Agent-System/cmd/server/main.go

package main

import (
	"context"
	"flag"
	"fmt"
	"net/http"
	"os"
	"os/signal"
	"syscall"
	"time"

	"chat-agent-system/internal/app"
	"chat-agent-system/internal/app/eventsconsumer"
	"chat-agent-system/internal/config"
	"chat-agent-system/internal/localpg"
	"chat-agent-system/internal/server"
	"chat-agent-system/internal/workflow"
)

func main() {
	envPath := flag.String("env", ".env", "环境配置文件路径")
	port := flag.Int("port", 0, "HTTP 服务端口 (默认读取配置文件或 8080)")
	host := flag.String("host", "", "HTTP 监听主机 (默认读取配置文件或 127.0.0.1)")
	backend := flag.String("backend", "", "智能体后端调度覆盖 (stub, pi-agent, pi-native)")
	flag.Parse()

	cfg, err := config.Load(*envPath)
	if err != nil {
		fmt.Fprintf(os.Stderr, "加载配置失败: %v\n", err)
		os.Exit(1)
	}

	if *port > 0 {
		cfg.HTTPPort = *port
	}
	if *host != "" {
		cfg.HTTPHost = *host
	}
	if *backend != "" {
		cfg.AgentBackend = *backend
	}

	rootCtx, stop := signal.NotifyContext(context.Background(), os.Interrupt, syscall.SIGTERM)
	defer stop()

	pg, err := localpg.Ensure(rootCtx, cfg)
	if err != nil {
		fmt.Fprintf(os.Stderr, "localpg: %v\n", err)
		os.Exit(1)
	}
	if pg != nil {
		defer func() { _ = pg.Stop() }()
		fmt.Printf("localpg managed PGDATA=%s port=%d\n", pg.PGData, pg.Port)
	}

	rt, err := app.New(cfg)
	if err != nil {
		fmt.Fprintf(os.Stderr, "业务服务初始化失败: %v\n", err)
		os.Exit(1)
	}
	// Inject executor/quality-gate key validators without giving workflowdef a production import of workflow.
	if rt.WFDef != nil {
		rt.WFDef.SetRegistryValidators(workflow.ValidateExecutorKey, workflow.ValidateQualityGateKey)
	}
	if err := rt.Start(rootCtx); err != nil {
		fmt.Fprintf(os.Stderr, "后台 worker 启动失败: %v\n", err)
		os.Exit(1)
	}

	var eventsH *eventsconsumer.Handle
	if eventsconsumer.Enabled() && rt.Pool != nil {
		eventsH, err = eventsconsumer.Start(rootCtx, rt.Pool, cfg)
		if err != nil {
			fmt.Fprintf(os.Stderr, "eventsconsumer: %v\n", err)
			os.Exit(1)
		}
		fmt.Printf("events in-process owner=%s mode=cluster-table\n", eventsH.Owner)
	}

	srv := server.NewFromRuntime(rt)
	if eventsH != nil {
		srv.SetEventsAlive(func() bool { return eventsH.Alive(30 * time.Second) })
	}

