Skip to content

V0.2.0

Latest
Latest

Choose a tag to compare

View all tags
@meirk-brd meirk-brd released this 01 Jan 10:58
v0.2.0
8ce8df9

v0.2.0

Highlights

This release expands the Web Scraper API with 44 supported datasets (up from 4), adds zone configuration for all tools, and includes several bug fixes for improved reliability.

What's New

Added

  • 40 new datasets for Web Scraper API covering:

    • E-commerce: Walmart, eBay, Home Depot, Zara, Etsy, Best Buy
    • Social Media: Instagram (profiles, posts, reels, comments), Facebook (posts, marketplace, reviews, events), TikTok (profiles, posts, shop, comments), YouTube (profiles, videos, comments), X/Twitter, Reddit
    • Business: Crunchbase, ZoomInfo, LinkedIn (job listings, posts, people search)
    • Other: Google Maps reviews, Google Shopping, Google Play Store, Apple App Store, GitHub, Yahoo Finance, Reuters, Zillow, Booking.com

  • Zone configuration for SERP API - set custom zones at initialization or per-request

    serp = BrightDataSERP(zone="my_custom_zone")
# Or per-request
serp.invoke({"query": "AI news", "zone": "different_zone"})

  • New input parameters for specialized datasets:

    • keyword for Amazon product search
    • first_name, last_name for LinkedIn people search
    • num_of_reviews for Facebook company reviews
    • num_of_comments for YouTube comments
    • days_limit for Google Maps reviews

Fixed

  • JSON parsing error when API returns newline-delimited JSON (NDJSON) - now returns raw text for proper handling
  • API key logging - removed debug print statement that exposed credentials
  • Empty parameters - filter out None and empty string values from API requests
  • Async polling - handle 202 responses with starting, building, running status during data collection

Changed

  • Package folder renamed from langchain-brightdata to langchain_brightdata for proper Python imports
  • Return type for dataset API changed from parsed JSON to raw text to handle NDJSON responses
  • Improved dataset type validation with Literal types for better LLM tool schema discovery
Assets 2
Loading