From 9f19108a7157db21e331c0b1c3df3f98611c1a80 Mon Sep 17 00:00:00 2001 From: sigoden Date: Tue, 25 Jun 2024 12:55:38 +0800 Subject: refactor: rag chunk contains path metadata (#647) --- src/rag/mod.rs | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'src/rag/mod.rs') diff --git a/src/rag/mod.rs b/src/rag/mod.rs index 3fdad02..ab7c3c7 100644 --- a/src/rag/mod.rs +++ b/src/rag/mod.rs @@ -262,8 +262,10 @@ impl Rag { ); let documents = load(&path, &extension) .with_context(|| format!("Failed to load file at '{path}'"))?; - let documents = - splitter.split_documents(&documents, &SplitterChunkHeaderOptions::default()); + let split_options = SplitterChunkHeaderOptions::default().with_chunk_header(&format!( + "\npath: {path}\n\n\n" + )); + let documents = splitter.split_documents(&documents, &split_options); rag_files.push(RagFile { path, documents }); progress( &progress_tx, -- cgit v1.2.3