Update app.py

app.py

@@ -247,15 +247,20 @@ def extract_main_body_llm(text: str) -> str:
     """
     prompt = f"""
 Extract ONLY the main body text from the following passage.
+
 Rules:
-- Completely EXCLUDE titles, headings, chapter labels, author names, source information,
-  footnotes, annotations, introductions, and any non-body content.
-- Preserve the original paragraph structure of the main text.
+- Completely EXCLUDE titles, headings, chapter labels, section numbers,
+  author names, source information, footnotes, annotations, and introductions.
+- Treat verse numbers, line numbers, and numbering markers (e.g., "001:001") as non-body content and remove them.
+- Do NOT rewrite, paraphrase, summarize, or otherwise modify the text content.
+- Preserve the original paragraph structure of the main body text.
+  (Do not treat line breaks caused by formatting or verse layout as paragraph breaks.)
 - Insert exactly ONE blank line between paragraphs.
 - Do NOT create new section breaks, chapter divisions, or headings.
 - Output only the extracted main body text.
 - Do not include explanations, comments, or metadata.
 - Do not include [TEXT START] and [TEXT END] in the output.
+
 [TEXT START]
 {text}
 [TEXT END]