From c36fceb1353dd5d009d94cf90173f08944e7b6a4 Mon Sep 17 00:00:00 2001
From: linearcombination <4829djaskdfj@gmail.com>
Date: Wed, 23 Oct 2024 10:30:45 -0700
Subject: [PATCH] Preserve text formatted as poetry in STET

---
 backend/document/stet/stet.py | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/backend/document/stet/stet.py b/backend/document/stet/stet.py
index 9700a30f..5af4afac 100644
--- a/backend/document/stet/stet.py
+++ b/backend/document/stet/stet.py
@@ -71,9 +71,12 @@ def split_chapter_into_verses(chapter: USFMChapter) -> dict[str, str]:
         verse_number = re.search(r'<sup class="versemarker">(\d+)</sup>', verse_span)
         if verse_number:
             verse_number_ = verse_number.group(1)
-            # Remove all <sup> and <div> tags and their content from the verse text
+            # Remove all <sup> tags and their content from the verse text
             verse_text = re.sub(r"<sup.*?>.*?</sup>", "", verse_span)
-            verse_text = re.sub(r"<div.*?>.*?</div>", "", verse_text)
+            # Remove <div> tags that do not have class matching "poetry-<integer>"
+            verse_text = re.sub(
+                r"<div(?!.*class=\"poetry-(\d+)\").*?>.*?</div>", "", verse_text
+            )
             # Remove the remaining HTML tags and strip extra spaces
             verse_text = re.sub(r"<.*?>", "", verse_text).strip()
             logger.debug("verse_number: %s, verse_text: %s", verse_number_, verse_text)