From 28194a4012ea15e49960a5a18b95bfe37eeb0770 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Cl=C3=A9mence=20Lesn=C3=A9?= <clemence@lesne.pro>
Date: Sat, 15 Jun 2024 18:45:47 +0200
Subject: [PATCH] quality: Unescape HTML entities in Document Intelligence

---
 persistence/document_intelligence.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/persistence/document_intelligence.py b/persistence/document_intelligence.py
index f4a49f1..62a64fe 100644
--- a/persistence/document_intelligence.py
+++ b/persistence/document_intelligence.py
@@ -13,6 +13,7 @@
     ParagraphRole,
 )
 import math
+import html
 
 
 class DocumentIntelligenceAnalyze(IAnalyze):
@@ -62,8 +63,8 @@ async def analyze(
                     None,  # Third, nothing
                 ),
             )
-            content = res.content
-            title = title_paragraph.content if title_paragraph else None
+            content = html.unescape(res.content)
+            title = html.unescape(title_paragraph.content) if title_paragraph else None
             langs = {lang.locale for lang in res.languages or [] if lang.confidence >= self._config.extract_lang_threshold}
 
         # Return title, content and langs