fix(bikegear): route scraper through residential proxy to bypass Cloudflare 403 on server IPs

Datacenter IPs are blocked by bikegear.in Cloudflare WAF.
Switches fetchHtml from native fetch to axios with BIKEGEAR_PROXY_URL support.
Set BIKEGEAR_PROXY_URL=http://user:pass@host:port in server .env to enable.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

src/business-logic/import-pipeline/sources/bikegear/scraper.js

@@ -3,13 +3,23 @@
  * 2-phase: listing pages → product URLs, then detail pages → full product data.
  *
  * Prices are in INR (Indian Rupees) as scraped from the site.
+ *
+ * Set BIKEGEAR_PROXY_URL (e.g. http://user:pass@host:port) to route through a
+ * residential proxy — required when running from a datacenter IP that bikegear.in
+ * Cloudflare blocks with 403.
  */
 
+const axios = require("axios");
 const { BRANDS } = require("./brands");
 
 const BASE_URL = "https://bikegear.in";
 const LISTING_CONCURRENCY = Number(process.env.BIKEGEAR_LISTING_CONCURRENCY ?? 3);
 const DETAIL_CONCURRENCY = Number(process.env.BIKEGEAR_DETAIL_CONCURRENCY ?? 2);
+const PROXY_URL = process.env.BIKEGEAR_PROXY_URL || null;
+
+if (PROXY_URL) {
+  console.log(`[BIKEGEAR] Using proxy: ${PROXY_URL.replace(/:\/\/[^@]+@/, "://<redacted>@")}`);
+}
 
 const FETCH_HEADERS = {
   "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36",
@@ -31,27 +41,54 @@ function sleep(ms) {
   return new Promise((resolve) => setTimeout(resolve, ms));
 }
 
+/** Build axios proxy config from a proxy URL string. */
+function buildProxyConfig(proxyUrl) {
+  const u = new URL(proxyUrl);
+  const cfg = {
+    host: u.hostname,
+    port: parseInt(u.port, 10),
+    protocol: u.protocol.replace(":", ""),
+  };
+  if (u.username) {
+    cfg.auth = { username: decodeURIComponent(u.username), password: decodeURIComponent(u.password) };
+  }
+  return cfg;
+}
+
 async function fetchHtml(url, attempt = 1) {
   try {
-    const res = await fetch(url, { headers: FETCH_HEADERS });
+    const axiosOpts = {
-
+      headers: FETCH_HEADERS,
-    if (res.status === 429 || res.status >= 500) {
+      timeout: 30000,
-      if (attempt <= 4) {
+      responseType: "text",
-        const wait = attempt * 2000;
+      maxRedirects: 5,
-        console.log(`[BIKEGEAR] RETRY ${url} HTTP ${res.status} — waiting ${wait}ms (${attempt}/4)`);
+    };
-        await sleep(wait);
+    if (PROXY_URL) {
-        return fetchHtml(url, attempt + 1);
+      axiosOpts.proxy = buildProxyConfig(PROXY_URL);
-      }
-      throw new Error(`HTTP ${res.status} after retries`);
     }
 
-    if (!res.ok) throw new Error(`HTTP ${res.status}`);
+    const res = await axios.get(url, axiosOpts);
-    return res.text();
+    return res.data;
   } catch (err) {
-    if (attempt <= 4 && (err.code === "UND_ERR_CONNECT_TIMEOUT" || err.code === "ECONNRESET" || err.message?.includes("fetch failed"))) {
+    const status = err.response?.status;
+
+    if ((status === 429 || status >= 500) && attempt <= 4) {
+      const wait = attempt * 2000;
+      console.log(`[BIKEGEAR] RETRY ${url} HTTP ${status} — waiting ${wait}ms (${attempt}/4)`);
+      await sleep(wait);
+      return fetchHtml(url, attempt + 1);
+    }
+
+    if (status) throw new Error(`HTTP ${status}`);
+
+    const isNetwork = err.code === "ECONNRESET" || err.code === "ECONNREFUSED" ||
+      err.code === "ETIMEDOUT" || err.code === "ENOTFOUND" ||
+      err.message?.includes("timeout") || err.message?.includes("connect");
+    if (isNetwork && attempt <= 4) {
       await sleep(attempt * 2000);
       return fetchHtml(url, attempt + 1);
     }
+
     throw err;
   }
 }