Files
voice-cat/server/src/session_registry.h

224 lines
9.8 KiB
C
Raw Normal View History

2026-06-15 23:48:44 +02:00
/*
* server/session_registry.h In-memory session, channel, and user registry.
*
* Tracks all authenticated sessions, the channel tree, user<>channel assignments,
* UDP endpoint bindings (M2), and SSRC<>session mappings (M2).
2026-06-15 23:48:44 +02:00
* Protected by a shared_mutex (many readers, few writers). All methods are thread-safe.
*/
#ifndef VOICECAT_SERVER_SESSION_REGISTRY_H
#define VOICECAT_SERVER_SESSION_REGISTRY_H
#ifdef VOICECAT_HAS_NET
#include <array>
#include <atomic>
2026-06-15 23:48:44 +02:00
#include <cstdint>
#include <memory>
#include <optional>
2026-06-15 23:48:44 +02:00
#include <shared_mutex>
#include <string>
#include <unordered_map>
#include <vector>
#define ASIO_STANDALONE 1
#include <asio.hpp>
#include "db.h"
2026-06-15 23:48:44 +02:00
#include "proto/voicecat.pb.h"
namespace voicecat::server {
class ConnSession;
struct ChannelEntry {
voicecat::v1::Channel proto;
};
struct UserEntry {
voicecat::v1::User proto;
uint64_t session_id{};
};
// Hashes asio::ip::udp::endpoint by "addr:port" string.
struct UdpEndpointHash {
size_t operator()(const asio::ip::udp::endpoint& ep) const {
std::string key = ep.address().to_string() + ':' + std::to_string(ep.port());
return std::hash<std::string>{}(key);
}
};
2026-06-15 23:48:44 +02:00
class SessionRegistry {
public:
explicit SessionRegistry(std::shared_ptr<Database> db);
2026-06-15 23:48:44 +02:00
// Load channels from the database, seeding defaults on first run.
void load_channels();
2026-06-15 23:48:44 +02:00
// Register a session (before auth). Returns the assigned session_id.
uint64_t register_session(std::weak_ptr<ConnSession> session);
// Remove a session (called on disconnect).
void unregister_session(uint64_t session_id);
// Add a user once authenticated. Returns the assigned user_id.
uint32_t add_user(uint64_t session_id, const voicecat::v1::User& user);
// Remove a user (called on disconnect after auth).
void remove_user(uint32_t user_id);
fix(net): broadcast LEFT on disconnect, add keepalive/reaper, cap PLC hiss Three reported bugs traced to one root cause plus two missing designed features: 1. Stale users + eternal PLC hiss (root cause): ConnSession::close() silently erased dropped users without broadcasting UserEvent::LEFT, so peers never learned the user left and their audio engines never called remove_stream — Opus PLC synthesized comfort noise forever. Fix: broadcast_left() helper + close() broadcasts LEFT before erasing. 2. PLC cap (defense-in-depth): on_playback now caps pure PLC at ~2s, then emits digital silence so a stale stream can never hiss forever even if remove_stream is skipped. Resets automatically on fresh packets. 3. No timeout / no ping: client never sent Ping, server had no last_seen / reaper, so half-open connections (NAT timeout, wifi loss, sleep) left ghost users forever. Fix: client Ping every 15s with RTT measurement, ConnSession::last_seen bumped on every inbound TCP/UDP frame, steady_timer reaper sweeps every 15s and drops sessions older than 45s (configurable via server::Config). 4. UDP KEEPALIVE: client sends plaintext kFrameKeepalive every 5s; server bumps last_seen + echoes back. Keeps NAT bindings alive and lets media activity defer the reaper independently of TCP. 5. Graceful client disconnect: vc_disconnect() sends Disconnect{code=0} via a flag-based io-thread exit (no double-close race); server handles client-sent Disconnect with immediate close() + LEFT broadcast. 3 new tests: disconnect_left, plc_cap, reaper_timeout. 21/21 ctest green. Docs: protocol.md §6/§7, voice.md §6, architecture.md §5, PROGRESS.md.
