delete_document
Remove documents from Paperlib MCP by deleting database entries and optionally associated PDF files to manage your academic literature collection.
Instructions
删除指定文档
从数据库删除文档及其所有关联数据（chunks、embeddings、导入记录等）。 可选择同时删除 MinIO 中的 PDF 文件。
Args: doc_id: 文档的唯一标识符 also_delete_object: 是否同时删除 MinIO 中的 PDF 文件，默认 False
Returns: 删除结果，包含删除的记录数量
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| doc_id | Yes | ||
| also_delete_object | No |
Implementation Reference
- src/paperlib_mcp/tools/fetch.py:363-438 (handler)The core handler function for the 'delete_document' tool. It handles deletion of the document from the database tables (ingest_jobs, documents, chunks, chunk_embeddings) and optionally deletes the associated PDF file from MinIO storage using delete_object. Returns a success dict with stats.@mcp.tool() def delete_document( doc_id: str, also_delete_object: bool = False, ) -> dict[str, Any]: """删除指定文档 从数据库删除文档及其所有关联数据（chunks、embeddings、导入记录等）。 可选择同时删除 MinIO 中的 PDF 文件。 Args: doc_id: 文档的唯一标识符 also_delete_object: 是否同时删除 MinIO 中的 PDF 文件，默认 False Returns: 删除结果，包含删除的记录数量 """ try: # 先获取文档信息 doc = query_one( "SELECT pdf_key FROM documents WHERE doc_id = %s", (doc_id,) ) if not doc: return { "success": False, "error": f"Document not found: {doc_id}", "doc_id": doc_id, } pdf_key = doc["pdf_key"] # 统计将要删除的数据 stats = query_one( """ SELECT (SELECT COUNT(*) FROM chunks WHERE doc_id = %s) as chunk_count, (SELECT COUNT(*) FROM chunk_embeddings ce JOIN chunks c ON ce.chunk_id = c.chunk_id WHERE c.doc_id = %s) as embedding_count, (SELECT COUNT(*) FROM ingest_jobs WHERE doc_id = %s) as job_count """, (doc_id, doc_id, doc_id) ) # 删除导入记录 execute("DELETE FROM ingest_jobs WHERE doc_id = %s", (doc_id,)) # 删除文档（级联删除 chunks 和 embeddings） execute("DELETE FROM documents WHERE doc_id = %s", (doc_id,)) result = { "success": True, "doc_id": doc_id, "deleted_chunks": stats["chunk_count"] if stats else 0, "deleted_embeddings": stats["embedding_count"] if stats else 0, "deleted_jobs": stats["job_count"] if stats else 0, "object_deleted": False, } # 可选删除 MinIO 对象 if also_delete_object and pdf_key: delete_result = delete_object(pdf_key) result["object_deleted"] = delete_result.get("deleted", False) result["pdf_key"] = pdf_key return result except Exception as e: return { "success": False, "error": str(e), "doc_id": doc_id, }
- src/paperlib_mcp/server.py:15-36 (registration)Registration of the fetch tools, including delete_document, by calling register_fetch_tools(mcp) in the main MCP server setup.from paperlib_mcp.tools.fetch import register_fetch_tools from paperlib_mcp.tools.writing import register_writing_tools # M2 GraphRAG 工具 from paperlib_mcp.tools.graph_extract import register_graph_extract_tools from paperlib_mcp.tools.graph_canonicalize import register_graph_canonicalize_tools from paperlib_mcp.tools.graph_community import register_graph_community_tools from paperlib_mcp.tools.graph_summarize import register_graph_summarize_tools from paperlib_mcp.tools.graph_maintenance import register_graph_maintenance_tools # M3 Review 工具 from paperlib_mcp.tools.review import register_review_tools # M4 Canonicalization & Grouping 工具 from paperlib_mcp.tools.graph_relation_canonicalize import register_graph_relation_canonicalize_tools from paperlib_mcp.tools.graph_claim_grouping import register_graph_claim_grouping_tools from paperlib_mcp.tools.graph_v12 import register_graph_v12_tools register_health_tools(mcp) register_import_tools(mcp) register_search_tools(mcp) register_fetch_tools(mcp)