Big rewrite. Now pulls data from __NEXT_DATA__ instead of scraping HTML, which gives way more fields and is less likely to break.
New fields added:
reviewId -- Trustpilot's internal review IDreviewUrl -- direct link to the reviewpublishedDate / experienceDate / updatedDate -- three separate dates instead of oneauthorCountry -- ISO country code (GB, US, DE, etc.)authorReviewCount -- how many reviews that person has writtenauthorImageUrl -- profile picture URLverificationLevel -- not just true/false, now shows the actual levelsource -- Organic vs Invitation (huge for sentiment analysis)language -- review language codelikesCount -- upvotes on the review
Business info improvements:
- Now extracts
websiteUrl, phone, email, address when available
- Gets data from
__NEXT_DATA__ first, falls back to JSON-LD
Under the hood:
- 3-tier extraction:
__NEXT_DATA__ > JSON-LD > HTML parsing
- 25 fields per review (up from 14)
- Field names cleaned up to match industry standard (e.g.
title instead of reviewTitle)
First release. Scrapes reviews, ratings, author info, and company replies. Supports star filtering, sorting, and pagination.