2026-06-18 01:18:33 +02:00
// Broadcast a UserEvent::LEFT for a user to all other sessions. Called by
// ConnSession::close() before remove_user() so remaining clients learn about an
// ungraceful disconnect (TCP drop, crash, network loss). Mirrors the first half of
// kick_user(). Takes the shared lock internally; safe to call from ConnSession::close.
void broadcast_left(uint32_t user_id, const std::string& reason);
2026-06-15 23:48:44 +02:00
// Move a user to a channel. Returns false if channel doesn't exist.
bool set_user_channel(uint32_t user_id, uint32_t channel_id);
// Snapshot for ServerStateSnapshot message.
std::vector<voicecat::v1::Channel> channel_snapshot() const;
std::vector<voicecat::v1::User> user_snapshot() const;
std::optional<voicecat::v1::User> user_snapshot_user(uint32_t user_id) const;
std::optional<std::string> user_nickname(uint32_t user_id) const;
2026-06-15 23:48:44 +02:00
// Resolve target sessions for a text message relay.
std::vector<std::shared_ptr<ConnSession>> resolve_text_targets(
uint64_t sender_session_id, voicecat::v1::TextScope scope, uint32_t target_id) const;
// Broadcast an envelope to all sessions except the excluded one.
void broadcast(const voicecat::v1::Envelope& env, uint64_t exclude_session_id = 0) const;
fix(net): broadcast LEFT on disconnect, add keepalive/reaper, cap PLC hiss Three reported bugs traced to one root cause plus two missing designed features: 1. Stale users + eternal PLC hiss (root cause): ConnSession::close() silently erased dropped users without broadcasting UserEvent::LEFT, so peers never learned the user left and their audio engines never called remove_stream — Opus PLC synthesized comfort noise forever. Fix: broadcast_left() helper + close() broadcasts LEFT before erasing. 2. PLC cap (defense-in-depth): on_playback now caps pure PLC at ~2s, then emits digital silence so a stale stream can never hiss forever even if remove_stream is skipped. Resets automatically on fresh packets. 3. No timeout / no ping: client never sent Ping, server had no last_seen / reaper, so half-open connections (NAT timeout, wifi loss, sleep) left ghost users forever. Fix: client Ping every 15s with RTT measurement, ConnSession::last_seen bumped on every inbound TCP/UDP frame, steady_timer reaper sweeps every 15s and drops sessions older than 45s (configurable via server::Config). 4. UDP KEEPALIVE: client sends plaintext kFrameKeepalive every 5s; server bumps last_seen + echoes back. Keeps NAT bindings alive and lets media activity defer the reaper independently of TCP. 5. Graceful client disconnect: vc_disconnect() sends Disconnect{code=0} via a flag-based io-thread exit (no double-close race); server handles client-sent Disconnect with immediate close() + LEFT broadcast. 3 new tests: disconnect_left, plc_cap, reaper_timeout. 21/21 ctest green. Docs: protocol.md §6/§7, voice.md §6, architecture.md §5, PROGRESS.md.
2026-06-18 01:18:33 +02:00
// Return all sessions whose last_seen is older than max_age_ms (steady_clock ms), i.e.
// have not had any inbound TCP or UDP activity in that span. The reaper (server.cpp)
// calls close() on each — which broadcasts UserEvent::LEFT via the Tier 1 fix. Locks
// only to collect the list; close() runs outside the lock (mirrors kick_user's pattern).
std::vector<std::shared_ptr<ConnSession>> find_stale_sessions(int64_t max_age_ms) const;
private:
void broadcast_unlocked(const voicecat::v1::Envelope& env,
uint64_t exclude_session_id = 0) const;
public:
// ── Permissions ────────────────────────────────────────────────────────────
void set_session_permissions(uint64_t session_id,
const voicecat::v1::Permissions& perms);
std::optional<voicecat::v1::Permissions> get_session_permissions(
uint64_t session_id) const;
// ── Moderation ─────────────────────────────────────────────────────────────
// Find a live session by its user_id. Returns nullptr if offline.
std::shared_ptr<ConnSession> find_session_by_user_id(uint32_t user_id) const;
// Forcibly disconnect a user with a reason. Broadcasts UserEvent::LEFT.
// Returns true if the user was online.
bool kick_user(uint32_t user_id, const std::string& reason);
// Kick a user and insert a persistent ban. Returns true if the user was online.
bool ban_user(uint32_t user_id, const std::string& reason, int64_t expires_at);
// Set server-mute/deafen flags on a user and broadcast the update.
bool set_server_mute(uint32_t user_id, bool muted, bool deafened);
// Move a user to a channel (permission-checked by caller).
bool move_user(uint32_t user_id, uint32_t channel_id);
// ── Channel CRUD ───────────────────────────────────────────────────────────
// Create a channel. Returns the new channel id, or 0 on error.
uint32_t create_channel(const voicecat::v1::Channel& ch, const std::string& password,
std::string& error);
// Update a channel. Returns false on error.
