Working Ebay Scraping

.gitea/workflows/deploy.yaml

@@ -38,9 +38,12 @@ jobs:
               git clone https://git.hudsonriggs.systems/HRiggs/Train-ID.git "$APP_DIR"
             fi
             cd "$APP_DIR"
-            git pull origin main
-            # Install Node.js deps and build
-            "$NPM" ci || "$NPM" install
+            # Reset any local changes (e.g., package-lock.json, build artifacts) and sync to origin/main
+            git fetch --prune origin
+            git reset --hard origin/main
+            git clean -fdx
+            # Install Node.js deps and build without modifying lockfile
+            "$NPM" ci --no-audit --no-fund
             "$NPM" run build
             # Ensure systemd service exists and restart
             if systemctl list-unit-files | grep -q "${SERVICE}.service"; then

README.md

@@ -23,6 +23,13 @@ db_port=3306
 db_user=trainid
 db_pass=changeme
 db_name=trainid
+
+# eBay API (optional - falls back to scraping if not configured)
+# Get credentials from https://developer.ebay.com/my/keys
+# Create an app and use the Client ID and Client Secret
+EBAY_CLIENT_ID=YourAppId...
+EBAY_CLIENT_SECRET=YourClientSecret...
+EBAY_SANDBOX=false  # Set to 'true' to use eBay sandbox environment
 ```
 
 ## Local development
@@ -42,6 +49,16 @@ Visit `http://localhost:3000`.
 - GET `/api/export/xlsx` → download XLSX export of inventory with embedded thumbnails
 - DELETE `/api/items/:id` → delete one
 - DELETE `/api/items` → wipe all
+- GET `/api/debug/ebay-prices?sku=...` → debug eBay price lookup (shows API and scraping attempts)
+- POST `/api/prices/update` → update cached prices for all SKUs from eBay
+- GET `/api/price-report` → get price report with item values
+
+### eBay Price Checking
+The system uses eBay's Browse API to search for items by SKU. **Important limitations:**
+- **Browse API only searches active listings**, not sold/completed ones
+- For sold/completed listings, the system falls back to web scraping
+- In sandbox mode, test data is limited - you may see 0 results even if the API is working correctly
+- The API method: `GET /buy/browse/v1/item_summary/search` with query parameter `q` (SKU)
 
 ## Debian 13 (Trixie) LXC install
 These steps assume a fresh Debian 13 LXC and deployment directory `/opt/Train-ID` with a system user `deployuser` that has passwordless sudo for service management.

package.json

@@ -18,8 +18,8 @@
   "license": "ISC",
   "type": "module",
   "dependencies": {
-    "cors": "^2.8.5",
     "cheerio": "^1.0.0",
+    "cors": "^2.8.5",
     "dotenv": "^17.2.3",
     "exceljs": "^4.4.0",
     "express": "^5.1.0",
@@ -27,7 +27,8 @@
     "morgan": "^1.10.1",
     "mysql2": "^3.15.3",
     "openai": "^6.7.0",
-    "pdfkit": "^0.17.2"
+    "pdfkit": "^0.17.2",
+    "puppeteer": "^24.28.0"
   },
   "devDependencies": {
     "@types/cheerio": "^0.22.35",

src/db/client.ts

@@ -151,6 +151,23 @@ export const db = {
     }
     return Array.from(set);
   },
+  async listDistinctSkuManufacturers() {
+    await getReady();
+    // Get distinct SKU + manufacturer pairs
+    const [rows] = await pool.query(`
+      select distinct sku, manufacturer 
+      from items 
+      where sku is not null and sku <> '' and manufacturer is not null and manufacturer <> ''
+    `);
+    const pairs: Array<{ sku: string; manufacturer: string }> = [];
+    for (const r of rows as any[]) {
+      const s = normalizeSku(r.sku as string);
+      if (s && r.manufacturer) {
+        pairs.push({ sku: s, manufacturer: r.manufacturer });
+      }
+    }
+    return pairs;
+  },
   async upsertSkuPrice(sku: string, price: number | null, currency: string = 'USD', source: string = 'ebay') {
     await getReady();
     const s = normalizeSku(sku);

src/public/index.html

@@ -86,6 +86,8 @@
           <table id="skuPrices">
             <thead>
               <tr>
+                <th>ID</th>
+                <th>Name</th>
                 <th>SKU</th>
                 <th>Price (USD)</th>
                 <th class="hide-sm">Updated</th>
@@ -203,7 +205,9 @@
           const tr = document.createElement('tr');
           const title = s.description || '';
           tr.innerHTML = `
