Files
smartmate/backend/memory/service/read_service.go
Losita bf1f1defa5 Version: 0.9.14.dev.260410
后端:
  1. LLM 客户端从 newAgent/llm 提升为 infra/llm 基础设施层
     - 删除 backend/newAgent/llm/(ark.go / ark_adapter.go / client.go / json.go)
     - 等价迁移至 backend/infra/llm/,所有 newAgent node 与 service 统一改引用 infrallm
     - 消除 newAgent 对模型客户端的私有依赖,为 memory / websearch 等多模块复用铺路
  2. RAG 基础设施完成可运行态接入(factory / runtime / observer / service 四层成型)
     - 新建 backend/infra/rag/factory.go / runtime.go / observe.go / observer.go /
  service.go:工厂创建、运行时生命周期、轻量观测接口、检索服务门面
     - 更新 infra/rag/config/config.go:补齐 Milvus / Embed / Reranker 全部配置项与默认值
     - 更新 infra/rag/embed/eino_embedder.go:增强 Eino embedding 适配,支持 BaseURL / APIKey 环境变量 / 超时 /
  维度等参数
     - 更新 infra/rag/store/milvus_store.go:完整实现 Milvus 向量存储(建集合 / 建 Index / Upsert / Search /
  Delete),支持 COSINE / L2 / IP 度量
     - 更新 infra/rag/core/pipeline.go:适配 Runtime 接口,Pipeline 由 factory 注入而非手动拼装
     - 更新 infra/rag/corpus/memory_corpus.go / vector_store.go:对接 Memory 模块数据源与 Store 接口扩展
  3. Memory 模块从 Day1 骨架升级为 Day2 完整可运行态
     - 新建 memory/module.go:统一门面 Module,对外封装 EnqueueExtract / ReadService / ManageService / WithTx /
  StartWorker,启动层只依赖这一个入口
     - 新建 memory/orchestrator/llm_write_orchestrator.go:LLM 驱动的记忆抽取编排器,替代原 mock 抽取
     - 新建 memory/service/read_service.go:按用户开关过滤 + 轻量重排 + 访问时间刷新的读取链路
     - 新建 memory/service/manage_service.go:记忆管理面能力(列出 / 软删除 / 开关读写),删除同步写审计日志
     - 新建 memory/service/common.go:服务层公共工具
     - 新建 memory/worker/loop.go:后台轮询循环 RunPollingLoop,定时抢占 pending 任务并推进
     - 新建 memory/utils/audit.go / settings.go:审计日志构造、用户设置过滤等纯函数
     - 更新 memory/model/item.go / job.go / settings.go / config.go / status.go:补齐 DTO 字段与状态常量
     - 更新 memory/repo/item_repo.go / job_repo.go / audit_repo.go / settings_repo.go:补齐 CRUD 与查询能力
     - 更新 memory/worker/runner.go:Runner 对接 Module 与 LLM 抽取器,任务状态机完整化
     - 更新 memory/README.md:同步模块现状说明
  4. newAgent 接入 Memory 读取注入与工具注册依赖预埋
     - 新建 service/agentsvc/agent_memory.go:定义 MemoryReader 接口 + injectMemoryContext,在 graph
  执行前统一补充记忆上下文
     - 更新 service/agentsvc/agent.go:新增 memoryReader 字段与 SetMemoryReader 方法
     - 更新 service/agentsvc/agent_newagent.go:调用 injectMemoryContext 注入 pinned block,检索失败仅降级不阻断主链路
     - 更新 newAgent/tools/registry.go:新增 DefaultRegistryDeps(含 RAGRuntime),工具注册表支持依赖注入
  5. 启动流程与事件处理器接线更新
     - 更新 cmd/start.go:初始化 RAG Runtime → Memory Module → 注册事件处理器 → 启动 Worker 后台轮询
     - 更新 service/events/memory_extract_requested.go:改用 memory.Module.WithTx(tx) 统一门面,事件处理器不再直接依赖
  repo/service 内部包
  6. 缓存插件与配置同步
     - 更新 middleware/cache_deleter.go:静默忽略 MemoryJob / MemoryItem / MemoryAuditLog / MemoryUserSetting
  等新模型,避免日志刷屏;清理冗余注释
     - 更新 config.example.yaml:补齐 rag / memory / websearch 配置段及默认值
     - 更新 go.mod / go.sum:新增 eino-ext/openai / json-patch / go-openai 依赖
  前端:无 仓库:无
2026-04-10 23:17:38 +08:00

348 lines
8.6 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 service
import (
"context"
"fmt"
"sort"
"strconv"
"strings"
"time"
infrarag "github.com/LoveLosita/smartflow/backend/infra/rag"
memorymodel "github.com/LoveLosita/smartflow/backend/memory/model"
memoryrepo "github.com/LoveLosita/smartflow/backend/memory/repo"
memoryutils "github.com/LoveLosita/smartflow/backend/memory/utils"
"github.com/LoveLosita/smartflow/backend/model"
)
const (
defaultRetrieveLimit = 5
maxRetrieveLimit = 20
)
// ReadService 负责 memory 模块内部的读取、门控与轻量重排。
//
// 职责边界:
// 1. 负责把 memory_items 读出来并做用户设置过滤;
// 2. 负责最小可用的排序与截断,为后续 prompt 注入提供稳定入口;
// 3. 不直接依赖 newAgent不负责真正把记忆拼进 prompt。
type ReadService struct {
itemRepo *memoryrepo.ItemRepo
settingsRepo *memoryrepo.SettingsRepo
