From ae22d053c7aa93ce019ec6b360eafec166a97ab1 Mon Sep 17 00:00:00 2001 From: lih Date: Thu, 20 Nov 2025 14:30:45 +0800 Subject: [PATCH] Fix: missing parameters in by_plaintext method for PDF naive mode --- rag/app/naive.py | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/rag/app/naive.py b/rag/app/naive.py index 49dca17af..099e2f6fd 100644 --- a/rag/app/naive.py +++ b/rag/app/naive.py @@ -111,7 +111,13 @@ def by_tcadp(filename, binary=None, from_page=0, to_page=100000, lang="Chinese", def by_plaintext(filename, binary=None, from_page=0, to_page=100000, callback=None, **kwargs): - if kwargs.get("layout_recognizer", "") == "Plain Text": + parser_config = kwargs.get("parser_config", {}) + layout_recognizer = parser_config.get("layout_recognize", "DeepDOC") + + if isinstance(layout_recognizer, bool): + layout_recognizer = "DeepDOC" if layout_recognizer else "Plain Text" + + if layout_recognizer == "Plain Text": pdf_parser = PlainParser() else: vision_model = LLMBundle(kwargs["tenant_id"], LLMType.IMAGE2TEXT, llm_name=kwargs.get("layout_recognizer", ""), lang=kwargs.get("lang", "Chinese"))