# Changelog of Naver Place Scraper - Korea Local Business & Map Data (`haketa/naver-place-scraper`) Actor

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

## Changelog

### 1.1.0 — 2026-11-01

- **Fresh verification run:** scraped **400 unique Naver Place records (0 duplicates) in ~2.3 minutes** across 8 area+keyword queries (`강남 맛집`, `홍대 카페`, `성수 카페`, `부산 맛집`, `제주 카페`, `이태원 맛집`, `명동 맛집`, `연남동 맛집`) via KR residential proxies. Pagination confirmed working — ~50–59 places per query across multiple pages, de-duplicated by `placeId`, stopping cleanly at the `maxItems` budget.
- **Field coverage confirmed:** identity + location fields at **100%** (`placeId`, `name`, `category`, `businessCategory`, `roadAddress`, `address`, `commonAddress`, `fullAddress`, `latitude`, `longitude`, `placeUrl`, `mobileUrl`); review counts, `newBusinessHours`, `hasNPay` and `imageCount` at **100%**; `imageUrl`/`imageUrls`/`totalReviewCount` ~99.8%; `options` 96%; `phone` 92%. Data-dependent fields (`visitorReviewScore`, `landlinePhone`, `michelinGuide`) populate only when the business exposes them — verified as expected Naver behavior, not a gap.
- **Quality:** no value bugs found — no `[object Object]`, all image/place URLs absolute `https://`, correct types (float coordinates, integer review counts, boolean `hasBooking`/`hasNPay`, object `newBusinessHours`, array `imageUrls`), and the `totalReviewCount` sum-fallback works for places without an explicit total.
- **README refresh:** replaced the Output example with a real record from this fresh run, updated the field-reference coverage note with measured per-field percentages, and added a verified pagination/scale benchmark.
- **Added `CHANGELOG.md`** (this file).