Free forever
Runs entirely on Cloudflare Pages free tier. ~1,000 files, each under 25 MiB — there is simply no metered axis to bill.
A Japan zipcode dataset for indie developers and OSS. 120,680 normalized entries are delivered from Cloudflare Pages, and each language SDK lets you take them all the way offline.
The form below hits jpzip.nadai.dev directly from your browser. If CORS clears you get real data; otherwise a built-in demo response.
The protocol is fixed, so each SDK is just a few dozen lines — with identical signatures across languages.
import { lookup, preload } from "@jpzip/jpzip" // 単発検索 — /p/231.json を fetch して結果を抽出 const entry = await lookup("2310017") // → { prefecture: "神奈川県", city: "横浜市中区", // towns: [{ town: "本町", kana: "ホンチョウ" }], ... } // オフライン化 — 全件を SDK 内にキャッシュ await preload({ scope: "all" }) // 以降の lookup() はネットワーク不要 const e2 = await lookup("1500001")
No Workers, no R2, no KV. Even at billions of requests, there is no metered axis to bill.
Runs entirely on Cloudflare Pages free tier. ~1,000 files, each under 25 MiB — there is simply no metered axis to bill.
preload({ scope: "all" }) caches the full dataset into the SDK. After that, every lookup works without network.
No API key needed. CORS allows all origins. Fetch directly from the browser.
GitHub Actions ingests the latest Japan Post data on the 1st and 15th of each month and redeploys the CDN.
KEN_ALL and KEN_ALL_ROME are pre-joined. The MIC municipality code (city_code) is included too.
SDKs are MIT. The data is effectively public domain (Japan Post). Commercial use and redistribution are free.
No signup, no API key, no billing surprises. Install from npm (or your language’s package manager) and run it in the browser, on a server, or at the edge.