-            <td title="${title.replaceAll('"', '\\"')}">${s.sku}</td>
+            <td>${s.id !== null ? s.id : ''}</td>
+            <td title="${title.replaceAll('"', '\\"')}">${title || ''}</td>
+            <td>${s.sku}</td>
             <td>${s.price !== null ? `$${s.price.toFixed(2)}` : ''}</td>
             <td class="hide-sm">${s.updatedAt ? new Date(s.updatedAt).toLocaleString() : ''}</td>
           `;

src/routes/api.ts

@@ -178,10 +178,14 @@ router.delete('/items', async (_req, res) => {
 router.get('/debug/ebay-prices', async (req, res) => {
   try {
     const skuParam = typeof req.query.sku === 'string' ? req.query.sku : '';
+    const manufacturerParam = typeof req.query.manufacturer === 'string' ? req.query.manufacturer : null;
     const norm = normalizeSku(skuParam);
     if (!norm) return res.status(400).json({ error: 'sku required' });
-    const details = await debugFetchSoldPricesUSDForSku(norm);
-    res.json({ sku: norm, ...details });
+    const details = await debugFetchSoldPricesUSDForSku(norm, manufacturerParam);
+    res.json({ 
+      ...details,
+      note: 'Scraping sold/completed listings using manufacturer + SKU for better search results.',
+    });
   } catch (err: any) {
     console.error(err);
     res.status(500).json({ error: err.message || 'Failed to debug ebay prices' });
@@ -191,22 +195,42 @@ router.get('/debug/ebay-prices', async (req, res) => {
 // Update cached prices for all distinct SKUs (on-demand only)
 router.post('/prices/update', async (_req, res) => {
   try {
-    const skus = await db.listDistinctSkus();
+    // Get SKU + manufacturer pairs for better search results
+    const skuManufacturers = await db.listDistinctSkuManufacturers();
     const out: Record<string, number | null> = {};
-      for (const sku of skus) {
+    
+    for (const { sku, manufacturer } of skuManufacturers) {
       try {
         const norm = normalizeSku(sku);
         if (!norm) { out[sku] = null; continue; }
-          const price = await fetchMedianSoldPriceUSDForSku(norm);
+        const price = await fetchMedianSoldPriceUSDForSku(norm, manufacturer);
         await db.upsertSkuPrice(norm, price, 'USD', 'ebay');
         out[norm] = price;
         // small delay to be gentle (reduce block risk)
-        await new Promise(r => setTimeout(r, 900));
+        await new Promise(r => setTimeout(r, 1200));
       } catch (e) {
         console.error('Failed to update price for', sku, e);
         out[sku] = null;
       }
     }
+    
+    // Also handle SKUs without manufacturer (fallback)
+    const skusWithoutMfr = await db.listDistinctSkus();
+    for (const sku of skusWithoutMfr) {
+      if (out[sku] !== undefined) continue; // Already processed
+      try {
+        const norm = normalizeSku(sku);
+        if (!norm) { out[sku] = null; continue; }
+        const price = await fetchMedianSoldPriceUSDForSku(norm, null);
+        await db.upsertSkuPrice(norm, price, 'USD', 'ebay');
+        out[norm] = price;
+        await new Promise(r => setTimeout(r, 1200));
+      } catch (e) {
+        console.error('Failed to update price for', sku, e);
+        out[sku] = null;
+      }
+    }
+    
     res.json({ updated: out });
   } catch (err: any) {
     console.error(err);
@@ -236,8 +260,8 @@ router.get('/price-report', async (_req, res) => {
       };
     });
 
-    // Build sku list with one description (first encountered)
-    const skuListMap: Record<string, { sku: string, price: number | null, currency: string, updatedAt: string | null, description: string } > = {};
+    // Build sku list with one description and ID (first encountered)
+    const skuListMap: Record<string, { sku: string, price: number | null, currency: string, updatedAt: string | null, description: string, id: number | null } > = {};
       for (const it of items) {
         const key = normalizeSku(it.sku);
         if (!key) continue;
@@ -248,7 +272,8 @@ router.get('/price-report', async (_req, res) => {
             price: p ? (p.price !== null ? Number(p.price) : null) : null,
             currency: p ? p.currency : 'USD',
             updatedAt: p ? p.updatedAt : null,
