feat: Replace模式清理未使用标签

Author: lyingbug

internal/application/repository/tag.go

@@ -156,3 +156,15 @@ func (r *knowledgeTagRepository) CountUntaggedReferences(
 	}
 	return
 }
+
+// DeleteUnusedTags deletes tags that are not referenced by any knowledge or chunk.
+// Returns the number of deleted tags.
+func (r *knowledgeTagRepository) DeleteUnusedTags(ctx context.Context, tenantID uint64, kbID string) (int64, error) {
+	// Delete tags that have no references in both knowledges and chunks tables
+	result := r.db.WithContext(ctx).
+		Where("tenant_id = ? AND knowledge_base_id = ?", tenantID, kbID).
+		Where("id NOT IN (SELECT DISTINCT tag_id FROM knowledges WHERE tenant_id = ? AND knowledge_base_id = ? AND tag_id IS NOT NULL AND tag_id != '')", tenantID, kbID).
+		Where("id NOT IN (SELECT DISTINCT tag_id FROM chunks WHERE tenant_id = ? AND knowledge_base_id = ? AND tag_id IS NOT NULL AND tag_id != '')", tenantID, kbID).
+		Delete(&types.KnowledgeTag{})
+	return result.RowsAffected, result.Error
+}

internal/application/service/knowledge.go

@@ -3067,6 +3067,14 @@ func (s *knowledgeService) executeFAQImport(ctx context.Context, taskID string,
 				return fmt.Errorf("failed to delete chunk vectors: %w", err)
 			}
 			logger.Infof(ctx, "FAQ import task %s: deleted %d chunks (including updates)", taskID, len(chunksToDelete))
+
+			// 清理不再被引用的Tag
+			deletedTags, err := s.tagRepo.DeleteUnusedTags(ctx, tenantID, kb.ID)
+			if err != nil {
+				logger.Warnf(ctx, "FAQ import task %s: failed to cleanup unused tags: %v", taskID, err)
+			} else if deletedTags > 0 {
+				logger.Infof(ctx, "FAQ import task %s: cleaned up %d unused tags", taskID, deletedTags)
+			}
 		}
 	} else {
 		// Append模式：查询已存在的条目，跳过未变化的

internal/types/interfaces/tag.go

@@ -54,4 +54,6 @@ type KnowledgeTagRepository interface {
 		tenantID uint64,
 		kbID string,
 	) (knowledgeCount int64, chunkCount int64, err error)
+	// DeleteUnusedTags deletes tags that are not referenced by any knowledge or chunk.
+	DeleteUnusedTags(ctx context.Context, tenantID uint64, kbID string) (int64, error)
 }