	addr := fmt.Sprintf("%s:%d", cfg.HTTPHost, cfg.HTTPPort)
	fmt.Println("==================================================================")
	fmt.Println("   🤖 医疗器械商务与研发智能参谋系统 - HTTP / SSE API 守护服务   ")
	fmt.Println("==================================================================")
	fmt.Printf("服务环境: %s | 智能体后端: %s\n", cfg.Env, cfg.AgentBackend)
	fmt.Printf("监听地址: http://%s\n", addr)
	fmt.Println("核心服务路由:")
	fmt.Println("  • [GET]  /api/v1/health             - 服务健康检查与基础设施状态")
	fmt.Println("  • [GET]  /api/v1/projects           - 查询项目当前上下文与黑板事实")
	fmt.Println("  • [POST] /api/v1/chat               - 标准阻塞式对话协同接口")
	fmt.Println("  • [POST] /api/v1/chat/stream        - 核心 SSE 流式长连接对话协议 (分层按需)")
	fmt.Println("  • [GET]  /api/v1/chat/stream        - SSE 流式长连接对话调试接口 (EventSource)")
	fmt.Println("  • [POST] /api/v1/tasks/run/stream   - 独立任务与阶段执行 SSE 监控流 (精简进度与拦截)")
	fmt.Println("  • [GET]  /api/v1/context/user       - 获取单用户私有上下文与习惯偏好画像")
	fmt.Println("  • [PUT]  /api/v1/context/user/preference - 整理或更新用户个性化产出风格")
	fmt.Println("  • [GET]  /api/v1/context/project    - 获取项目多方协同黑板事实与里程碑决策流")
	fmt.Println("  • [GET]  /api/v1/chat/sessions      - 分页查询用户/项目历史会话列表")
	fmt.Println("  • [GET]  /api/v1/chat/messages      - 分页拉取指定会话多轮历史消息记录")
	fmt.Println("  • [GET]  /api/v1/context/cases/recommend - 往期受控标杆案例与历史报价语义自推荐")
	fmt.Println("  • [GET]  /api/v1/docgov/document    - 查询受控文档治理状态看板")
	fmt.Println("  • [POST] /api/v1/docgov/claim       - 查阅并认领受控文档 (正式稿方可认领)")
	fmt.Println("  • [POST] /api/v1/docgov/proposal    - 对已认领文档发起变更提案")
	fmt.Println("  • [POST] /api/v1/docgov/proposal/decide - 编制负责人拍板同意/拒绝变更提案")
	fmt.Println("  • [POST] /api/v1/docgov/submit-review - 编制完成提交审核 (文档正文锁定)")
	fmt.Println("  • [POST] /api/v1/docgov/review-decision - 审核通过或驳回 (驳回必填原因且解锁)")
	fmt.Println("  • [POST] /api/v1/docgov/approver-decision - 最终批准或驳回")
	fmt.Println("  • [GET/POST] /api/v1/docgov/comments - 批注与修改建议管理")
	fmt.Println("  • [GET]  /api/v1/docgov/audit-logs  - 全生命周期审计轨迹流")
	fmt.Println("==================================================================")

	go func() {
		if err := srv.Start(); err != nil && err != http.ErrServerClosed {
			fmt.Fprintf(os.Stderr, "服务启动失败: %v\n", err)
			os.Exit(1)
		}
	}()

	<-rootCtx.Done()
	fmt.Println("\n正在优雅停止服务...")
	if eventsH != nil {
		eventsH.Stop()
	}
	// T20-B7：drain 窗口与 AGENT_TIMEOUT 量级对齐（默认 min(AgentTimeout, 35m)+30s 缓冲），
	// 消除「5s Shutdown vs 30m 长任务」错配；可由 SHUTDOWN_DRAIN_TIMEOUT 显式覆盖。
	drain := shutdownDrainTimeout(cfg)
	ctx, cancel := context.WithTimeout(context.Background(), drain)
	defer cancel()
	fmt.Printf("等待 in-flight 请求/长任务排空（drain 上限 %s）...\n", drain)
	if err := srv.Shutdown(ctx); err != nil {
		fmt.Fprintf(os.Stderr, "服务停止异常: %v\n", err)
	} else {
		fmt.Println("服务已安全退出。")
	}
}

// shutdownDrainTimeout 解析 T20-B7 drain 窗口：
//   - SHUTDOWN_DRAIN_TIMEOUT 显式设置则优先（校验 >0）；
//   - 否则默认 min(AGENT_TIMEOUT, 35m) + 30s 缓冲（compose stop_grace_period 需 ≥ 此值）。
func shutdownDrainTimeout(cfg *config.Config) time.Duration {
	if v := os.Getenv("SHUTDOWN_DRAIN_TIMEOUT"); v != "" {
		if d, err := time.ParseDuration(v); err == nil && d > 0 {
			return d
		}
	}
	at := 30 * time.Minute
	if cfg != nil && cfg.AgentTimeout > 0 {
		at = cfg.AgentTimeout
	}
	cap := 35 * time.Minute
	if at > cap {
		at = cap
	}
	return at + 30*time.Second
}