-            description: it.description
+            description: it.description,
+            id: it.id
           };
         }
     }

src/services/ebay.ts

@@ -1,5 +1,7 @@
 import * as cheerio from 'cheerio';
+import puppeteer from 'puppeteer';
 
+// Scraping result type
 type ScrapeResult = {
   prices: number[];
   blocked: boolean;
@@ -13,105 +15,285 @@ async function scrapeSoldPricesUSDPage(html: string, wantDiagnostics = false): P
   const sampleTexts: string[] = [];
   const samplePrices: number[] = [];
 
-  const blocked = html.includes('To continue, please verify') || html.toLowerCase().includes('robot check');
+  // Detect various eBay bot detection pages
+  const blocked = html.includes('To continue, please verify') 
+    || html.toLowerCase().includes('robot check')
+    || html.includes('Pardon Our Interruption')
+    || html.includes('access to this page has been denied')
+    || (html.length < 50000 && html.includes('ebaystatic.com') && !html.includes('srp-results')); // Suspiciously small HTML with eBay assets but no results
+
+  // Try multiple listing item selectors (eBay uses different structures)
+  const listingSelectors = [
+    'li.s-item',
+    'li[class*="s-item"]',
+    'ul.srp-results li',
+    'div[class*="srp-item"]',
+    'li[data-view*="item"]',
+  ];
 
   const priceSelectors = [
     '.s-item__price',
     '.s-item__detail--primary .s-item__price',
     'span[class*="s-item__price"]',
+    'span[class*="price"]',
+    'div[class*="price"]',
   ];
 
-  $('li.s-item').each((_i, el) => {
+  // Try each listing selector
+  let foundListings = false;
+  for (const listingSel of listingSelectors) {
+    const listings = $(listingSel);
+    if (listings.length > 0) {
+      foundListings = true;
+      if (wantDiagnostics) selectorHits[`listing_${listingSel}`] = listings.length;
+      
+      listings.each((_i, el) => {
         const $el = $(el);
         let text: string | null = null;
+        
+        // Try price selectors
         for (const sel of priceSelectors) {
           const t = $el.find(sel).first().text().trim();
           if (t) {
-        text = t; selectorHits[sel] = (selectorHits[sel] || 0) + 1; break;
+            text = t; 
+            selectorHits[`price_${sel}`] = (selectorHits[`price_${sel}`] || 0) + 1; 
+            break;
           }
         }
+        
+        // Regex fallback: look for price patterns in the listing HTML
         if (!text) {
-      // regex fallback within this listing's HTML
           const htmlFrag = $el.html() || '';
-      const m = htmlFrag.match(/\$\s*[0-9]{1,3}(?:,[0-9]{3})*(?:\.[0-9]{2})?/);
-      if (m) text = m[0];
+          const priceMatches = htmlFrag.match(/\$\s*[0-9]{1,3}(?:,[0-9]{3})*(?:\.[0-9]{2})?/g);
+          if (priceMatches && priceMatches.length > 0) {
+            // Take the first price match, but prefer ones that look like sold prices
+            text = priceMatches[0];
+            selectorHits['price_regex'] = (selectorHits['price_regex'] || 0) + 1;
           }
+        }
+        
         if (!text) return;
         if (wantDiagnostics && sampleTexts.length < 10) sampleTexts.push(text);
         if (!text.includes('$')) return;
-    const single = text.split(' to ')[0];
+        
+        // Extract price value
+        const single = text.split(' to ')[0].split(' ')[0]; // Handle ranges and extra text
         const num = single.replace(/[^0-9.]/g, '');
         if (!num) return;
         const value = Number(num);
-    if (!Number.isFinite(value) || value <= 0) return;
+        if (!Number.isFinite(value) || value <= 0 || value > 1000000) return; // Sanity check
         prices.push(value);
         if (wantDiagnostics && samplePrices.length < 10) samplePrices.push(value);
       });
       
+      if (prices.length > 0) break; // Found prices, no need to try other listing selectors
+    }
+  }
+
+  // If no listings found with standard selectors, try broad regex search as last resort
+  if (!foundListings && prices.length === 0 && !blocked) {
+    const priceMatches = html.match(/\$\s*[0-9]{1,3}(?:,[0-9]{3})*(?:\.[0-9]{2})?/g);
+    if (priceMatches) {
