# Changelog of MercadoLibre Search Scraper (`parsebird/mercadolibre-search-scraper`) Actor

- **URL**: https://apify.com/parsebird/mercadolibre-search-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/parsebird/mercadolibre-search-scraper.md

## Changelog

### v1.0.5 — 2026-08-29

- Block image, media, and font downloads in the browser — all product data
  comes from a JSON blob in the page HTML, so this cuts residential-proxy
  transfer per run by ~25–70% with no change to the output
- Detect hard challenge pages (login wall / CAPTCHA) by URL and rotate the
  proxy session immediately instead of polling a dead page for 40 seconds
- Shorter page-hydration poll window (12s) and fewer retries per page (3)
- Added a wall-clock time budget: the crawl stops and finishes cleanly well
  before the platform timeout
- Circuit breaker: stop after two targets in a row return nothing (bot wall)
- Runs that save zero products now finish as FAILED with an explanation
  instead of SUCCEEDED with an empty dataset

### v1.0 — 2026-08-22

- Initial release
- Keyword search and direct results-URL scraping across five MercadoLibre marketplaces: Colombia, Peru, Uruguay, Ecuador, Dominican Republic
- Filters: price range, condition (new/used), free shipping, official store, sort (relevance, lowest/highest price)
- Extracts title, price, discount, installment text, shipping badge, seller/official-store status, sponsored flag, rating, category, condition, and image for each listing
- Automatic pagination with duplicate-safe collection