Create snapshots on web.archive.org for bookmarks (#150)

* Implement initial background tasks concept

* fix property reference

* update requirements.txt

* simplify bookmark null check

* improve web archive url display

* add background tasks test

* add basic supervisor setup

* schedule missing snapshot creation on login

* remove task locks and clear task history before starting background task processor

* batch create snapshots after import

* fix script reference in supervisord.conf

* add option to disable background tasks

* restructure feature overview

bookmarks/services/bookmarks.py

@@ -6,6 +6,7 @@ from django.utils import timezone
 from bookmarks.models import Bookmark, parse_tag_string
 from bookmarks.services.tags import get_or_create_tags
 from bookmarks.services.website_loader import load_website_metadata
+from bookmarks.services import tasks
 
 
 def create_bookmark(bookmark: Bookmark, tag_string: str, current_user: User):
@@ -27,10 +28,16 @@ def create_bookmark(bookmark: Bookmark, tag_string: str, current_user: User):
     # Update tag list
     _update_bookmark_tags(bookmark, tag_string, current_user)
     bookmark.save()
+    # Create snapshot on web archive
+    tasks.create_web_archive_snapshot(bookmark.id, False)
+
     return bookmark
 
 
 def update_bookmark(bookmark: Bookmark, tag_string, current_user: User):
+    # Detect URL change
+    original_bookmark = Bookmark.objects.get(id=bookmark.id)
+    has_url_changed = original_bookmark.url != bookmark.url
     # Update website info
     _update_website_metadata(bookmark)
     # Update tag list
@@ -38,6 +45,10 @@ def update_bookmark(bookmark: Bookmark, tag_string, current_user: User):
     # Update dates
     bookmark.date_modified = timezone.now()
     bookmark.save()
+    # Update web archive snapshot, if URL changed
+    if has_url_changed:
+        tasks.create_web_archive_snapshot(bookmark.id, True)
+
     return bookmark
 
 

bookmarks/services/importer.py

@@ -5,6 +5,7 @@ from django.contrib.auth.models import User
 from django.utils import timezone
 
 from bookmarks.models import Bookmark, parse_tag_string
+from bookmarks.services import tasks
 from bookmarks.services.parser import parse, NetscapeBookmark
 from bookmarks.services.tags import get_or_create_tags
 from bookmarks.utils import parse_timestamp
@@ -38,6 +39,9 @@ def import_netscape_html(html: str, user: User):
             logging.exception('Error importing bookmark: ' + shortened_bookmark_tag_str)
             result.failed = result.failed + 1
 
+    # Create snapshots for newly imported bookmarks
+    tasks.schedule_bookmarks_without_snapshots(user.id)
+
     return result
 
 

bookmarks/services/tasks.py

@@ -0,0 +1,62 @@
+import logging
+
+import waybackpy
+from background_task import background
+from django.conf import settings
+from django.contrib.auth import get_user_model
+from waybackpy.exceptions import WaybackError
+
+from bookmarks.models import Bookmark
+
+logger = logging.getLogger(__name__)
+
+
+def when_background_tasks_enabled(fn):
+    def wrapper(*args, **kwargs):
+        if settings.LD_DISABLE_BACKGROUND_TASKS:
+            return
+        return fn(*args, **kwargs)
+
+    # Expose attributes from wrapped TaskProxy function
+    attrs = vars(fn)
+    for key, value in attrs.items():
+        setattr(wrapper, key, value)
+
+    return wrapper
+
+
+@when_background_tasks_enabled
+@background()
+def create_web_archive_snapshot(bookmark_id: int, force_update: bool):
+    try:
+        bookmark = Bookmark.objects.get(id=bookmark_id)
+    except Bookmark.DoesNotExist:
+        return
+
+    # Skip if snapshot exists and update is not explicitly requested
+    if bookmark.web_archive_snapshot_url and not force_update:
+        return
+
+    logger.debug(f'Create web archive link for bookmark: {bookmark}...')
+
+    wayback = waybackpy.Url(bookmark.url)
+
+    try:
+        archive = wayback.save()
+    except WaybackError as error:
+        logger.exception(f'Error creating web archive link for bookmark: {bookmark}...', exc_info=error)
+        raise
+
+    bookmark.web_archive_snapshot_url = archive.archive_url
+    bookmark.save()
+    logger.debug(f'Successfully created web archive link for bookmark: {bookmark}...')
+
+
+@when_background_tasks_enabled
+@background()
+def schedule_bookmarks_without_snapshots(user_id: int):
+    user = get_user_model().objects.get(id=user_id)
+    bookmarks_without_snapshots = Bookmark.objects.filter(web_archive_snapshot_url__exact='', owner=user)
+
+    for bookmark in bookmarks_without_snapshots:
+        create_web_archive_snapshot(bookmark.id, False)