# Changelog of OLX Brazil Scraper — Ads, Prices, Location & Seller Data (`haketa/olx-brasil-scraper`) Actor

- **URL**: https://apify.com/haketa/olx-brasil-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/haketa/olx-brasil-scraper.md

## Changelog

### 1.0.0 — 2026-11-01

- **Fresh verification run:** scraped **500 unique OLX Brasil ads** for the keyword `iphone` in **44 seconds** on the default Brazil residential proxy, with **0% duplicates** on `listId` and `url`.
- **Core-field coverage 100%:** `listId`, `title`, `price`, `priceText`, `category`, `categoryId`, `isProfessional`, `state`, `city`, `ddd`, `locationLabel`, `publishedAt`, `imageUrl`, `images`, `attributes` and `url` all populated on every record; `neighbourhood` 96%. Verified `attributes` ships as a structured object and `images` as a proper array (no stringified values).
- **Price intelligence verified:** `oldPrice` captured on 18.8% of ads (all genuine reductions where the previous price exceeds the current one) alongside the separate `priceDropped` badge flag; numeric `price` parsed correctly from the Brazilian `R$ 1.234,56` format with no null-price errors.
- **No value bugs:** scanned the full sample — no `[object Object]`, no scheme-less URLs, no garbled text, correct types throughout. Live `publishedAt` timestamps confirm real-time data.
- **README refreshed** to the gold standard: output example and field reference regrounded in the fresh run, pagination & scale note, expanded FAQ and related-scraper links.
- Confirmed the App Router RSC parser (`self.__next_f.push`) with the legacy `__NEXT_DATA__` fallback, session rotation + warm-up, and `o`-param pagination all working against the live site.