refactor: 重构文档向量相关代码结构

common/eino/indexer.go

@@ -100,9 +100,9 @@ func (i *PGVectorIndexer) doStore(ctx context.Context, docs []*schema.Document,
 	}
 
 	// 转成业务实体
-	var chunks []*dto.VectorDocumentChunkMsg
+	var chunks []*dto.VectorDocumentVectorMsg
 	for idx, doc := range docs {
-		ck := new(dto.VectorDocumentChunkMsg)
+		ck := new(dto.VectorDocumentVectorMsg)
 		err = gconv.Struct(doc.MetaData, ck)
 		if err != nil {
 			glog.Errorf(ctx, "doStore err: %v", err)
@@ -126,7 +126,7 @@ func (i *PGVectorIndexer) doStore(ctx context.Context, docs []*schema.Document,
 		return
 	}
 	// 入库
-	rows, err = dao.DocumentChunk.BatchInsert(ctx, chunks)
+	rows, err = dao.DocumentVector.BatchInsert(ctx, chunks)
 
 	return
 }

common/eino/retriever.go

@@ -210,7 +210,7 @@ func (r *PGVectorRetriever) doRetrieveVector(ctx context.Context, query string,
 	}
 	datasetIds := gconv.Int64s(opts.DSLInfo["dataset_ids"])
 
-	rows, err := dao.DocumentChunk.GetAllByVector(ctx, datasetIds, queryVec, topK)
+	rows, err := dao.DocumentVector.GetAllByVector(ctx, datasetIds, queryVec, topK)
 	if err != nil {
 		return nil, err
 	}
@@ -239,7 +239,7 @@ func (r *PGVectorRetriever) doRetrieveMeilisearch(ctx context.Context, query str
 	datasetIds := gconv.Int64s(opts.DSLInfo["dataset_ids"])
 
 	// 调用你已有的 Meilisearch DAO
-	rows, err := dao.DocumentChunk.SearchByKeywords(ctx, query, datasetIds, topK)
+	rows, err := dao.DocumentVector.SearchByKeywords(ctx, query, datasetIds, topK)
 	if err != nil {
 		return nil, err
 	}

common/task/base_task.go

@@ -1,69 +0,0 @@
-package task
-
-import (
-	"time"
-
-	"gitea.com/red-future/common/beans"
-)
-
-type baseTaskCol struct {
-	beans.SQLBaseCol
-	TaskType       string
-	Status         string
-	Priority       string
-	ParentTaskID   string
-	TotalItems     string
-	ProcessedItems string
-	Progress       string
-	StartTime      string
-	EndTime        string
-	Duration       string
-	SuccessCount   string
-	FailCount      string
-	Executor       string
-	DocumentID     string
-	Remark         string
-}
-
-var BaseTaskCol = baseTaskCol{
-	SQLBaseCol:     beans.DefSQLBaseCol,
-	TaskType:       "task_type",
-	Status:         "status",
-	Priority:       "task_priority",
-	ParentTaskID:   "parent_task_id",
-	TotalItems:     "total_items",
-	ProcessedItems: "processed_items",
-	Progress:       "progress",
-	StartTime:      "start_time",
-	EndTime:        "end_time",
-	Duration:       "duration",
-	SuccessCount:   "success_count",
-	FailCount:      "fail_count",
-	Executor:       "executor",
-	DocumentID:     "document_id",
-	Remark:         "remark",
-}
-
-// SQLBaseTask 任务基类 - SQL版本
-type SQLBaseTask struct {
-	beans.SQLBaseDO `orm:",inline"`
-	// 任务核心信息
-	TaskType     TaskType     `orm:"task_type" json:"taskType" dc:"任务类型"`
-	Status       TaskStatus   `orm:"status" json:"status" dc:"任务状态"`
-	Priority     TaskPriority `orm:"task_priority" json:"priority,omitempty" dc:"任务优先级"`
-	ParentTaskID int64        `orm:"parent_task_id" json:"parentTaskId,omitempty" dc:"父任务ID"`
-	// 任务进度
-	TotalItems     int64   `orm:"total_items" json:"totalItems" dc:"总数"`
-	ProcessedItems int64   `orm:"processed_items" json:"processedItems" dc:"已处理数"`
-	Progress       float64 `orm:"progress" json:"progress" dc:"进度"` // 0~100 百分比
-	// 任务结果
-	StartTime    *time.Time `orm:"start_time" json:"startTime" dc:"开始时间"`
-	EndTime      *time.Time `orm:"end_time" json:"endTime,omitempty" dc:"结束时间"`
-	Duration     int64      `orm:"duration" json:"duration,omitempty" dc:"耗时(毫秒)"`
-	SuccessCount int64      `orm:"success_count" json:"successCount" dc:"成功数"`
-	FailCount    int64      `orm:"fail_count" json:"failCount" dc:"失败数"`
-	// 其他
-	Executor   string `orm:"executor" json:"executor,omitempty" dc:"执行器标识"`
-	DocumentID int64  `orm:"document_id" json:"documentId,omitempty" dc:"文档ID"`
-	Remark     string `orm:"remark" json:"remark,omitempty" dc:"备注/错误信息"`
-}

common/task/consts.go

@@ -1,30 +0,0 @@
-package task
-
-// TaskType 任务类型枚举：文档解析的三个子任务
-type TaskType string
-
-const (
-	TaskTypeExtractKeywords TaskType = "EXTRACT_KEYWORDS" // 提取关键词
-	TaskTypeGenerateVector  TaskType = "GENERATE_VECTOR"  // 生成向量
-	TaskTypeFullTextSearch  TaskType = "FULL_TEXT_SEARCH" // 全文检索
-	TaskTypeDocParse        TaskType = "DOC_PARSE"        // 顶层文档解析总任务
-)
-
-// TaskStatus 任务状态枚举
-type TaskStatus string
-
-const (
-	TaskStatusPending   TaskStatus = "PENDING"   // 待执行
-	TaskStatusRunning   TaskStatus = "RUNNING"   // 执行中
-	TaskStatusCompleted TaskStatus = "COMPLETED" // 已完成
-	TaskStatusFailed    TaskStatus = "FAILED"    // 执行失败
-)
-
-// TaskPriority 任务优先级
-type TaskPriority int
-
-const (
-	TaskPriorityLow    TaskPriority = 1 // 低
-	TaskPriorityMedium TaskPriority = 2 // 中
-	TaskPriorityHigh   TaskPriority = 3 // 高
-)