Files
EchoChat/backend/go-service/app/ws/handler.go
bujinyuan 3b83c79036 feat(phase2e-2): 落地会议生命周期状态机 + Router 幂等双层防御(Task 8)
核心交付:
- 新建 MeetingLifecycleService(6 钩子 + sync.Map 本地 timer + Redis key 双保险 + RescheduleFromRedis)
- 新建 MeetingCleanupTask(启动重建 timer + 每 N 秒扫 host_grace/empty_ttl 兜底 + 4h stale active 回收)
- MediaOrchestrator 新增 ResolveRouterID;HTTPMediaOrchestrator.CreateRouter 入口 sync.Map 幂等防御
- 业务层 JoinRoom 移除 CreateRouter 调用改走 CancelEmptyTTL + ResolveRouterID;LeaveRoom 空房分支改调 OnAllMembersLeft 不再立即销毁
- MeetingSignalService 新增 OnWSDisconnect 实现 ws.MeetingDisconnectHook;OnRoomJoin 追加 host 重连钩子
- ws.handler 定义 MeetingDisconnectHook 接口 + SetMeetingDisconnectHook,解耦 ws→meeting 反向依赖
- config 新增 MeetingConfig{HostGrace=120, EmptyRoomTTL=300, CleanupInterval=30, StaleRoomHours=4}

关键设计决策:
- Redis key TTL = 业务时长 + max(CleanupIntervalSeconds*2, 30s) buffer:避免本地 timer 与
  Redis 自动过期同步到期导致 DEL 返回 0 被误判为"已被其他路径处理"而跳过业务逻辑
- Router 幂等双层防御(决策 q2_router_dedup=a2_both):业务层不重复调 + HTTP 层 sync.Map 命中直接返回
- 普通成员 WS 断开仅清 media 资源不动 participant 表(决策 q1_nonhost_disconnect=a1_keep_current)

E2E 验证:docs/verify/meeting_t8_verify.mjs PASS=20 FAIL=0,覆盖 5 场景:
- S1 host 宽限期过期自动转让(meeting.host.changed + DB host_id 更新)
- S2 宽限期内重连保留身份
- S3 empty_ttl 期内新成员加入复活房间
- S4 empty_ttl 过期 → 房间 Ended + 新 join 被拒
- S5 CreateRoom +1 Router / JoinRoom 不再创建新 Router(通过 media-server /internal/info stats.routers 断言)

media-server:/internal/info 响应追加 stats.routers + routers[] 供 E2E 断言 Router 幂等

文档同步:
- docs/progress/CURRENT_STATUS.md 头部 + 新增 Task 8 交付条目
- docs/plans/2026-04-21-phase2e-2-implementation.plan.md Task 8 标记完成 + 实际产出/决策/验证
- docs/api/frontend/meeting.md 补充 host.changed.auto_reason / room.ended.reason=system_error / 空房 TTL 复活语义 + Task 8 验证记录
- docs/architecture/system-architecture.md meeting 模块职责补充"会议生命周期状态机"
- .cursor/rules/project-context.mdc 追加 Task 8 条目并更新 Phase 2e-2 进度(Task 0-8 )

