From be751685702500bc8bea470034268bdecb10eb2d Mon Sep 17 00:00:00 2001
From: Andrew Peabody <andrewpeabody@google.com>
Date: Mon, 31 Aug 2026 04:41:46 -0700
Subject: [PATCH] fix(backup): add utf-8 encoding when writing url list files
 (#4345)

---
 .../blueprint/backups/__init__.py             |  4 +-
 changedetectionio/tests/test_backup.py        | 44 ++++++++++++++++++-
 2 files changed, 45 insertions(+), 3 deletions(-)

diff --git a/changedetectionio/blueprint/backups/__init__.py b/changedetectionio/blueprint/backups/__init__.py
index 0d1626945..d3c97a61a 100644
--- a/changedetectionio/blueprint/backups/__init__.py
+++ b/changedetectionio/blueprint/backups/__init__.py
@@ -60,13 +60,13 @@ def create_backup(datastore_path, watches: dict, tags: dict = None):
 
         # Create a list file with just the URLs, so it's easier to port somewhere else in the future
         list_file = "url-list.txt"
-        with open(os.path.join(datastore_path, list_file), "w") as f:
+        with open(os.path.join(datastore_path, list_file), "w", encoding="utf-8") as f:
             for uuid in watches:
                 url = watches[uuid]["url"]
                 f.write("{}\r\n".format(url))
         list_with_tags_file = "url-list-with-tags.txt"
         with open(
-                os.path.join(datastore_path, list_with_tags_file), "w"
+                os.path.join(datastore_path, list_with_tags_file), "w", encoding="utf-8"
         ) as f:
             for uuid in watches:
                 url = watches[uuid].get('url')
diff --git a/changedetectionio/tests/test_backup.py b/changedetectionio/tests/test_backup.py
index 476d2f1ad..00d5a17d2 100644
--- a/changedetectionio/tests/test_backup.py
+++ b/changedetectionio/tests/test_backup.py
@@ -258,4 +258,46 @@ def test_backup_restore_zip_bomb_rejected(client, live_server, measure_memory_us
                 include_watches_replace=True,
             )
     finally:
-        restore_mod._MAX_DECOMPRESSED_BYTES = original_limit
\ No newline at end of file
+        restore_mod._MAX_DECOMPRESSED_BYTES = original_limit
+
+
+def test_backup_with_extended_utf8_urls_and_tags(datastore_path):
+    """Test create_backup handles extended UTF-8 characters in URLs and tags without encoding errors (issue #1805)."""
+    import os
+    import tempfile
+    from unittest.mock import MagicMock
+
+    from changedetectionio.blueprint.backups import create_backup
+
+    with tempfile.TemporaryDirectory() as tmp_dir:
+        watch_mock = MagicMock()
+        watch_mock.data_dir = tmp_dir
+        watch_mock.__getitem__.side_effect = lambda key: (
+            "https://example.com/products/café-öl-日本語-🎉?query=test#section" if key == "url" else None
+        )
+        watch_mock.get.side_effect = lambda key, default=None: (
+            "https://example.com/products/café-öl-日本語-🎉?query=test#section"
+            if key == "url"
+            else (["Größe", "日本語タグ", "🏷️special"] if key == "tags" else default)
+        )
+
+        watches = {"test-uuid-1": watch_mock}
+        # Should create backup without UnicodeEncodeError
+        create_backup(datastore_path=tmp_dir, watches=watches, tags={})
+
+        url_list_path = os.path.join(tmp_dir, "url-list.txt")
+        url_list_tags_path = os.path.join(tmp_dir, "url-list-with-tags.txt")
+
+        assert os.path.isfile(url_list_path)
+        assert os.path.isfile(url_list_tags_path)
+
+        with open(url_list_path, encoding="utf-8") as f:
+            content = f.read()
+            assert "café-öl-日本語-🎉" in content
+
+        with open(url_list_tags_path, encoding="utf-8") as f:
+            content = f.read()
+            assert "café-öl-日本語-🎉" in content
+            assert "Größe" in content
+            assert "日本語タグ" in content
+            assert "🏷️special" in content