From f9cc2c2d40c2083549c333064582183c96162e05 Mon Sep 17 00:00:00 2001
From: Bryan Newbold <bnewbold@archive.org>
Date: Tue, 21 Jul 2020 17:05:05 -0700
Subject: strip <em> tags explicitly

---
 fatcat_scholar/schema.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fatcat_scholar/schema.py b/fatcat_scholar/schema.py
index 19f148b..64e9268 100644
--- a/fatcat_scholar/schema.py
+++ b/fatcat_scholar/schema.py
@@ -227,6 +227,7 @@ def scrub_text(raw: str, mimetype: str = None) -> Optional[str]:
     text = re.sub(r"[`‘’‛⸂⸃⸌⸍⸜⸝]", "'", text)
     text = re.sub(r"[„“]|(\'\')|(,,)", '"', text)
     text = re.sub(r"\s+", " ", text).strip()
+    text = text.replace("<em>", "").replace("</em>", "")
 
     # hack to remove abstract prefixes
     for prefix in UNWANTED_ABSTRACT_PREFIXES:
-- 
cgit v1.2.3