shopify-product-scraper/dist/index.js

A lightweight JavaScript library for scraping products from Shopify stores. This package provides a simple and efficient way to extract product data from Shopify websites.

var __async = (__this, __arguments, generator) => {
  return new Promise((resolve, reject) => {
    var fulfilled = (value) => {
      try {
        step(generator.next(value));
      } catch (e) {
        reject(e);
      }
    };
    var rejected = (value) => {
      try {
        step(generator.throw(value));
      } catch (e) {
        reject(e);
      }
    };
    var step = (x) => x.done ? resolve(x.value) : Promise.resolve(x.value).then(fulfilled, rejected);
    step((generator = generator.apply(__this, __arguments)).next());
  });
};

// src/url.ts
var shopifyIndicators = [
  /shopify\.com/i,
  /cdn\.shopify\.com/i,
  /Shopify\.Buy\.SDK/i,
  /var Shopify =/i,
  /{{ '.*' | asset_url }}/i
];
function isValidURL(url) {
  return __async(this, null, function* () {
    try {
      const html = yield fetch(url).then((res) => res.text());
      return shopifyIndicators.some((pattern) => pattern.test(html));
    } catch (e) {
      return false;
    }
  });
}

// src/index.ts
function scrape(_0) {
  return __async(this, arguments, function* (url, { limit = Infinity, onProgress } = {}) {
    var _a;
    const isValid = yield isValidURL(url);
    if (!isValid) return null;
    const data = [];
    let page = 0;
    while (data.length < limit) {
      const pageLimit = Math.min(250, limit - data.length);
      const requestedData = yield fetch(
        `${url}/products.json/?limit=${pageLimit}&page=${page}`
      ).then((res) => res.json());
      if (!((_a = requestedData == null ? void 0 : requestedData.products) == null ? void 0 : _a.length)) break;
      data.push(...requestedData.products);
      if (onProgress)
        onProgress({
          progress: data.length / limit,
          products: data
        });
      page++;
    }
    return data;
  });
}
export {
  scrape
};