Reliable Batching and Cost Controls
This release introduces highly requested TypeScript type-safety for batch scraping responses, ensuring complete clarity when validating bulk operations. We have also hardened billing and scheduling mechanisms to prevent incorrect debits for blocked runs and ensure accurate refund processing. Finally, task cancellation and batch payload validation have been improved for a more robust developer experience.
The Node SDK now fully exposes accepted and rejected task details for bulk scraping. This provides complete TypeScript compile-time safety when inspecting the results of your batch operations.
Credit entitlement checks are now performed before charging balances for scheduled scraping tasks. This ensures you are never charged for a scheduled run that gets denied or blocked during pre-dispatch checks.
Improvements
4Typed batch scraping responses
The Node SDK now fully exposes accepted and rejected task details for bulk scraping. This provides complete TypeScript compile-time safety when inspecting the results of your batch operations.
Documented batch response schemas
The OpenAPI schema now formally defines the terminal HTTP 200 response body for batch scraping requests. This simplifies custom client generation and bulk request integration.
Enhanced TypeScript type discovery
The Node SDK now successfully detects and parses inline or intersection generic types inside raw request helper calls. This ensures your custom type definitions are correctly captured during development.
General bug fixes and improvements
Plus 9 internal improvements for better reliability and performance.
Bug Fixes
4No debits for denied runs
Credit entitlement checks are now performed before charging balances for scheduled scraping tasks. This ensures you are never charged for a scheduled run that gets denied or blocked during pre-dispatch checks.
Reliable scraping task cancellation
Task cancellations are now properly preserved and raised during background billing retry cycles. This ensures that stopped scraping requests immediately cease processing and do not consume your resources.
Accurate concurrent refund tracking
A concurrency lock has been introduced to prevent concurrent refund replays from overwriting newer balance updates. This ensures your account credit balances remain perfectly accurate and synchronized.
Robust batch payload validation
The batch endpoint now gracefully validates and rejects non-object JSON inputs before queue processing. This prevents malformed bulk requests from stalling or halting execution pipelines.
Security
1Secure browser profile isolation
Access restrictions have been tightened to block external access to browser profile directories within served downloads. This secures session-specific data and prevents potential cross-tenant leakage.
Plus 9 internal changes for stability and performance.