mcp-scholar

Instructions

获取论文详细信息

Args:
    paper_id: 论文ID

Returns:
    Dict: 论文详细信息

Implementation Reference

• src/mcp_scholar/server.py:285-319 (handler)

  The MCP tool handler for 'paper_detail' that takes a paper_id, calls the helper get_paper_detail, processes the result (adding URLs), and returns success/error dict.

  @mcp.tool()
  async def paper_detail(ctx: Context, paper_id: str) -> Dict[str, Any]:
      """
      获取论文详细信息
  
      Args:
          paper_id: 论文ID
  
      Returns:
          Dict: 论文详细信息
      """
      try:
          # 移除进度显示
          logger.info(f"正在获取论文ID为 {paper_id} 的详细信息...")
          detail = await get_paper_detail(paper_id)
  
          if detail:
              # 确保URL信息被返回
              if "url" not in detail and detail.get("pub_url"):
                  detail["url"] = detail["pub_url"]
  
              # 如果有DOI，添加DOI URL
              if "doi" in detail and "doi_url" not in detail:
                  detail["doi_url"] = f"https://doi.org/{detail['doi']}"
  
              return {"status": "success", "detail": detail}
          else:
              # 移除错误通知
              logger.warning(f"未找到ID为 {paper_id} 的论文")
              return {"status": "error", "message": f"未找到ID为 {paper_id} 的论文"}
      except Exception as e:
          # 移除错误通知
          logger.error(f"获取论文详情失败: {str(e)}", exc_info=True)
          return {"status": "error", "message": "论文详情服务暂时不可用", "error": str(e)}

• src/mcp_scholar/scholar.py:248-336 (helper)

  Core helper function that queries the OpenAlex API to fetch detailed paper information based on paper_id (supports DOI, OpenAlex ID, arXiv), processes abstract from inverted index, extracts authors, venue, DOI, PDF, concepts.

  async def get_paper_detail(paper_id: str) -> Optional[Dict[str, Any]]:
      """
      通过OpenAlex API获取论文详情
  
      Args:
          paper_id: 论文ID，可以是OpenAlex ID、DOI或ArXiv ID
  
      Returns:
          Dict: 论文详细信息
      """
      try:
          # 设置电子邮件参数（礼貌请求）
          email_param = f"?mailto={EMAIL}" if EMAIL else ""
  
          # 确定使用什么ID类型
          if paper_id.startswith("10."):  # 看起来是DOI
              api_url = f"{OPENALEX_API}/works/doi:{paper_id}{email_param}"
          elif paper_id.startswith("W"):  # OpenAlex ID
              api_url = f"{OPENALEX_API}/works/{paper_id}{email_param}"
          elif paper_id.lower().startswith("arxiv:"):  # arXiv ID
              api_url = f"{OPENALEX_API}/works/arxiv:{paper_id.replace('arxiv:', '')}{email_param}"
          else:  # 尝试作为OpenAlex ID (不带前缀的)
              api_url = f"{OPENALEX_API}/works/W{paper_id}{email_param}"
  
          async with httpx.AsyncClient(timeout=10.0) as client:
              response = await client.get(api_url)
  
              if response.status_code == 200:
                  data = response.json()
  
                  # 提取论文详细信息
                  result = {
                      "title": data.get("title", "未知标题"),
                      "abstract": "",  # 默认为空，稍后处理
                      "citations": data.get("cited_by_count", 0),
                      "year": data.get("publication_year", "未知年份"),
                      "venue": "",  # 需要从期刊/会议信息中提取
                      "paper_id": data.get("id", "").replace("https://openalex.org/", ""),
                      "url": data.get("id", ""),
                  }
  
                  # 处理摘要（OpenAlex 摘要是倒排索引格式）
                  if data.get("abstract_inverted_index"):
                      result["abstract"] = convert_inverted_index_to_text(
                          data.get("abstract_inverted_index", {})
                      )
  
                  # 处理作者信息
                  authors = data.get("authorships", [])
                  author_names = []
                  for author in authors:
                      if author.get("author", {}).get("display_name"):
                          author_names.append(author["author"]["display_name"])
                  result["authors"] = ", ".join(author_names)
  
                  # 处理期刊/会议信息
                  if data.get("host_venue", {}).get("display_name"):
                      result["venue"] = data["host_venue"]["display_name"]
  
                  # 处理DOI信息
                  if data.get("doi"):
                      result["doi"] = data["doi"]
                      result["doi_url"] = f"https://doi.org/{result['doi']}"
  
                  # 添加PDF链接（如果有）
                  if data.get("open_access", {}).get("oa_url"):
                      result["pdf_url"] = data["open_access"]["oa_url"]
  
                  # 添加关键概念
                  if data.get("concepts"):
                      concepts = []
                      for concept in data["concepts"]:
                          if (
                              concept.get("display_name")
                              and concept.get("score", 0) > 0.5
                          ):  # 只添加相关性高的概念
                              concepts.append(concept["display_name"])
                      if concepts:
                          result["concepts"] = ", ".join(concepts)
  
                  return result
              else:
                  print(f"获取论文详情错误: {response.status_code} - {response.text}")
                  return None
  
      except Exception as e:
          print(f"获取论文详情时出错: {str(e)}")
          return None