+      for (const match of priceMatches.slice(0, 20)) { // Limit to first 20 matches
+        const num = match.replace(/[^0-9.]/g, '');
+        const value = Number(num);
+        if (Number.isFinite(value) && value > 0 && value < 1000000) {
+          prices.push(value);
+          if (wantDiagnostics && samplePrices.length < 10) samplePrices.push(value);
+        }
+      }
+      if (wantDiagnostics) selectorHits['fallback_regex'] = priceMatches.length;
+    }
+  }
+
   return { prices, blocked, diagnostics: wantDiagnostics ? { selectorHits, sampleTexts, samplePrices } : undefined };
 }
 
-async function fetchSoldSearchHtml(query: string, page = 1): Promise<{ ok: boolean; html: string }> {
-  const url = `https://www.ebay.com/sch/i.html?_nkw=${encodeURIComponent(query)}&LH_Sold=1&LH_Complete=1&rt=nc&_ipg=200&_pgn=${page}`; 
-  const res = await fetch(url, {
-    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',
-      'Accept-Language': 'en-US,en;q=0.9'
-    } as any
-  } as any);
-  const html = await res.text();
-  return { ok: res.ok, html };
+/**
+ * Build search query combining manufacturer and SKU
+ */
+function buildSearchQuery(manufacturer: string | null, sku: string): string[] {
+  const queries: string[] = [];
+  
+  if (manufacturer && manufacturer.trim()) {
+    const cleanMfr = manufacturer.trim();
+    const cleanSku = sku.trim();
+    
+    // Try various combinations for better matching
+    queries.push(`"${cleanMfr}" "${cleanSku}"`);  // Exact match both
+    queries.push(`${cleanMfr} ${cleanSku}`);      // Both without quotes
+    queries.push(`"${cleanMfr} ${cleanSku}"`);    // Combined exact match
+    queries.push(`${cleanMfr} "${cleanSku}"`);    // Mfr loose, SKU exact
+  }
+  
+  // Fallback to SKU only if no manufacturer
+  queries.push(`"${sku}"`);
+  queries.push(sku);
+  
+  return queries;
 }
 
-export async function fetchMedianSoldPriceUSDForSku(sku: string): Promise<number | null> {
-  // Try quoted exact search first
-  const tryQueries = [
-    `"${sku}"`,
-    sku // fallback without quotes
-  ];
-  for (const q of tryQueries) {
-    let all: number[] = [];
-    for (let page = 1; page <= 2; page++) {
-      const { ok, html } = await fetchSoldSearchHtml(q, page);
+// Puppeteer browser instance (reused across requests)
+let browserInstance: any = null;
+
+/**
+ * Get or create Puppeteer browser instance
+ */
+async function getBrowser() {
+  if (!browserInstance) {
+    browserInstance = await puppeteer.launch({
+      headless: true,
+      args: [
+        '--no-sandbox',
+        '--disable-setuid-sandbox',
+        '--disable-dev-shm-usage',
+        '--disable-blink-features=AutomationControlled',
+      ],
+    });
+  }
+  return browserInstance;
+}
+
+/**
+ * Fetch sold/completed listings HTML from eBay search using Puppeteer
+ */
+async function fetchSoldSearchHtml(query: string, pageNum = 1): Promise<{ ok: boolean; html: string }> {
+  const url = `https://www.ebay.com/sch/i.html?_nkw=${encodeURIComponent(query)}&LH_Sold=1&LH_Complete=1&rt=nc&_ipg=200&_pgn=${pageNum}`; 
+  
+  try {
+    const browser = await getBrowser();
+    const page = await browser.newPage();
+    
+    // Set realistic viewport and user agent
+    await page.setViewport({ width: 1920, height: 1080 });
+    await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36');
+    
+    // Navigate to the search page
+    await page.goto(url, { 
+      waitUntil: 'networkidle2',
+      timeout: 30000 
+    });
+    
+    // Wait for listings to load (check for s-item elements or price elements)
+    try {
+      await page.waitForSelector('li.s-item, ul.srp-results li, [class*="s-item"]', { timeout: 10000 });
+    } catch (e) {
+      // Listings might not have loaded, but continue anyway
+    }
+    
+    // Get the rendered HTML
+    const html = await page.content();
+    await page.close();
+    
+    return { ok: true, html };
+  } catch (error: any) {
+    console.error(`Failed to fetch eBay search with Puppeteer: ${error.message}`);
+    return { ok: false, html: '' };
+  }
+}
+
+/**
+ * Fetch average price for a SKU by scraping sold/completed listings
+ * Uses the last 3 sold listings (or 2 or 1 if less available)