Made-with: Cursor
2026-04-21 18:21:04 +08:00

215 lines
7.2 KiB
Go
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

// Package ws 提供 WebSocket 连接处理
// 负责 HTTP → WebSocket 升级、JWT 认证、消息路由分发
package ws
import (
"context"
"net/http"
"github.com/echochat/backend/config"
"github.com/echochat/backend/pkg/logs"
"github.com/echochat/backend/pkg/utils"
"github.com/echochat/backend/pkg/ws"
"github.com/gin-gonic/gin"
"github.com/gorilla/websocket"
"go.uber.org/zap"
)
// TokenValidator 有状态 JWT 验证接口(检查 Token 是否在 Redis 中有效)
// 由 auth.AuthService 实现,用于防止已登出用户建立 WebSocket 连接
type TokenValidator interface {
ValidateAccessToken(ctx context.Context, userID int64, clientType, token string) bool
}
// OfflineMessagePusher 离线消息推送接口
// 由 im.handler.OfflinePusher 实现WebSocket 连接建立后触发推送
type OfflineMessagePusher interface {
PushOfflineMessages(ctx context.Context, userID int64)
}
// NotifyConnectHook 通知未读补偿钩子接口
// 由 notify.service.NotifyService 隐式实现
// WebSocket 连接建立后触发,向客户端推送 notify.unread.total 事件
// 用于断线重连场景下的徽标状态同步
type NotifyConnectHook interface {
PushUnreadTotalOnConnect(ctx context.Context, userID int64)
}
// MeetingDisconnectHook 会议 WS 断线钩子接口Phase 2e-2 Task 8
// 由 meeting.service.MeetingSignalService 隐式实现
// 触发时机:用户最后一条 WS 连接被移除(最终下线)
// 职责:清理该用户在会议中的媒体资源;若为 host 则启动 host 宽限期
// 抽象为接口避免 ws 包反向依赖 meeting 包引发循环引用
type MeetingDisconnectHook interface {
OnWSDisconnect(ctx context.Context, userID int64)
}
var upgrader = websocket.Upgrader{
ReadBufferSize: 1024,
WriteBufferSize: 1024,
CheckOrigin: func(r *http.Request) bool {
return true // TODO: 生产环境通过配置限制 allowed origins
},
}
// Handler WebSocket 连接处理器
type Handler struct {
hub *ws.Hub
pubsub *ws.PubSub
jwtCfg *config.JWTConfig
onlineService *OnlineService
tokenValidator TokenValidator
offlinePusher OfflineMessagePusher
notifyConnectHook NotifyConnectHook
meetingDisconnectHook MeetingDisconnectHook // Task 8 注入
}
// NewHandler 创建 WebSocket Handler 实例
func NewHandler(hub *ws.Hub, pubsub *ws.PubSub, jwtCfg *config.JWTConfig, onlineService *OnlineService, tokenValidator TokenValidator) *Handler {
return &Handler{
hub: hub,
pubsub: pubsub,
jwtCfg: jwtCfg,
onlineService: onlineService,
tokenValidator: tokenValidator,
}
}
// SetOfflinePusher 设置离线消息推送器(由 IM 模块在初始化时注入)
func (h *Handler) SetOfflinePusher(pusher OfflineMessagePusher) {
h.offlinePusher = pusher
}
// SetNotifyConnectHook 设置通知未读补偿钩子(由 notify 模块在初始化时注入)
func (h *Handler) SetNotifyConnectHook(hook NotifyConnectHook) {
h.notifyConnectHook = hook
}
// SetMeetingDisconnectHook 设置会议 WS 断线钩子(由 meeting 模块在初始化时注入)
// Phase 2e-2 Task 8WS 最终下线时触发 host 宽限期 / 媒体资源清理
func (h *Handler) SetMeetingDisconnectHook(hook MeetingDisconnectHook) {
h.meetingDisconnectHook = hook
}
// Upgrade 处理 WebSocket 升级请求
// GET /ws?token=xxx → JWT 认证 → 升级连接 → 注册 Hub → 订阅 Redis 频道
func (h *Handler) Upgrade(c *gin.Context) {
funcName := "ws.handler.Upgrade"
token := c.Query("token")
if token == "" {
utils.ResponseUnauthorized(c, "缺少认证 Token")
return
}
claims, err := utils.ParseToken(h.jwtCfg, token)
