Forward crawler status messages to the run status message #1188

crawlee-python crawlers emit Event.CRAWLER_STATUS with a status line such as "Crawled 120/500 pages, 3 failed requests, desired concurrency 8." on every status interval (source). Nothing in the Python SDK listens to it, so a crawler running inside an Actor never updates the run's status message in Console unless the user wires it up.

The JS SDK subscribes to crawlee's STATUS_MESSAGE event in Actor.init() and forwards each message to the run without logging it again (source). On exit it unsubscribes after the listeners finish, so a late crawler message cannot overwrite the terminal status message (source).

Acceptance criteria

• On the platform, the Actor registers a CRAWLER_STATUS listener during init that updates the run status message, without logging it a second time.
• The listener is removed before the terminal status message is set on exit.
• A failed update logs a warning and does not break the crawler.

Implementation notes

• In _ActorType.__aenter__, after the event manager is entered (_actor.py#L219), register a crawlee.events.Event.CRAWLER_STATUS listener only when is_at_home(). Crawlers emit on the same event manager through service_locator, so no crawler changes are needed.
• The listener calls set_status_message(data.message) (non-terminal). Crawlee already logs the status line, and set_status_message logs only off the platform, so nothing is logged twice. Catch exceptions in the listener and log a warning.
• In __aexit__, remove the listener before finalize() sets the terminal status message (#L295-L300), so a late crawler message can't overwrite it.
• Add unit tests to tests/unit/actor/test_actor_lifecycle.py: an emitted CRAWLER_STATUS event updates the run status message on the platform, nothing happens locally, and no update follows the terminal message.

✍️ Drafted by Claude Code