Rakuma Scraper – Rakuten Flea Market Japan (楽天ラクマ)
Pricing
from $2.00 / 1,000 results
Scrape Rakuma (fril.jp, 楽天ラクマ) — Rakuten's flea-market app, Japan's #2 C2C marketplace: fashion, brand bags, sneakers, hobby. Search by keyword, brand, category, price, condition, sold-out; get normalized size, condition rank, seller and images. JP Data Works series.
Simple keyword searches, one per line — brand names, item types, model numbers, Japanese or English (e.g. KAPITAL, リーバイス 501, Porter タンカー). Each keyword is one search and inherits the defaults below. Optional when searches is set.
[]Advanced searches as JSON objects. Every field is optional, but each search needs at least one of keyword, brand, brandId, categoryPath, categoryId.
keyword — free text (site keyword search).brand — brand name as listed on Rakuma (e.g. KAPITAL, COMME des GARCONS); resolved to the site's brand id through one keyword search. brandId — that id directly (1785 = KAPITAL).categoryPath — mens, ladies, kids, or a path such as mens > jackets, メンズ > パンツ > デニム/ジーンズ, ladies > バッグ; categoryId — the site's category id (542 = メンズ > ジャケット/アウター > Gジャン/デニムジャケット).priceMinJpy / priceMaxJpy — price bounds (client-side; the site's price filter is robots-disallowed).conditionMin — minimum condition rank 2–6 (6 = 新品、未使用, 5 = 未使用に近い, 4 = 目立った傷や汚れなし, 3 = やや傷や汚れあり, 2 = 傷や汚れあり). Needs the item page, so it turns on enrichDetail.status — on_sale (default), sold (sold-out listings only — price research), all.excludeWords — drop items whose title contains any of these (client-side).sellerId — keep only one seller's items (client-side).A keyword combined with a brand or category runs the keyword search and filters the cards by brand / category client-side (the site's brand_id / category_id query parameters are robots-disallowed). Without a keyword the brand page, brand × category page or category page is crawled directly.
Values set here override the defaults below.
[]items (default): one dataset row per listing. summary: one market-summary row per search instead of the listings — sample size, price min / p25 / median / mean / p75 / max, price histogram, status / condition-rank / size / brand / category breakdowns, cheapest listings — billed once per search as the summary event (the sampled listings are not billed). both: listings plus the summary row.
Applied to every search without its own status. sold returns sold-out listings with their last price (price research); the site shows them in the same listing.
Lower price bound for every search without its own priceMinJpy (client-side).
Upper price bound for every search without its own priceMaxJpy (client-side).
Only listings with this condition rank or better: 6 新品、未使用, 5 未使用に近い, 4 目立った傷や汚れなし, 3 やや傷や汚れあり, 2 傷や汚れあり. The condition is only on the item page, so this turns on item-page enrichment. Leave empty for all.
Drop items whose title contains any of these words (case-insensitive).
[]Stop after this many listings for each keyword / search. The site shows 40 listings per page, up to 100 pages.
Hard cap on listings saved in this run across all searches. Leave empty for no cap (Max total charge still applies).
Save a listing only once when several searches return it.
Open each item page for condition, size, shipping (payer, method, days, prefecture), seller (name, shop, review counts), likes, comments, description, full category path and the full-size gallery. One extra request and one detail-page event per item.
Cap on item-page requests per run. Items beyond the cap are returned without detail fields.
Minimum pause between requests. Please keep ≥ 2000.
Optional Apify Proxy or your own proxies. Not needed under normal use.