diff --git a/langchain/document_loaders/pdf.py b/langchain/document_loaders/pdf.py index b7e9cd1c..0ff912b8 100644 --- a/langchain/document_loaders/pdf.py +++ b/langchain/document_loaders/pdf.py @@ -156,6 +156,7 @@ class PyMuPDFLoader(BasePDFLoader): page_content=page.get_text(**kwargs).encode("utf-8"), metadata=dict( { + "source": file_path, "file_path": file_path, "page_number": page.number + 1, "total_pages": len(doc),