if err != nil {
logs.Warn(nil, funcName, "WebSocket Token 验证失败", zap.Error(err))
utils.ResponseUnauthorized(c, "Token 无效或已过期")
return
}
clientType := claims.ClientType
if clientType == "" {
clientType = "frontend"
}
if h.tokenValidator != nil && !h.tokenValidator.ValidateAccessToken(c.Request.Context(), claims.UserID, clientType, token) {
logs.Warn(nil, funcName, "WebSocket Token 已失效Redis 校验)",
zap.Int64("user_id", claims.UserID))
utils.ResponseUnauthorized(c, "认证已失效,请重新登录")
return
}
conn, err := upgrader.Upgrade(c.Writer, c.Request, nil)
if err != nil {
logs.Error(nil, funcName, "WebSocket 升级失败",
zap.Int64("user_id", claims.UserID), zap.Error(err))
return
}
client := ws.NewClient(h.hub, conn, claims.UserID)
client.SetOnDisconnect(func(userID int64) {
if client.IsClosedByHub() && h.hub.IsOnline(userID) {
logs.Info(nil, "ws.handler.onDisconnect", "连接被新连接替换,跳过下线清理",
zap.Int64("user_id", userID))
return
}
h.pubsub.Unsubscribe(userID)
h.onlineService.UserOffline(context.Background(), userID)
// Task 8通知会议模块处理 host 宽限期 / 媒体资源清理
// 钩子为可选(测试或 meeting 模块未加载场景安全跳过)
if h.meetingDisconnectHook != nil {
go h.meetingDisconnectHook.OnWSDisconnect(context.Background(), userID)
}
})
h.hub.Register(client)
h.pubsub.Subscribe(claims.UserID)
h.onlineService.UserOnline(c.Request.Context(), claims.UserID, c.ClientIP())
logs.Info(nil, funcName, "WebSocket 连接建立",
zap.Int64("user_id", claims.UserID),
zap.String("ip", c.ClientIP()))
go client.WritePump()
go client.ReadPump(h.createReadHandler(claims.UserID))
if h.offlinePusher != nil {
go h.offlinePusher.PushOfflineMessages(context.Background(), claims.UserID)
}
if h.notifyConnectHook != nil {
go h.notifyConnectHook.PushUnreadTotalOnConnect(context.Background(), claims.UserID)
}
}
// createReadHandler 创建带生命周期管理的消息处理函数
// 优先查 Hub 事件路由表(业务模块注册的处理器),未命中再走内置 fallback
func (h *Handler) createReadHandler(userID int64) ws.MessageHandler {
return func(client *ws.Client, msg *ws.Message) {
funcName := "ws.handler.onMessage"
logs.Debug(nil, funcName, "收到 WebSocket 消息",
zap.Int64("user_id", client.UserID),
zap.String("event", msg.Event),
zap.Int64("seq", msg.Seq))
// 优先查事件路由表IM、Meeting 等模块注册的处理器)
if h.hub.DispatchEvent(client, msg) {
return
}
// 内置事件 fallback
switch msg.Event {
case "heartbeat":
h.onlineService.HeartbeatRenew(context.Background(), userID)
resp := ws.NewResponse(msg.Event, msg.Seq, 0, "pong", nil)
data, err := ws.MarshalResponse(resp)
if err != nil {
logs.Error(nil, funcName, "序列化心跳响应失败", zap.Error(err))
return
}
client.Send(data)
default:
logs.Warn(nil, funcName, "未知事件类型",
zap.String("event", msg.Event),
zap.Int64("user_id", client.UserID))
resp := ws.NewResponse(msg.Event, msg.Seq, -1, "未知事件", nil)
data, err := ws.MarshalResponse(resp)
if err != nil {
logs.Error(nil, funcName, "序列化响应失败", zap.Error(err))
return
}
client.Send(data)
}
}
}
// GetHub 返回 Hub 实例(供在线状态等模块访问)
func (h *Handler) GetHub() *ws.Hub {
return h.hub
}
// GetPubSub 返回 PubSub 实例(供业务模块发送推送)
func (h *Handler) GetPubSub() *ws.PubSub {
return h.pubsub
}