首次提交：初始化项目代码

2026-07-31 14:44:48 +08:00
commit a4fe393571
119 changed files with 29105 additions and 0 deletions
@@ -0,0 +1,165 @@
+package document
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"fmt"
+
+	"github.com/google/uuid"
+	"resume-platform/internal/model"
+	"resume-platform/internal/parser"
+	"resume-platform/internal/repository"
+	"resume-platform/internal/service/knowledge"
+	"resume-platform/internal/service/resume"
+	"resume-platform/pkg/logger"
+)
+
+type DocumentService interface {
+	UploadDocument(ctx context.Context, userID string, fileName string, fileType string, fileSize int64, content string) (*model.Document, error)
+	GetDocuments(ctx context.Context, userID string) ([]*model.Document, error)
+	GetDocument(ctx context.Context, userID string, documentID string) (*model.Document, error)
+	DeleteDocument(ctx context.Context, userID string, documentID string) error
+	ProcessDocument(ctx context.Context, documentID string) error
+	ParseAndGenerateResume(ctx context.Context, userID string, fileName string, content []byte) (*model.Resume, error)
+}
+
+type documentService struct {
+	repo              repository.ResumeRepository
+	resumeGenerator   *resume.ResumeGeneratorService
+	knowledgeBaseService *knowledge.KnowledgeBaseService
+}
+
+func NewDocumentService(repo repository.ResumeRepository, resumeGenerator *resume.ResumeGeneratorService) DocumentService {
+	return &documentService{repo: repo, resumeGenerator: resumeGenerator}
+}
+
+func NewDocumentServiceWithKnowledge(repo repository.ResumeRepository, resumeGenerator *resume.ResumeGeneratorService, knowledgeBaseService *knowledge.KnowledgeBaseService) DocumentService {
+	return &documentService{repo: repo, resumeGenerator: resumeGenerator, knowledgeBaseService: knowledgeBaseService}
+}
+
+func (s *documentService) UploadDocument(ctx context.Context, userID string, fileName string, fileType string, fileSize int64, content string) (*model.Document, error) {
+	document := &model.Document{
+		ID:       generateID(),
+		UserID:   userID,
+		Name:     fileName,
+		Type:     fileType,
+		Size:     fileSize,
+		Content:  content,
+		Status:   "uploading",
+	}
+	
+	err := s.repo.CreateDocument(document)
+	if err != nil {
+		return nil, err
+	}
+	
+	go s.ProcessDocument(ctx, document.ID)
+	
+	return document, nil
+}
+
+func (s *documentService) GetDocuments(ctx context.Context, userID string) ([]*model.Document, error) {
+	return s.repo.GetDocumentsByUserID(userID)
+}
+
+func (s *documentService) GetDocument(ctx context.Context, userID string, documentID string) (*model.Document, error) {
+	return s.repo.GetDocument(documentID, userID)
+}
+
+func (s *documentService) DeleteDocument(ctx context.Context, userID string, documentID string) error {
+	return s.repo.DeleteDocument(documentID, userID)
+}
+
+func (s *documentService) ProcessDocument(ctx context.Context, documentID string) error {
+	document, err := s.repo.GetDocumentByID(documentID)
+	if err != nil {
+		return err
+	}
+	
+	if document.Content == "" {
+		document.Status = "failed"
+		document.ErrorMsg = "document content is empty"
+		return s.repo.UpdateDocument(document)
+	}
+	
+	chunks := s.splitText(document.Content, 512)
+	
+	for i, chunk := range chunks {
+		metadata := map[string]interface{}{
+			"document_id": document.ID,
+			"chunk_index": i,
+			"user_id":     document.UserID,
+		}
+		metadataJSON, _ := json.Marshal(metadata)
+		
+		documentChunk := &model.DocumentChunk{
+			ID:         generateID(),
+			DocumentID: document.ID,
+			ChunkIndex: i,
+			Content:    chunk,
+			Metadata:   string(metadataJSON),
+		}
+		
+		err := s.repo.CreateDocumentChunk(documentChunk)
+		if err != nil {
+			document.Status = "failed"
+			document.ErrorMsg = err.Error()
+			s.repo.UpdateDocument(document)
+			return err
+		}
+	}
+	
+	if s.knowledgeBaseService != nil {
+		err = s.knowledgeBaseService.AddDocument(ctx, document.ID, document.Content, document.UserID)
+		if err != nil {
+			logger.Warnf("Failed to add document to knowledge base: %v", err)
+		} else {
+			logger.Infof("Document %s added to knowledge base", document.ID)
+		}
+	}
+	
+	document.Status = "processed"
+	return s.repo.UpdateDocument(document)
+}
+
+func (s *documentService) splitText(text string, chunkSize int) []string {
+	var chunks []string
+	words := []rune(text)
+	
+	for i := 0; i < len(words); i += chunkSize {
+		end := i + chunkSize
+		if end > len(words) {
+			end = len(words)
+		}
+		chunks = append(chunks, string(words[i:end]))
+	}
+	
+	return chunks
+}
+
+func generateID() string {
+	return uuid.New().String()
+}
+
+func (s *documentService) ParseAndGenerateResume(ctx context.Context, userID string, fileName string, content []byte) (*model.Resume, error) {
+	documentText, err := parser.ParseFile(fileName, bytes.NewReader(content))
+	if err != nil {
+		logger.Warnf("Failed to parse document %s: %v", fileName, err)
+		return nil, err
+	}
+	
+	if documentText == "" {
+		return nil, fmt.Errorf("document content is empty after parsing")
+	}
+	
+	resume, err := s.resumeGenerator.GenerateFromDocument(ctx, documentText)
+	if err != nil {
+		return nil, err
+	}
+	
+	resume.UserID = userID
+	fmt.Printf("DEBUG document.go: resume.UserID set to: '%s'\n", resume.UserID)
+	
+	return resume, nil
+}