feat: add price scraper with JSON-LD/meta/regex fallback

- Implement extractPrice() to extract price from HTML with three fallback methods:
  1. JSON-LD Product schema (json-ld)
  2. og:price:amount meta tag (meta)
  3. Euro symbol regex pattern (regex)
- Implement fetchAndExtractPrice() for fetching URLs and extracting prices
- Includes helper functions for parsing and normalizing price values
- Add comprehensive test suite with fixtures for each extraction method
- All tests passing (14/14)

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013gwc3aCiyQxmCWQke8RvEz

src/scraper.js

@@ -0,0 +1,94 @@
+const cheerio = require('cheerio');
+
+function roundMoney(n) {
+  return Math.round(n * 100) / 100;
+}
+
+function parsePriceValue(raw) {
+  if (raw == null) return null;
+  const cleaned = String(raw).trim().replace(',', '.');
+  const value = parseFloat(cleaned);
+  return Number.isFinite(value) ? roundMoney(value) : null;
+}
+
+function searchJsonLdNode(node) {
+  if (Array.isArray(node)) {
+    for (const item of node) {
+      const price = searchJsonLdNode(item);
+      if (price != null) return price;
+    }
+    return null;
+  }
+  if (node && typeof node === 'object') {
+    const type = node['@type'];
+    const isProduct = type === 'Product' || (Array.isArray(type) && type.includes('Product'));
+    if (isProduct && node.offers) {
+      const offers = Array.isArray(node.offers) ? node.offers[0] : node.offers;
+      const price = parsePriceValue(offers && offers.price);
+      if (price != null) return price;
+    }
+    if (Array.isArray(node['@graph'])) {
+      const price = searchJsonLdNode(node['@graph']);
+      if (price != null) return price;
+    }
+  }
+  return null;
+}
+
+function findPriceInJsonLd($) {
+  const scripts = $('script[type="application/ld+json"]').toArray();
+  for (const el of scripts) {
+    let data;
+    try {
+      data = JSON.parse($(el).contents().text());
+    } catch {
+      continue;
+    }
+    const price = searchJsonLdNode(data);
+    if (price != null) return price;
+  }
+  return null;
+}
+
+function findPriceInMeta($) {
+  const selectors = ['meta[property="product:price:amount"]', 'meta[property="og:price:amount"]'];
+  for (const selector of selectors) {
+    const price = parsePriceValue($(selector).attr('content'));
+    if (price != null) return price;
+  }
+  return null;
+}
+
+function findPriceInText($) {
+  const text = $('body').text();
+  const match = text.match(/€\s?(\d+[.,]\d{2})/);
+  return match ? parsePriceValue(match[1]) : null;
+}
+
+function extractPrice(html) {
+  const $ = cheerio.load(html);
+  let price = findPriceInJsonLd($);
+  if (price != null) return { price, method: 'json-ld' };
+  price = findPriceInMeta($);
+  if (price != null) return { price, method: 'meta' };
+  price = findPriceInText($);
+  if (price != null) return { price, method: 'regex' };
+  return { price: null, method: null };
+}
+
+async function fetchAndExtractPrice(url, { fetchImpl = fetch } = {}) {
+  let response;
+  try {
+    response = await fetchImpl(url);
+  } catch (err) {
+    return { price: null, method: null, error: err };
+  }
+  if (!response.ok) {
+    return { price: null, method: null, error: new Error(`HTTP ${response.status}`) };
+  }
+  const html = await response.text();
+  const { price, method } = extractPrice(html);
+  return { price, method, error: null };
+}
+
+module.exports = { extractPrice, fetchAndExtractPrice, parsePriceValue };