ragRuntime infrarag.Runtime
cfg memorymodel.Config
}
func NewReadService(
itemRepo *memoryrepo.ItemRepo,
settingsRepo *memoryrepo.SettingsRepo,
ragRuntime infrarag.Runtime,
cfg memorymodel.Config,
) *ReadService {
return &ReadService{
itemRepo: itemRepo,
settingsRepo: settingsRepo,
ragRuntime: ragRuntime,
cfg: cfg,
}
}
// Retrieve 读取可供后续注入使用的候选记忆。
func (s *ReadService) Retrieve(ctx context.Context, req memorymodel.RetrieveRequest) ([]memorymodel.ItemDTO, error) {
if s == nil || s.itemRepo == nil || s.settingsRepo == nil {
return nil, nil
}
if req.UserID <= 0 {
return nil, nil
}
now := req.Now
if now.IsZero() {
now = time.Now()
}
setting, err := s.settingsRepo.GetByUserID(ctx, req.UserID)
if err != nil {
return nil, err
}
effectiveSetting := memoryutils.EffectiveUserSetting(setting, req.UserID)
if !effectiveSetting.MemoryEnabled {
return nil, nil
}
limit := normalizeLimit(req.Limit, defaultRetrieveLimit, maxRetrieveLimit)
if s.cfg.RAGEnabled && s.ragRuntime != nil && strings.TrimSpace(req.Query) != "" {
items, ragErr := s.retrieveByRAG(ctx, req, effectiveSetting, limit, now)
if ragErr == nil && len(items) > 0 {
return items, nil
}
}
return s.retrieveByLegacy(ctx, req, limit, now, effectiveSetting)
}
func (s *ReadService) retrieveByLegacy(
ctx context.Context,
req memorymodel.RetrieveRequest,
limit int,
now time.Time,
effectiveSetting model.MemoryUserSetting,
) ([]memorymodel.ItemDTO, error) {
if !effectiveSetting.MemoryEnabled {
return nil, nil
}
query := memorymodel.ItemQuery{
UserID: req.UserID,
ConversationID: req.ConversationID,
AssistantID: req.AssistantID,
RunID: req.RunID,
Statuses: []string{model.MemoryItemStatusActive},
MemoryTypes: normalizeRetrieveMemoryTypes(req.MemoryTypes),
IncludeGlobal: true,
OnlyUnexpired: true,
Limit: normalizeLimit(limit*3, limit*3, maxRetrieveLimit*3),
Now: now,
}
items, err := s.itemRepo.FindByQuery(ctx, query)
if err != nil {
return nil, err
}
items = memoryutils.FilterItemsBySetting(items, effectiveSetting)
if len(items) == 0 {
return nil, nil
}
sort.SliceStable(items, func(i, j int) bool {
left := scoreRetrievedItem(items[i], now, req.ConversationID)
right := scoreRetrievedItem(items[j], now, req.ConversationID)
if left == right {
return items[i].ID > items[j].ID
}
return left > right
})
if len(items) > limit {
items = items[:limit]
}
_ = s.itemRepo.TouchLastAccessAt(ctx, collectMemoryIDs(items), now)
return toItemDTOs(items), nil
}
func (s *ReadService) retrieveByRAG(
ctx context.Context,
req memorymodel.RetrieveRequest,
effectiveSetting model.MemoryUserSetting,
limit int,
now time.Time,
) ([]memorymodel.ItemDTO, error) {
if !effectiveSetting.MemoryEnabled {
return nil, nil
}
result, err := s.ragRuntime.RetrieveMemory(ctx, infrarag.MemoryRetrieveRequest{
Query: req.Query,
TopK: limit,
Threshold: s.cfg.Threshold,
Action: "search",
UserID: req.UserID,
ConversationID: req.ConversationID,
AssistantID: req.AssistantID,
RunID: req.RunID,
MemoryTypes: normalizeRetrieveMemoryTypes(req.MemoryTypes),
})
if err != nil || result == nil || len(result.Items) == 0 {
return nil, err
}
items := make([]memorymodel.ItemDTO, 0, len(result.Items))
ids := make([]int64, 0, len(result.Items))
for _, hit := range result.Items {
dto, memoryID := buildMemoryDTOFromRetrieveHit(hit)
if !effectiveSetting.ImplicitMemoryEnabled && !dto.IsExplicit {
continue
}
if !effectiveSetting.SensitiveMemoryEnabled && dto.SensitivityLevel > 0 {
continue
}
if dto.ID <= 0 && memoryID > 0 {
dto.ID = memoryID
}
items = append(items, dto)
if dto.ID > 0 {
ids = append(ids, dto.ID)
}
}
if len(items) > limit {
items = items[:limit]
}
