Add a saved-page corpus and a Chromium test harness

.gitattributes

@@ -0,0 +1 @@
+tests/pages/**/*.html -text -diff

.github/workflows/ci.yml

@@ -43,3 +43,35 @@ jobs:
           name: store-zip
           path: dist-zips/*.zip
           if-no-files-found: error
+
+  e2e:
+    runs-on: ubuntu-latest
+    timeout-minutes: 15
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          # The upgrade scenarios check out the v2.0 store commit.
+          fetch-depth: 0
+
+      - uses: actions/setup-node@v4
+        with:
+          node-version: 22
+          cache: npm
+
+      - run: npm ci
+
+      # Full Chromium, not the headless shell: extensions need it. The browser
+      # version is pinned by the @playwright/test version in package.json.
+      - name: Install Chromium
+        run: npx playwright install --with-deps --no-shell chromium
+
+      - name: Chromium harness
+        run: npx playwright test
+
+      - uses: actions/upload-artifact@v4
+        if: failure()
+        with:
+          name: e2e-report
+          path: |
+            tests/e2e/.report
+            tests/e2e/.results

.gitignore

@@ -10,3 +10,8 @@ dist/
 build/
 *.zip
 dist-zips/
+
+# e2e harness output
+tests/e2e/.build/
+tests/e2e/.results/
+tests/e2e/.report/

CLAUDE.md

@@ -18,6 +18,8 @@ AmazonSellerScraper/
 │   │   ├── scraper.js        # DOM scraping on Amazon pages
 │   │   ├── chatbot.js        # Floating AI chatbot widget (Shadow DOM)
 │   │   └── offer-fetcher.js  # Seller price fetching for spread analysis
+│   ├── lib/
+│   │   └── parsers.js        # Pure search/offer parsing (global Parsers)
 │   ├── background/
 │   │   └── service-worker.js # Message routing + Gemini API calls
 │   └── modules/
@@ -93,7 +95,7 @@ AmazonSellerScraper/
 
