Rene Fichtmueller 2e852e0a2f fix(scrapers): replace bot User-Agents with Chrome UA + disable dead domain
- 16 commercial scrapers: replace TIP-Bot/1.0 with Chrome/120 UA
  (GBICS confirmed returning 0 bytes for bot UA, Chrome UA returns 200KB)
- gbics.ts: fix User-Agent (was returning empty HTML, now returns products)
- optictransceiver.ts: disable — domain repurposed as plant shop (2026-04-06)
  Alocasia Regal Shield is not a transceiver.
2026-04-06 02:17:50 +02:00

238 lines
9.0 KiB
TypeScript

/**
* Fluxlight Scraper — US-based compatible transceiver vendor
*
* www.fluxlight.com — BigCommerce, server-rendered HTML with real prices.
* ~144+ products across 6 pages. Uses pagination via ?page=N.
*
* Rate limited: 1 req/2sec.
*/
import { pool, findOrCreateScrapedTransceiver, ensureVendor, upsertPriceObservation } from "../utils/db";
import { contentHash } from "../utils/hash";
const BASE = "https://fluxlight.com";
const CATALOG_PATH = "/transceivers/";
const HEADERS = {
"User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
Accept: "text/html,application/xhtml+xml",
};
interface Product {
partNumber: string;
name: string;
url: string;
price?: number;
formFactor: string;
speed: string;
speedGbps: number;
reachLabel?: string;
reachMeters?: number;
fiberType?: string;
wavelength?: string;
}
function sleep(ms: number): Promise<void> {
return new Promise((resolve) => setTimeout(resolve, ms));
}
function detectFormFactor(text: string): { formFactor: string; speed: string; speedGbps: number } {
const lower = text.toLowerCase();
if (lower.includes("osfp") && !lower.includes("qsfp")) return { formFactor: "OSFP", speed: "400G", speedGbps: 400 };
if (lower.includes("qsfp-dd")) return { formFactor: "QSFP-DD", speed: "400G", speedGbps: 400 };
if (lower.includes("qsfp28")) return { formFactor: "QSFP28", speed: "100G", speedGbps: 100 };
if (lower.includes("qsfp+") || lower.includes("qsfp plus")) return { formFactor: "QSFP+", speed: "40G", speedGbps: 40 };
if (lower.includes("sfp56")) return { formFactor: "SFP56", speed: "50G", speedGbps: 50 };
if (lower.includes("sfp28") || lower.includes("25g")) return { formFactor: "SFP28", speed: "25G", speedGbps: 25 };
if (lower.includes("sfp+") || lower.includes("10gbase") || lower.includes("10g")) return { formFactor: "SFP+", speed: "10G", speedGbps: 10 };
if (lower.includes("xfp")) return { formFactor: "XFP", speed: "10G", speedGbps: 10 };
if (lower.includes("1000base") || lower.includes("1g")) return { formFactor: "SFP", speed: "1G", speedGbps: 1 };
if (lower.includes("sfp") && !lower.includes("qsfp")) return { formFactor: "SFP", speed: "1G", speedGbps: 1 };
return { formFactor: "SFP+", speed: "10G", speedGbps: 10 };
}
function detectReach(text: string): { label: string; meters: number } | undefined {
const patterns: [RegExp, string, number][] = [
[/\b80\s*km\b/i, "80km", 80000],
[/\b40\s*km\b/i, "40km", 40000],
[/\b20\s*km\b/i, "20km", 20000],
[/\b10\s*km\b/i, "10km", 10000],
[/\b2\s*km\b/i, "2km", 2000],
[/\b550\s*m\b/i, "550m", 550],
[/\b500\s*m\b/i, "500m", 500],
[/\b300\s*m\b/i, "300m", 300],
[/\b100\s*m\b/i, "100m", 100],
[/\bLR4\b/, "10km", 10000],
[/\bLR\b/, "10km", 10000],
[/\bER4?\b/, "40km", 40000],
[/\bZR4?\b/, "80km", 80000],
[/\bSR4?\b/, "300m", 300],
[/\bDR4?\b/, "500m", 500],
[/\bFR4?\b/, "2km", 2000],
];
for (const [regex, label, meters] of patterns) {
if (regex.test(text)) return { label, meters };
}
return undefined;
}
function detectFiber(text: string): string {
if (/single.?mode|smf|[^a-z]lx[^a-z]|[^a-z]lr[^a-z]|[^a-z]er[^a-z]|[^a-z]zr[^a-z]|bidi|cwdm|dwdm/i.test(text)) return "SMF";
if (/multi.?mode|mmf|[^a-z]sx[^a-z]|[^a-z]sr[^a-z]/i.test(text)) return "MMF";
if (/copper|dac|twinax|rj45|base-t|catx/i.test(text)) return "Copper";
return "";
}
function detectWavelength(text: string): string {
const match = text.match(/(\d{3,4})\s*nm/i);
if (match) return match[1];
return "";
}
function parseProductList(html: string): Product[] {
const products: Product[] = [];
// BigCommerce product card pattern: product link + price
// Pattern: <a href="https://www.fluxlight.com/PARTNUM-FL/">Product Name</a> ... $29.99
const productRegex = /href="(https?:\/\/(?:www\.)?fluxlight\.com\/[^"]*-FL\/)"[^>]*>\s*([^<]{10,})<\/a>/gi;
let match;
while ((match = productRegex.exec(html)) !== null) {
const url = match[1];
const name = match[2].trim();
if (name.length < 10 || name.length > 200) continue;
