels, sample_rate)
*
* user_id / stream_id : identify the sender (same values as VC_EVENT_STREAM_STARTED).
* pcm : decoded int16 PCM, interleaved when channels == 2.
* samples_per_channel : samples per channel for this frame (typically 960 @ 48 kHz).
* channels : 1 or 2, matching the sender's stream configuration.
* sample_rate : always 48000 in the current implementation.
*
* Pass cb = NULL to disable (default: disabled; hardware playback only).
* The callback MUST NOT block, lock, or allocate copy what you need and return.
* PCM is still delivered to the hardware playback device regardless (dual output). */
typedef void (*vc_pcm_sink_cb)(void* user, uint32_t user_id, uint32_t stream_id,
const int16_t* pcm, size_t samples_per_channel,
uint32_t channels, uint32_t sample_rate);
VC_API vc_result vc_set_pcm_sink(vc_client* c, vc_pcm_sink_cb cb, void* user);
/* ── 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);
feat(M4): Windows WinForms client, TOFU identity pinning, VAD threshold + always-on mode Core ABI extensions (voicecat.h): - vc_list_channels / vc_list_users / vc_list_user_streams — pull-based snapshot getters for the channel-tree and user-list UI; session_model_mu_ guards cross-thread reads - VC_EVENT_JOIN_RESULT / vc_join_channel — channel join with optional password - VC_EVENT_SERVER_IDENTITY + vc_confirm_server_identity — TOFU gate that blocks io_thread_ until the UI approves or rejects; pins TLS leaf-cert SHA-256 (not declared Ed25519) - vc_get_server_identity_display — Ed25519 fingerprint for human-readable display only - VC_INPUT_ALWAYS_ON = 2 in vc_input_mode — transmit unconditionally, no VAD gate - vc_set_vad_threshold — live RMS threshold update (0.0–1.0); EnergyVadProcessor stores it atomically so the audio RT path reads without a lock C++ implementation: - SessionModel::apply_snapshot / apply_channel_event fixed to populate parent_id, password_protected, and max_users (were permanently zeroed) - TlsContext::peer_cert_fingerprint — SHA-256 of peer leaf cert DER via mbedTLS - TofuStore split into peek (read-only) + pin (write) so first-connect only persists after user approval; tofu_store_path in vc_config for per-user pin file location - TcpAcceptor uses dual-stack IPv6+IPv4 fallback (fixes localhost → ::1 on Windows) - windows-client CMake preset: Release shared DLL, static MinGW runtime, no tools/tests - New C++ tests: test_channel_user_list_abi, test_tofu_flow (14/14 green) Windows client (clients/windows/ — .NET 10 WinForms): - VoiceCat.Interop: LibraryImport P/Invoke surface, UnmanagedCallersOnly callbacks, Channel<VoiceCatEvent> event delivery drained by 30ms WinForms Timer - VoiceCat.App: ConnectDialog (saved servers, DPAPI password storage), ServerIdentity- Dialog (TOFU first-connect / mismatch warning), MainForm (channel TreeView, user ListBox, RichTextBox chat, voice controls, device pickers, VAD/PTT/always-on mode, per-user gain/mute/NR tuning, VAD sensitivity TrackBar, level meter ProgressBar) - PttKeyCaptureDialog — focus-scoped PTT key capture (documented limitation) - PerUserTuningDialog — real-time gain/mute/NR applied to all of a user's streams - Accessibility: explicit AccessibleName/Description on every control, & mnemonics, Activity log ListBox as durable screen-reader record, AutomationNotification for curated live announcements Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-17 00:35:16 +02:00