 ## Testing
 
-### JavaScript (Jest + JSDOM) — 214 tests
+### JavaScript (Jest + JSDOM)
 
 ```bash
 npm test              # Run all JS tests

README.md

@@ -142,6 +142,8 @@ For local Firebase work, `npm run build:dev` points the build at the emulators u
 
 `npm test` runs the Jest suite and the tool tests. `npm run check` builds, then runs the permission lock (nothing may be added over `tools/live-manifest.json`, the published v2.0 manifest), the version gate and the secret scan. `npm run zip` writes the store package to `dist-zips/` and refuses a dev build, a stray file, uncommitted changes (`node tools/zip.mjs --allow-dirty` overrides that for local tries), or any gate failure. CI runs all of these.
 
+The Jest suite includes a golden corpus of saved Amazon pages (`tests/pages/`, see its README). `npm run test:e2e` loads the built extension into Chromium and runs scrape scenarios against those pages, with every request answered locally. Run `npx playwright install --no-shell chromium` once first. Known bugs run as expected failures tagged with their audit finding id; `PROSCAN_SHOW_KNOWN=1 npm run test:e2e` shows what they fail on.
+
 ## Usage
 
 1. Navigate to any Amazon search results or seller page
@@ -179,9 +181,11 @@ AmazonSellerScraper/
 │   └── popup.js                  # UI state management and export handling
 ├── scripts/
 │   ├── content/
-│   │   ├── scraper.js            # DOM scraping with cascading selectors
+│   │   ├── scraper.js            # Scrape loop: storage, messages, pagination
 │   │   ├── chatbot.js            # Floating AI chatbot (Shadow DOM)
 │   │   └── offer-fetcher.js      # Seller offer page fetching for spread analysis
+│   ├── lib/
+│   │   └── parsers.js            # Pure search and offer page parsing
 │   ├── background/
 │   │   └── service-worker.js     # Message routing + Gemini API
 │   └── modules/

jest.config.js

@@ -14,6 +14,7 @@ module.exports = {
   coverageDirectory: 'coverage',
   collectCoverageFrom: [
     'scripts/modules/*.js',
+    'scripts/lib/*.js',
     'scripts/content/scraper.js',
     'scripts/content/offer-fetcher.js',
     '!**/node_modules/**'

manifest.json

@@ -29,6 +29,7 @@
             "matches": ["*://*.amazon.com/*"],
             "js": [
                 "scripts/modules/price.js",
+                "scripts/lib/parsers.js",
                 "scripts/modules/delta.js",
                 "scripts/content/scraper.js",
                 "scripts/content/chatbot.js",

package.json

@@ -9,12 +9,14 @@
     "test:unit": "jest tests/unit --verbose",
     "test:integration": "jest tests/integration --verbose",
     "test:tools": "node --test --test-concurrency=1 \"tools/tests/*.test.mjs\"",
+    "test:e2e": "playwright test",
     "test:coverage": "jest --coverage",
     "lock": "node tools/permission-lock.mjs && node tools/version-gate.mjs",
     "scan:secrets": "node tools/secret-scan.mjs",
     "check": "npm run build && npm run lock && npm run scan:secrets"
   },
   "devDependencies": {
+    "@playwright/test": "1.63.0",
     "adm-zip": "^0.5.17",
     "esbuild": "^0.28.0",
     "jest": "^29.7.0",

playwright.config.mjs

@@ -0,0 +1,14 @@
+import { defineConfig } from '@playwright/test';
+
+// One Chromium at a time: every test loads the extension into its own profile.
+export default defineConfig({
+  testDir: 'tests/e2e',
+  testMatch: '*.spec.mjs',
+  globalSetup: './tests/e2e/global-setup.mjs',
+  workers: 1,
+  fullyParallel: false,
+  retries: 0,
+  timeout: 90_000,
+  reporter: process.env.CI ? [['list'], ['html', { open: 'never', outputFolder: 'tests/e2e/.report' }]] : 'list',
+  outputDir: 'tests/e2e/.results',
+});

popup/popup.html

@@ -5,6 +5,7 @@
     <link rel="stylesheet" href="popup.css">
     <script src="../libs/xlsx.full.min.js"></script>
     <script src="../scripts/modules/price.js"></script>
+    <script src="../scripts/lib/parsers.js"></script>
     <script src="../scripts/modules/storage.js"></script>
     <script src="../scripts/modules/analyzer.js"></script>
     <script src="../scripts/modules/spread-analyzer.js"></script>

scripts/content/offer-fetcher.js

@@ -22,102 +22,9 @@
 /** @type {boolean} Whether spread analysis is currently running */
 let isAnalyzing = false;
 
-/**
- * CSS selectors for extracting prices from Amazon offer listing pages.
- * Multiple selectors for resilience against layout changes.
- * @const {Object}
- */
-const OFFER_SELECTORS = {
-    // AOD (All Offers Display) format — modern Amazon
-    aodPriceBlock: '.aod-information-block .a-price .a-offscreen',
-
-    // Classic offer listing page format
-    offerListPrice: '#olpOfferList .a-price .a-offscreen',
-
-    // Legacy offer listing format
-    legacyOfferPrice: '.olpOfferPrice',
-
-    // Scoped fallback — prices inside an offer/seller container only.
-    // Must NOT be the global '.a-price .a-offscreen' selector, which would
-    // scoop up buy-box, sponsored, and accessory prices as fake seller offers.
-    generalPrice: '#aod-offer-list .a-price .a-offscreen, [id^="aod-offer"] .a-price .a-offscreen, .olpOffer .a-price .a-offscreen'
-};
-
-/**
- * Build the offer listing URL for a given ASIN.
- * Uses the classic offer listing page format filtered to new condition.
- *
- * @param {string} asin - Amazon Standard Identification Number
- * @returns {string} Full URL for the offer listing page
- */
-function buildOfferUrl(asin) {
-    return `https://www.amazon.com/gp/offer-listing/${asin}/ref=dp_olp_all_mbc?ie=UTF8&condition=new`;
-}
-
-/**
- * Build the AOD (All Offers Display) AJAX URL for a given ASIN.
- * This endpoint returns HTML fragments with all seller offers.
- *
- * @param {string} asin - Amazon Standard Identification Number
- * @returns {string} AJAX endpoint URL
- */
-function buildAodUrl(asin) {
-    return `https://www.amazon.com/gp/aod/ajax?asin=${asin}&condition=new&m=&qid=&smid=&sourcecustomerorglistid=&sourcecustomerorglistitemid=&sr=&pc=dp`;
-}
-
-/**
- * Parse a price string from Amazon's DOM into a numeric value.
- * Handles currency symbols, commas, and whitespace.
- *
- * @param {string} text - Raw price text (e.g., "$19.99", "$1,299.00")
- * @returns {number} Parsed price, or 0 if unparseable
- */
-function parseOfferPrice(text) {
-    if (!text) return 0;
-    const cleaned = text.replace(/[^0-9.]/g, '');
-    const price = parseFloat(cleaned);
-    return isNaN(price) ? 0 : price;
-}
-
-/**
- * Extract all seller prices from an HTML document using cascading selectors.
- * Tries multiple selector strategies in order of specificity.
- *
- * @param {Document} doc - Parsed HTML document (from DOMParser)
- * @returns {number[]} Array of valid seller prices (> 0)
- */
-function extractPricesFromDocument(doc) {
-    const prices = [];
-
-    // Try each selector strategy in order
-    const selectorStrategies = [
-        OFFER_SELECTORS.aodPriceBlock,
-        OFFER_SELECTORS.offerListPrice,
-        OFFER_SELECTORS.legacyOfferPrice,
-        OFFER_SELECTORS.generalPrice
-    ];
-
-    for (const selector of selectorStrategies) {
-        const elements = doc.querySelectorAll(selector);
-        if (elements.length > 0) {
-            elements.forEach(el => {
-                const price = parseOfferPrice(el.textContent);
-                if (price > 0) {
-                    prices.push(price);
-                }
-            });
-            // If we found prices with this selector, don't try fallbacks
-            if (prices.length > 0) break;
-        }
-    }
-
-    // Return all prices as-is. Do NOT dedup: multiple distinct sellers
-    // legitimately list the same price, and each is a real data point for
-    // the spread statistics (sellerCount, mean, stdDev, CV, confidence factor).
-    // The cascade `break`s after the first selector that yields prices, so a
-    // single physical price element is never matched twice.
-    return prices;
-}
+// Pure parsing lives in scripts/lib/parsers.js. These names stay for the fetch
+// loop below and for the unit tests that load this file.
+var { buildOfferUrl, buildAodUrl, parseOfferPrice, extractPricesFromDocument } = Parsers;
 
 /**
  * Fetch the offer listing page for a single ASIN and extract seller prices.