From 535a60233ad07c05cd4451a7bd26fbdabde4c1ec Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Z=C3=BCbeyde=20Civelek?= <zubeydeecivelek@gmail.com>
Date: Mon, 16 Feb 2026 13:36:55 +0100
Subject: [PATCH] fix(serializer): do not remove HTML tags for titles

Co-authored-by: Cursor <cursoragent@cursor.com>
---
 cds/modules/records/serializers/json.py |  8 +-------
 tests/unit/test_records.py              |  2 +-
 tests/unit/test_serializer.py           | 17 +++++------------
 3 files changed, 7 insertions(+), 20 deletions(-)

diff --git a/cds/modules/records/serializers/json.py b/cds/modules/records/serializers/json.py
index ef5a103c7..e4843faaf 100644
--- a/cds/modules/records/serializers/json.py
+++ b/cds/modules/records/serializers/json.py
@@ -31,7 +31,7 @@
     has_read_record_eos_path_permission,
     has_read_record_permission,
 )
-from ..utils import HTMLTagRemover, parse_video_chapters, remove_html_tags
+from ..utils import HTMLTagRemover, parse_video_chapters
 from marshmallow_utils.html import sanitize_html, ALLOWED_HTML_ATTRS, ALLOWED_CSS_STYLES
 
 CUSTOM_ALLOWED_ATTRS = {
@@ -61,9 +61,6 @@ def _sanitize_metadata(self, metadata):
             if "title" in metadata and "title" in metadata["title"]:
                 title = metadata["title"]["title"]
                 title = self.html_tag_remover.unescape(title)
-                metadata["title"]["title"] = remove_html_tags(
-                    self.html_tag_remover, title
-                )
 
             if "description" in metadata:
                 description = metadata["description"]
@@ -79,9 +76,6 @@ def _sanitize_metadata(self, metadata):
                     if "title" in t and "title" in t["title"]:
                         t_title = t["title"]["title"]
                         t_title = self.html_tag_remover.unescape(t_title)
-                        t["title"]["title"] = remove_html_tags(
-                            self.html_tag_remover, t_title
-                        )
 
                     if "description" in t:
                         t_desc = t["description"]
diff --git a/tests/unit/test_records.py b/tests/unit/test_records.py
index 626156cc0..ecc55476c 100644
--- a/tests/unit/test_records.py
+++ b/tests/unit/test_records.py
@@ -158,7 +158,7 @@ def test_records_rest(
         res = client.get(url2, headers=json_headers)
         assert res.status_code == 200
         video_dict = json.loads(res.data.decode("utf-8"))
-        assert video_dict["metadata"]["title"]["title"] == "My english title"
+        assert video_dict["metadata"]["title"]["title"] == "My <b>english</b> title"
         expect_desc = "in tempor reprehenderit enim eiusmod <b><i>html</i></b>"
         assert video_dict["metadata"]["description"] == expect_desc
 
diff --git a/tests/unit/test_serializer.py b/tests/unit/test_serializer.py
index 9c39f94b8..3e61446a9 100644
--- a/tests/unit/test_serializer.py
+++ b/tests/unit/test_serializer.py
@@ -191,20 +191,13 @@ def test_cds_json_serializer_sanitization(video_record_metadata):
     assert 'Safe content' in description
     # Keep safe HTML tags like <b>
     assert '<b>bold</b>' in description
-    
-    # Remove everything in title
+
+    # Title: only unescape, no HTML tag removal
     title = result['metadata']['title']['title']
-    assert '<script>' not in title
-    assert '</script>' not in title
     assert 'Test' in title and 'Title' in title
-    assert '<b>' not in title
-    
-    # --- Translations checks ---
+    assert '<b>bold</b>' in title
+
+    # Translations: descriptions sanitized, titles only unescaped
     translations = result['metadata']['translations']
     for tr in translations:
-        # description
         assert '<script>' not in tr['description']
-        # title
-        assert '<script>' not in tr['title']['title']
-        assert '<b>' not in tr['title']['title']
-