Apply the offset in a SingleFile saved date

SingleFile writes the browser's local time with its offset ("... 18:36:23 GMT-0700").  The parser dropped
the offset and treated the local time as UTC, so an uploaded Archive's download_datetime was off by the
browser's UTC offset, and a Collection reorganize renamed its files with that shifted timestamp.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

modules/archive/models.py

@@ -1,6 +1,7 @@
 import json
 import pathlib
 import re
+from datetime import datetime
 from typing import Iterable, List, Optional
 
 import pytz
@@ -27,6 +28,20 @@ MATCH_URL = re.compile(r'^\s+?url:\s+?(http.*)', re.MULTILINE)
 MATCH_DATE = re.compile(r'^\s+?saved date:\s+?(.*)', re.MULTILINE)
 
 
+def parse_singlefile_saved_date(saved_date: str) -> datetime:
+    """Parse SingleFile's "saved date", which is the browser's local time and its offset, into UTC.
+
+    >>> parse_singlefile_saved_date('Thu Nov 07 2024 18:36:23 GMT-0700 (Mountain Standard Time)')
+    datetime.datetime(2024, 11, 8, 1, 36, 23, tzinfo=<UTC>)
+    """
+    parts = saved_date.split(' ')
+    try:
+        return datetime.strptime(' '.join(parts[:6]), '%a %b %d %Y %H:%M:%S GMT%z').astimezone(pytz.UTC)
+    except ValueError:
+        # No offset; assume GMT.
+        return dates.strpdate(' '.join(parts[:5])).replace(tzinfo=pytz.timezone('GMT'))
+
+
 class Archive(Base, ModelHelper):
     __tablename__ = 'archive'
     __table_args__ = (
@@ -331,11 +346,7 @@ class Archive(Base, ModelHelper):
             if not self.file_group.download_datetime:
                 try:
                     if match := MATCH_DATE.findall(head):
-                        dt = match[0].strip()
-                        dt = ' '.join(dt.split(' ')[:5])
-                        # SingleFile uses GMT.
-                        dt = dates.strpdate(dt).replace(tzinfo=pytz.timezone('GMT'))
-                        self.file_group.download_datetime = dt
+                        self.file_group.download_datetime = parse_singlefile_saved_date(match[0].strip())
                 except Exception as e:
                     logger.error(f'Could not get archive date from singlefile {path}', exc_info=e)
 

modules/archive/test/test_lib.py

@@ -2,7 +2,7 @@ import json
 import pathlib
 import re
 import tempfile
-from datetime import datetime, timedelta
+from datetime import datetime, timedelta, timezone
 from http import HTTPStatus
 from unittest import mock
 
@@ -633,6 +633,29 @@ async def test_refresh_archives_index(test_session, make_files_structure, refres
     assert archive.file_group.indexed is True
 
 
+@pytest.mark.asyncio
+@pytest.mark.parametrize('saved_date,expected', [
+    ('Mon May 16 2022 23:51:35 GMT+0000 (Coordinated Universal Time)', datetime(2022, 5, 16, 23, 51, 35)),
+    # The browser writes its local time; the offset must be applied, not dropped.
+    ('Thu Nov 07 2024 18:36:23 GMT-0700 (Mountain Standard Time)', datetime(2024, 11, 8, 1, 36, 23)),
+    ('Tue Mar 03 2026 09:15:00 GMT+0200 (Eastern European Standard Time)', datetime(2026, 3, 3, 7, 15, 0)),
+])
+async def test_singlefile_saved_date_timezone(test_session, make_files_structure, refresh_files, saved_date, expected):
+    """An Archive's download_datetime is the SingleFile "saved date" converted to UTC."""
+    make_files_structure({
+        'archive/example.com/NA.html': f'''<!DOCTYPE html> <html lang="en"><!--
+ Page saved with SingleFile
+ url: https://example.com
+ saved date: {saved_date}
+--><head><meta charset="utf-8"></head></html>''',
+    })
+
+    await refresh_files()
+
+    archive: Archive = test_session.query(Archive).one()
+    assert archive.file_group.download_datetime == expected.replace(tzinfo=timezone.utc)
+
+
 @pytest.mark.asyncio
 async def test_archive_meta(async_client, test_session, make_files_structure, refresh_files):
     # The start of a typical singlefile html file.