# Changelog of Bing News Scraper (`searchapi/bing-news-scraper`) Actor

- **URL**: https://apify.com/searchapi/bing-news-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/searchapi/bing-news-scraper.md

## Changelog

### \[3.0.1] - 2026-08-29

- Replaced obsolete web-search `ex1` time filters with Bing News' current `interval` controls.
- Added output-side publication-range enforcement so target-side filter drift cannot leak older or undated records into filtered datasets.
- Pinned patched transitive dependencies and added Actor-generation metadata.
- Revalidated search, URL, pagination, no-results, schema, and dataset behavior locally and in Apify Cloud.

### \[3.0.0] - 2026-07-20

#### Added

- Market-aware search URL construction and current Bing numbered pagination.
- `maxPages`, scroll, concurrency, retry, timeout, proxy, headless, and debug controls.
- Source-backed Bing publisher-search links, public high-quality thumbnails, and source dimensions.
- Explicit no-results, response content-type, status, and challenge-page classification.
- Stable sessions, cookies, and Crawlee browser fingerprints.
- Unit, fixture, and per-JSON dataset validation coverage.

#### Changed

- Buffers normalized records, sorts them by page/card order, assigns deterministic positions, and pushes one final dataset batch.
- Uses a focused news record normalizer instead of the generated multi-result normalizer.
- Converts Bing relative publication values such as `2 hours ago`, `1d`, and `11mon` from one run timestamp.
- Uses Bing's high-quality thumbnail as the primary image when the visible image is only an embedded placeholder.
- Updated Apify, Crawlee, Playwright, and the matching Actor Docker image.

#### Fixed

- Enforces `maxItems` globally across pages and concurrent URL-mode requests.
- Deduplicates by Bing article ID or a canonical article URL without title-only collisions.
- Keeps `publishedAtRaw` and `relativeTime` as Bing displayed them and uses one consistent `scrapedAt` value.
- Removed fabricated Google favicon URLs, embedded base64 logos, and misleading publisher-as-author data.
- Aligned all emitted fields with the dataset schema and documented optional fields.

### \[2.0.1] - 2026-06-13

- Verified the original 20-record DOM extraction flow.

### \[2.0.0] - 2026-06-12

- Added Bing IDs, rich aliases, thumbnails, publication metadata, and full-details views.

### \[1.0.0] - 2025-07-01

- Initial release.