Repository navigation
Implement or remove the unused Request.skip_navigation flag #2311
Description
CrawleeRequestData.skip_navigation (_request.py) is declared, persisted and serialized as skipNavigation, but nothing in src/crawlee reads it. Setting it has no effect.
Either implement it or remove it in 2.0. JS v4 implements it: the crawler skips opening the page and navigating, runs the handler, and accessing navigation-only context members (the response, the loaded URL) raises a NavigationSkippedError.
This will be a breaking change.
JS references: apify/crawlee#3107, apify/crawlee#3381
✍️ Drafted by Claude Code
Activity
janbuchar commentedon Oct 14, 2026
Implement it. It isn't breaking. Because it's not possible to make it type-safe, we made accessing context properties that are only present after navigation throw on runtime so that you don't have to handle
undefinedfields in the 99% use cases where you don't useskipNavigation.enriquealonso01 commentedon Oct 16, 2026
Picking this up per janbuchar's direction — PR incoming.
Approach mirrors JS SDK v4 (apify/crawlee#3107, #3381):
BasicCrawlerroutesskip_navigationrequests around the context pipeline and calls the router directly; navigation-onlyrequest.loaded_urlraises a new publicNavigationSkippedError; other attribute reads/writes pass through so router state updates still work. The existingskipNavigationserialization is unchanged.