6

backend/app/services/doc_service.py

@@ -41,6 +41,7 @@ class DocumentService:
         kb_id: str,
         filename: str,
         content: bytes,
+        category_id: str | None = None,
     ) -> Document:
         """上传文档。
 
@@ -111,6 +112,7 @@ class DocumentService:
             doc_token_hash=doc_token_hash,
             doc_token_encrypted=doc_token_encrypted,
             doc_token_hint=doc_token_hint,
+            category_id=category_id,
         )
 
         # 更新存储路径中的 document_id
@@ -200,3 +202,87 @@ class DocumentService:
 
         processor = LocalDocumentProcessor(self._session)
         processor.process(doc.id)
+
+    def create_text(
+        self,
+        user: User,
+        kb_id: str,
+        title: str,
+        content: str,
+        content_format: str = "markdown",
+        category_id: str | None = None,
+    ) -> Document:
+        """创建文本内容文档（无需上传文件）。"""
+        # 校验 KB 归属
+        kb = self._kb_repo.get_by_id(kb_id)
+        if kb is None or kb.user_id != user.id or kb.status == "DELETED":
+            raise NotFoundError("知识库不存在。")
+
+        # 生成文档 token
+        doc_token = generate_token()
+        doc_token_hash = hash_token(doc_token)
+        doc_token_encrypted = encrypt_token(doc_token)
+        doc_token_hint = doc_token[-8:] if len(doc_token) >= 8 else doc_token
+
+        # 计算内容大小
+        content_size = len(content.encode("utf-8"))
+
+        # 配额检查
+        self._check_quota(user, content_size)
+
+        # 从内容提取摘要和关键词
+        summary = self._extract_summary(content)
+        keywords = self._extract_keywords(content)
+
+        # 创建文档记录
+        doc = Document(
+            knowledge_base_id=kb_id,
+            user_id=user.id,
+            category_id=category_id,
+            original_filename=title,
+            storage_path=None,  # 文本内容不存储文件
+            markdown_path=None,
+            content=content,
+            content_format=content_format,
+            file_size=content_size,
+            mime_type="text/markdown" if content_format == "markdown" else "text/plain",
+            file_ext=".md" if content_format == "markdown" else ".txt",
+            sha256=None,
+            doc_token_hash=doc_token_hash,
+            doc_token_encrypted=doc_token_encrypted,
+            doc_token_hint=doc_token_hint,
+            title=title,
+            content_summary=summary,
+            keywords=keywords,
+            status="READY",  # 文本内容直接就绪
+        )
+        self._session.add(doc)
+
+        # 扣减配额
+        self._deduct_quota(user, content_size)
+
+        self._session.commit()
+        return doc
+
+    def _extract_summary(self, content: str, max_len: int = 200) -> str:
+        """提取摘要。"""
+        import re
+
+        text = re.sub(r"[#*_`\[\]()>]", "", content)
+        text = re.sub(r"\s+", " ", text).strip()
+        if len(text) <= max_len:
+            return text
+        return text[:max_len] + "..."
+
+    def _extract_keywords(self, content: str, top_k: int = 10) -> str:
+        """提取关键词。"""
+        try:
+            import jieba.analyse
+            import re
+
+            text = re.sub(r"[#*_`\[\]()>]", "", content)
+            text = re.sub(r"\s+", " ", text).strip()
+            keywords = jieba.analyse.extract_tags(text, topK=top_k)
+            return ",".join(keywords)
+        except Exception:
+            return ""