bool update_channel(const voicecat::v1::Channel& ch, const std::string& password,
std::string& error);
// Delete a channel. Remaining users are moved to Lobby (id=1). Returns false on error.
bool delete_channel(uint32_t channel_id, std::string& error);
// Return a channel proto by id, or nullopt.
std::optional<voicecat::v1::Channel> get_channel(uint32_t channel_id) const;
// Check a channel password.
bool check_channel_password(uint32_t channel_id, const std::string& password) const;
// ── M2: UDP / media ────────────────────────────────────────────────────────
// Register a session's UDP token (called at auth success).
void register_udp_token(const std::array<uint8_t, 16>& token, uint64_t session_id);
// Locate a session by its UDP binding token (called by MediaRelay on UDP_BINDING).
std::shared_ptr<ConnSession> find_by_udp_token(const std::array<uint8_t, 16>& token) const;
// Associate a UDP endpoint with a session (called by MediaRelay after token verification).
void register_udp_endpoint(asio::ip::udp::endpoint ep, uint64_t session_id);
// Locate the session that owns a UDP sender endpoint (called per incoming voice packet).
std::shared_ptr<ConnSession> find_by_udp_endpoint(const asio::ip::udp::endpoint& ep) const;
// Assign an SSRC for a new stream. Returns the assigned SSRC.
uint32_t assign_ssrc(uint64_t session_id);
// Add/replace a stream entry on a user (called when StreamAnnounce succeeds).
// Returns the updated User proto for broadcasting, or nullopt if user not found.
std::optional<voicecat::v1::User> set_user_stream(uint32_t user_id,
const voicecat::v1::StreamInfo& info);
// Remove a stream entry from a user (called on StreamStop). Returns the updated
// User proto for broadcasting, or nullopt if user not found.
std::optional<voicecat::v1::User> clear_user_stream(uint32_t user_id, uint32_t stream_id);
// Get all sessions in a channel except the one excluded (for SFU relay).
std::vector<std::shared_ptr<ConnSession>> find_channel_sessions(
uint32_t channel_id, uint64_t exclude_session_id = 0) const;
// Return the channel_id of a user (0 if not found).
uint32_t user_channel(uint32_t user_id) const;
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
// Return a channel's authoritative AudioConfig (M3 per-channel Opus tuning), or nullopt
// if the channel doesn't exist. There is no per-id Channel getter today otherwise —
// channel_snapshot() copies every channel, which callers needing just one config should
// avoid.
std::optional<voicecat::v1::AudioConfig> channel_audio_config(uint32_t channel_id) const;
2026-06-15 23:48:44 +02:00
private:
void seed_default_channels();
2026-06-15 23:48:44 +02:00
mutable std::shared_mutex mu_;
std::shared_ptr<Database> db_;
2026-06-15 23:48:44 +02:00
uint64_t next_session_id_{1};
uint32_t next_user_id_{1};
uint32_t next_channel_id_{3}; // 1 and 2 are reserved for Lobby, Music Room
2026-06-15 23:48:44 +02:00
std::unordered_map<uint64_t, std::weak_ptr<ConnSession>> sessions_;
std::unordered_map<uint32_t, UserEntry> users_;
std::unordered_map<uint32_t, ChannelEntry> channels_;
std::unordered_map<uint64_t, voicecat::v1::Permissions> session_permissions_;
// M2: token → session_id (populated at auth, cleared on disconnect)
struct TokenHash {
size_t operator()(const std::array<uint8_t, 16>& t) const {
// FNV-1a over 16 bytes
size_t h = 14695981039346656037ULL;
for (auto b : t) { h ^= b; h *= 1099511628211ULL; }
return h;
}
};
std::unordered_map<std::array<uint8_t, 16>, uint64_t, TokenHash> udp_tokens_;
// M2: UDP endpoint → session_id (populated after UDP binding packet arrives)
std::unordered_map<asio::ip::udp::endpoint, uint64_t, UdpEndpointHash> udp_endpoints_;
// M2: ssrc → session_id (populated when StreamAnnounce is processed)
std::unordered_map<uint32_t, uint64_t> ssrc_to_session_;
std::atomic<uint32_t> next_ssrc_{1};
2026-06-15 23:48:44 +02:00
};
} // namespace voicecat::server
#endif // VOICECAT_SERVER_SESSION_REGISTRY_H
2026-06-15 23:48:44 +02:00
#endif // VOICECAT_SERVER_SESSION_REGISTRY_H