get_document_chunks

Retrieve all text chunks from a document to analyze content structure. Provides chunk IDs, page numbers, and text summaries for comprehensive document examination.

Instructions

获取指定文档的所有 chunks 列表

根据 doc_id 获取该文档的所有文本块摘要信息。

Args: doc_id: 文档的唯一标识符

Returns: chunks 列表，每个包含 chunk_id、页码和文本摘要

Input Schema

TableJSON Schema

Name	Required	Description	Default
`doc_id`	Yes

Implementation Reference

src/paperlib_mcp/tools/fetch.py:190-243 (handler)

The handler function for the 'get_document_chunks' tool. It retrieves all chunks for a given document ID from the database, providing summaries including chunk_id, index, pages, token count, snippet, and embedding status.

@mcp.tool()
def get_document_chunks(doc_id: str) -> dict[str, Any]:
    """获取指定文档的所有 chunks 列表
    
    根据 doc_id 获取该文档的所有文本块摘要信息。
    
    Args:
        doc_id: 文档的唯一标识符
        
    Returns:
        chunks 列表，每个包含 chunk_id、页码和文本摘要
    """
    try:
        chunks = query_all(
            """
            SELECT 
                c.chunk_id,
                c.chunk_index,
                c.page_start,
                c.page_end,
                c.token_count,
                LEFT(c.text, 100) as snippet,
                CASE WHEN ce.chunk_id IS NOT NULL THEN true ELSE false END as has_embedding
            FROM chunks c
            LEFT JOIN chunk_embeddings ce ON c.chunk_id = ce.chunk_id
            WHERE c.doc_id = %s
            ORDER BY c.chunk_index
            """,
            (doc_id,)
        )
        
        return {
            "doc_id": doc_id,
            "chunk_count": len(chunks),
            "chunks": [
                {
                    "chunk_id": c["chunk_id"],
                    "chunk_index": c["chunk_index"],
                    "page_start": c["page_start"],
                    "page_end": c["page_end"],
                    "token_count": c["token_count"],
                    "snippet": c["snippet"] + "..." if len(c["snippet"]) >= 100 else c["snippet"],
                    "has_embedding": c["has_embedding"],
                }
                for c in chunks
            ],
        }
        
    except Exception as e:
        return {
            "error": str(e),
            "doc_id": doc_id,
            "chunks": [],
        }

src/paperlib_mcp/server.py:36-36 (registration)
Registers the fetch tools module, which includes the get_document_chunks tool, on the MCP server instance.
```
register_fetch_tools(mcp)
```

Paperlib MCP