// Look for price in surrounding context — BigCommerce uses data-product-price-without-tax
const context = html.slice(Math.max(0, match.index - 300), match.index + 1200);
const priceMatch = context.match(/data-product-price-without-tax[^>]*>\s*\$([\d,]+\.?\d{0,2})/)
|| context.match(/class="price price--withoutTax">\s*\$([\d,]+\.?\d{0,2})/)
|| context.match(/\$\s*([\d,]+\.?\d{0,2})/);
const price = priceMatch ? parseFloat(priceMatch[1].replace(",", "")) : undefined;
const ff = detectFormFactor(name);
const reach = detectReach(name);
const partNum = url.split("/").filter(Boolean).pop() || name.replace(/\s+/g, "-").slice(0, 80);
products.push({
partNumber: partNum,
name,
url,
price: price && price > 0 && price < 50000 ? price : undefined,
...ff,
reachLabel: reach?.label,
reachMeters: reach?.meters,
fiberType: detectFiber(name),
wavelength: detectWavelength(name),
});
}
// Fallback: broader link pattern
if (products.length === 0) {
const simpleRegex = /href="(https?:\/\/(?:www\.)?fluxlight\.com\/[^"]+)"[^>]*>([^<]{10,}(?:SFP|QSFP|XFP|Base)[^<]*)<\/a>/gi;
while ((match = simpleRegex.exec(html)) !== null) {
const url = match[1];
const name = match[2].trim();
if (products.find((p) => p.url === url)) continue;
const context = html.slice(match.index, match.index + 500);
const priceMatch = context.match(/\$\s*([\d,]+\.?\d{0,2})/);
const price = priceMatch ? parseFloat(priceMatch[1].replace(",", "")) : undefined;
const ff = detectFormFactor(name);
const reach = detectReach(name);
products.push({
partNumber: url.split("/").filter(Boolean).pop() || name.replace(/\s+/g, "-").slice(0, 80),
name, url,
price: price && price > 0 && price < 50000 ? price : undefined,
...ff,
reachLabel: reach?.label, reachMeters: reach?.meters,
fiberType: detectFiber(name), wavelength: detectWavelength(name),
});
}
}
const seen = new Set<string>();
return products.filter((p) => {
if (seen.has(p.url)) return false;
seen.add(p.url);
return true;
});
}
/** Detect max page by probing — page 1 may not have pagination links */
const MAX_PAGES = 6;
async function fetchPage(url: string): Promise<string> {
const resp = await fetch(url, { headers: HEADERS, signal: AbortSignal.timeout(30000) });
if (!resp.ok) throw new Error(`HTTP ${resp.status} for ${url}`);
return resp.text();
}
export async function scrapeFluxlight(): Promise<void> {
console.log("=== Fluxlight Scraper Starting ===\n");
const vendorId = await ensureVendor("Fluxlight", "compatible", "https://fluxlight.com", "https://www.fluxlight.com/transceivers/");
let allProducts: Product[] = [];
for (let page = 1; page <= MAX_PAGES; page++) {
try {
const url = page === 1 ? BASE + CATALOG_PATH : `${BASE}${CATALOG_PATH}?page=${page}`;
const html = await fetchPage(url);
const pageProducts = parseProductList(html);
allProducts.push(...pageProducts);
console.log(` Page ${page}: ${pageProducts.length} products`);
if (pageProducts.length === 0) {
console.log(` Empty page ${page}, continuing...`);
}
if (page < MAX_PAGES) await sleep(2000);
} catch (err) {
console.warn(` Page ${page} failed: ${(err as Error).message}`);
}
}
// Dedupe
const seen = new Set<string>();
allProducts = allProducts.filter((p) => {
if (seen.has(p.url)) return false;
seen.add(p.url);
return true;
});
console.log(`\nTotal unique products: ${allProducts.length}`);
let totalProducts = 0;
let priceUpdates = 0;
for (const product of allProducts) {
try {
const txId = await findOrCreateScrapedTransceiver({
partNumber: product.partNumber, vendorId,
formFactor: product.formFactor, speedGbps: product.speedGbps,
speed: product.speed, reachMeters: product.reachMeters,
reachLabel: product.reachLabel, fiberType: product.fiberType,
wavelengths: product.wavelength, category: "DataCenter",
});
if (product.price && product.price > 0) {
const hash = contentHash(JSON.stringify({ price: product.price, part: product.partNumber }));
const updated = await upsertPriceObservation({
transceiverId: txId, sourceVendorId: vendorId,
price: product.price, currency: "USD",
stockLevel: "in_stock", url: product.url, contentHash: hash,
});
if (updated) priceUpdates++;
}
totalProducts++;
} catch (err) {
console.warn(` Error: ${(err as Error).message.slice(0, 80)}`);
}
}
console.log(`\n=== Fluxlight Complete: ${totalProducts} products, ${priceUpdates} prices ===`);
}
if (require.main === module) {
scrapeFluxlight()
.then(() => pool.end())
.catch((err) => { console.error("Fatal:", err); pool.end(); process.exit(1); });
}