_ = s.itemRepo.TouchLastAccessAt(ctx, ids, now)
return items, nil
}
func normalizeRetrieveMemoryTypes(raw []string) []string {
normalized := normalizeMemoryTypes(raw)
if len(normalized) > 0 {
return normalized
}
return []string{
memorymodel.MemoryTypeConstraint,
memorymodel.MemoryTypePreference,
memorymodel.MemoryTypeTodoHint,
memorymodel.MemoryTypeFact,
}
}
func scoreRetrievedItem(item model.MemoryItem, now time.Time, conversationID string) float64 {
score := 0.35*clamp01(item.Importance) + 0.3*clamp01(item.Confidence) + 0.2*recencyScore(item, now)
if item.IsExplicit {
score += 0.1
}
if strValue(item.ConversationID) != "" && strValue(item.ConversationID) == conversationID {
score += 0.08
}
switch item.MemoryType {
case memorymodel.MemoryTypeConstraint:
score += 0.12
case memorymodel.MemoryTypePreference:
score += 0.08
case memorymodel.MemoryTypeTodoHint:
score += 0.05
}
return score
}
func recencyScore(item model.MemoryItem, now time.Time) float64 {
base := item.UpdatedAt
if base == nil {
base = item.CreatedAt
}
if base == nil || now.Before(*base) {
return 0.5
}
age := now.Sub(*base)
switch {
case age <= 24*time.Hour:
return 1
case age <= 7*24*time.Hour:
return 0.85
case age <= 30*24*time.Hour:
return 0.65
case age <= 90*24*time.Hour:
return 0.45
default:
return 0.25
}
}
func clamp01(v float64) float64 {
if v < 0 {
return 0
}
if v > 1 {
return 1
}
return v
}
func collectMemoryIDs(items []model.MemoryItem) []int64 {
if len(items) == 0 {
return nil
}
ids := make([]int64, 0, len(items))
for _, item := range items {
if item.ID <= 0 {
continue
}
ids = append(ids, item.ID)
}
return ids
}
func buildMemoryDTOFromRetrieveHit(hit infrarag.RetrieveHit) (memorymodel.ItemDTO, int64) {
memoryID := parseMemoryIDFromDocumentID(hit.DocumentID)
metadata := hit.Metadata
dto := memorymodel.ItemDTO{
ID: memoryID,
UserID: int(readFloatLike(metadata["user_id"])),
ConversationID: readString(metadata["conversation_id"]),
AssistantID: readString(metadata["assistant_id"]),
RunID: readString(metadata["run_id"]),
MemoryType: readString(metadata["memory_type"]),
Title: readString(metadata["title"]),
Content: strings.TrimSpace(hit.Text),
Confidence: readFloatLike(metadata["confidence"]),
Importance: readFloatLike(metadata["importance"]),
SensitivityLevel: int(readFloatLike(metadata["sensitivity_level"])),
IsExplicit: readBoolLike(metadata["is_explicit"]),
Status: readString(metadata["status"]),
TTLAt: readTimeLike(metadata["ttl_at"]),
}
return dto, memoryID
}
func parseMemoryIDFromDocumentID(documentID string) int64 {
documentID = strings.TrimSpace(documentID)
if !strings.HasPrefix(documentID, "memory:") {
return 0
}
raw := strings.TrimPrefix(documentID, "memory:")
if strings.HasPrefix(raw, "uid:") {
return 0
}
parsed, err := strconv.ParseInt(raw, 10, 64)
if err != nil {
return 0
}
return parsed
}
func readString(v any) string {
if v == nil {
return ""
}
return strings.TrimSpace(fmt.Sprintf("%v", v))
}
func readFloatLike(v any) float64 {
switch value := v.(type) {
case float64:
return value
case float32:
return float64(value)
case int:
return float64(value)
case int64:
return float64(value)
case string:
parsed, err := strconv.ParseFloat(strings.TrimSpace(value), 64)
if err == nil {
return parsed
}
}
return 0
}
func readBoolLike(v any) bool {
switch value := v.(type) {
case bool:
return value
case string:
return strings.EqualFold(strings.TrimSpace(value), "true")
default:
return false
}
}
func readTimeLike(v any) *time.Time {
text := readString(v)
if text == "" {
return nil
}
parsed, err := time.Parse(time.RFC3339, text)
if err != nil {
return nil
}
return &parsed
}