+ * Uses manufacturer + SKU for better search results
+ */
+export async function fetchMedianSoldPriceUSDForSku(sku: string, manufacturer: string | null = null): Promise<number | null> {
+  const queries = buildSearchQuery(manufacturer, sku);
+  
+  for (const q of queries) {
+    // Get prices from first page (most recent sold listings)
+    const { ok, html } = await fetchSoldSearchHtml(q, 1);
     if (!ok) continue;
     const { prices, blocked } = await scrapeSoldPricesUSDPage(html);
-      if (blocked) return null;
-      all = all.concat(prices);
-      if (all.length === 0) {
-        // small delay before next page to be gentle
+    if (blocked) {
+      console.warn(`eBay blocked request for query: ${q}`);
+      // Try next query instead of giving up
+      continue;
+    }
+    
+    // Take the last 3 sold listings (first 3 prices from the page, as eBay shows most recent first)
+    const recentPrices = prices.slice(0, 3);
+    
+    if (recentPrices.length > 0) {
+      // Calculate average of available prices (3, 2, or 1)
+      const sum = recentPrices.reduce((a, b) => a + b, 0);
+      const avg = sum / recentPrices.length;
+      return Number(avg.toFixed(2));
+    }
+    
+    // Small delay before trying next query
+    if (queries.indexOf(q) < queries.length - 1) {
       await new Promise(r => setTimeout(r, 600));
     }
   }
-    if (all.length > 0) {
-      all.sort((a, b) => a - b);
-      const mid = Math.floor(all.length / 2);
-      return Number((all.length % 2 === 0 ? (all[mid - 1] + all[mid]) / 2 : all[mid]).toFixed(2));
-    }
-  }
+  
   return null;
 }
 
-export async function debugFetchSoldPricesUSDForSku(sku: string) {
-  const sequences = [`"${sku}"`, sku];
+/**
+ * Debug function to inspect eBay price scraping attempts
+ */
+export async function debugFetchSoldPricesUSDForSku(sku: string, manufacturer: string | null = null) {
   const attempts: any[] = [];
-  for (const q of sequences) {
+  const queries = buildSearchQuery(manufacturer, sku);
+
+  for (const q of queries) {
     let total = 0;
     let blocked = false;
     const diagnostics: any = { pages: [] };
-    for (let page = 1; page <= 2; page++) {
+    
+    for (let page = 1; page <= 3; page++) {
       const { ok, html } = await fetchSoldSearchHtml(q, page);
       const diag = await scrapeSoldPricesUSDPage(html, true);
-      diagnostics.pages.push({ page, ok, count: diag.prices.length, selectorHits: diag.diagnostics?.selectorHits, sampleTexts: diag.diagnostics?.sampleTexts, samplePrices: diag.diagnostics?.samplePrices });
+      
+      // Additional HTML diagnostics
+      const htmlDiagnostics: any = {
+        htmlLength: html.length,
+        containsSold: html.toLowerCase().includes('sold'),
+        containsResults: html.toLowerCase().includes('results') || html.toLowerCase().includes('result'),
+        containsLi: html.includes('<li'),
+        containsUl: html.includes('<ul'),
+        sampleHtml: html.substring(0, 500), // First 500 chars for inspection
+      };
+      
+      diagnostics.pages.push({ 
+        page, 
+        ok, 
+        count: diag.prices.length, 
+        selectorHits: diag.diagnostics?.selectorHits, 
+        sampleTexts: diag.diagnostics?.sampleTexts, 
+        samplePrices: diag.diagnostics?.samplePrices,
+        htmlDiagnostics
+      });
       total += diag.prices.length;
       blocked = blocked || diag.blocked;
+      
       if (diag.prices.length === 0 && page === 1) {
+        // No results on first page, try next query
+        break;
+      }
+      
+      // Small delay between pages
+      if (page < 3) {
         await new Promise(r => setTimeout(r, 600));
       }
     }
-    attempts.push({ query: q, totalCount: total, blocked, details: diagnostics });
-    if (total > 0) break;
+    
+    attempts.push({ 
+      method: 'scraping', 
+      query: q, 
+      totalCount: total, 
+      blocked, 
+      details: diagnostics 
+    });
+    
+    if (total > 0) {
+      // Found results, but continue to show all attempts in debug
+      break;
     }
-  return { attempts };
+  }
+  
+  return { attempts, manufacturer, sku };
 }