/* ── Channel / user / stream enumeration (M4; mirrors vc_list_devices above) ─────────────── */
VC_API vc_result vc_list_channels(vc_client* c, vc_channel_list* out);
VC_API void vc_free_channel_list(vc_channel_list* list);
VC_API vc_result vc_list_users(vc_client* c, vc_user_list* out);
VC_API void vc_free_user_list(vc_user_list* list);
/* Streams currently owned by user_id (their mic/screen-audio/aux), per the last snapshot/
* event. VC_ERR_INVALID_ARG if user_id is unknown. */
VC_API vc_result vc_list_user_streams(vc_client* c, uint32_t user_id,
vc_stream_summary_list* out);
VC_API void vc_free_stream_summary_list(vc_stream_summary_list* list);
/* ── TOFU server-identity confirmation (M4) — see VC_EVENT_SERVER_IDENTITY/vc_tofu_status ── */
/* Accept or reject the pending server-identity check for the in-progress connect(). Must be
* called after a VC_EVENT_SERVER_IDENTITY event; the io_thread_ holds the connection open
* (ClientHello/auth deferred) until this is called, up to a generous internal timeout (after
* which it's treated as a reject). accept=0 aborts the connection (emits
* VC_EVENT_DISCONNECTED, result=VC_ERR_CRYPTO) and does NOT update the pin file. accept=1 on
* FIRST_CONNECT/MISMATCH updates the pin file to the new fingerprint and proceeds; accept=1 on
* MATCHED is a no-op confirmation (always safe) and proceeds. VC_ERR_INVALID_ARG if no
* identity confirmation is currently pending. */
VC_API vc_result vc_confirm_server_identity(vc_client* c, int accept /* bool */);
/* The Ed25519 identity fingerprint from ServerHello, hex-formatted for display (e.g. "this
* server also identifies as <hex>"). Purely informational — NOT the value
* vc_confirm_server_identity gates on (see vc_tofu_status's doc comment). Empty string if not
* yet available. Pass out_buf=NULL to query the required buffer size via *out_len first;
* otherwise out_buf must be >= *out_len + 1 bytes (NUL-terminated UTF-8/ASCII hex). */
VC_API vc_result vc_get_server_identity_display(vc_client* c, char* out_buf, size_t buf_cap,
size_t* out_len);
/* ── M5: Moderation & admin ─────────────────────────────────────────────────
* All calls are async; the result arrives as VC_EVENT_GENERIC_RESULT (or
* VC_EVENT_ACCOUNT_LIST for vc_list_accounts). They require VC_STATE_CONNECTED and,
* on the server side, the appropriate permission. */
VC_API vc_result vc_kick_user(vc_client* c, uint32_t user_id, const char* reason);
VC_API vc_result vc_ban_user(vc_client* c, uint32_t user_id, const char* reason,
uint64_t expires_unix_ms);
VC_API vc_result vc_set_permission(vc_client* c, uint32_t user_id,
const vc_permissions* perms);
VC_API vc_result vc_set_server_mute(vc_client* c, uint32_t user_id, int muted, int deafened);
VC_API vc_result vc_move_user(vc_client* c, uint32_t user_id, uint32_t channel_id);
VC_API vc_result vc_create_channel(vc_client* c, const vc_channel_info* info);
VC_API vc_result vc_edit_channel(vc_client* c, const vc_channel_info* info);
VC_API vc_result vc_delete_channel(vc_client* c, uint32_t channel_id);
VC_API vc_result vc_create_account(vc_client* c, const char* username, const char* password);
VC_API vc_result vc_reset_password(vc_client* c, const char* username,
const char* new_password);
VC_API vc_result vc_delete_account(vc_client* c, const char* username);
VC_API vc_result vc_list_accounts(vc_client* c);
/* Pull the last received account list (populated when VC_EVENT_ACCOUNT_LIST fires).
* Caller must free the list with vc_free_account_list. */
VC_API vc_result vc_get_account_list(vc_client* c, vc_account_list* out);
VC_API void vc_free_account_list(vc_account_list* list);
/* Pull the caller's own permissions (from the last AuthResult). */
VC_API vc_result vc_get_permissions(vc_client* c, vc_permissions* out);
/* AVAudioSession interruption hooks (iOS M6). Pause/resume miniaudio device I/O for
* AVAudioSession interruptions (phone call, Siri, etc.) and backgrounding. Call
* vc_audio_suspend() when an interruption begins; call vc_audio_resume() after the
* session is re-activated. No-op if the audio engine is not running. */
VC_API vc_result vc_audio_suspend(vc_client* c);
VC_API vc_result vc_audio_resume(vc_client* c);
fix(ios): stereo mic + A2DP output, add vc_audio_restart ABI Diagnosed by comparing against TeamTalk5 (Client/iTeamTalk), which achieves stereo mic + A2DP output. Five fixes: 1. configureStereoCapture now calls setPreferredInput + setInputDataSource (mirroring TeamTalk5's SoundDevicesModel). Previously omitted based on incorrect diagnosis that setPreferredInput collapsed A2DP — the real culprit was setPreferredInputNumberOfChannels(2), which neither project uses. 2. New C ABI: vc_audio_restart (full stop + re-init, unlike suspend/resume which only stop/start). Swift wrapper added. The withAudioSuspend wrapper that used it was removed after on-device testing showed it killed all audio (including VoiceOver) when switching presets — the core's set_capture_channels handles engine restart internally. 3. Bluetooth options: Voice Chat preset now includes BOTH .allowBluetoothHFP AND .allowBluetoothA2DP (matching TeamTalk5's UtilSound.swift:228). Previously HFP-only blocked A2DP headphones. 4. Capture channels now reset when switching stereo→mono via selectCaptureChannels/applyPreset. AudioSessionManager tracks activeMicStreamId (set by SessionState on join/leave voice). 5. Docs synced: voice.md, tech-stack.md, architecture.md, PROGRESS.md. Removed stale setPreferredInputNumberOfChannels(2) references. Verified: ctest --preset dev 21/21 green, iOS client builds. Stereo mic + A2DP output still needs on-device debugging — the core recipe is correct but iOS 26 route behavior requires hands-on testing with a debugger.
2026-06-19 16:58:21 +02:00
/* Full audio engine restart (iOS M6). Unlike vc_audio_suspend/resume which only stop/start
* the existing miniaudio devices (leaving them bound to the route that was active when they
* were opened), vc_audio_restart() uninitializes and re-initializes the capture and playback
* devices so they pick up a new AVAudioSession route. Call this from the Swift layer AFTER
* reconfiguring AVAudioSession (setCategory, setPreferredInput, setPreferredPolarPattern, etc.)
* so the core's devices reopen against the new route.
fix(ios): stereo mic + A2DP output, add vc_audio_restart ABI Diagnosed by comparing against TeamTalk5 (Client/iTeamTalk), which achieves stereo mic + A2DP output. Five fixes: 1. configureStereoCapture now calls setPreferredInput + setInputDataSource (mirroring TeamTalk5's SoundDevicesModel). Previously omitted based on incorrect diagnosis that setPreferredInput collapsed A2DP — the real culprit was setPreferredInputNumberOfChannels(2), which neither project uses. 2. New C ABI: vc_audio_restart (full stop + re-init, unlike suspend/resume which only stop/start). Swift wrapper added. The withAudioSuspend wrapper that used it was removed after on-device testing showed it killed all audio (including VoiceOver) when switching presets — the core's set_capture_channels handles engine restart internally. 3. Bluetooth options: Voice Chat preset now includes BOTH .allowBluetoothHFP AND .allowBluetoothA2DP (matching TeamTalk5's UtilSound.swift:228). Previously HFP-only blocked A2DP headphones. 4. Capture channels now reset when switching stereo→mono via selectCaptureChannels/applyPreset. AudioSessionManager tracks activeMicStreamId (set by SessionState on join/leave voice). 5. Docs synced: voice.md, tech-stack.md, architecture.md, PROGRESS.md. Removed stale setPreferredInputNumberOfChannels(2) references. Verified: ctest --preset dev 21/21 green, iOS client builds. Stereo mic + A2DP output still needs on-device debugging — the core recipe is correct but iOS 26 route behavior requires hands-on testing with a debugger.
2026-06-19 16:58:21 +02:00
* Safe to call when the engine is not running (it will just start it). */
VC_API vc_result vc_audio_restart(vc_client* c);
#if defined(__cplusplus)
} /* extern "C" */
#endif
#endif /* VOICECAT_H */