Add Playwright service and integrate with tools.py for web search functionality

Dockerfile

@@ -39,6 +39,12 @@ USER user
 COPY --chown=user requirements.txt requirements.txt
 RUN pip install --no-cache-dir --upgrade -r requirements.txt
 
+# Install Node.js and Playwright
+RUN curl -fsSL https://deb.nodesource.com/setup_20.x | bash - && \
+    apt-get install -y nodejs && \
+    npm install -g npm && \
+    npm install express playwright
+
 # Copy nginx configuration
 COPY --chown=user nginx.conf /etc/nginx/sites-available/default
 

playwright_service.js

@@ -0,0 +1,38 @@
+import express from 'express';
+import { chromium } from 'playwright';
+import fs from 'fs';
+
+const app = express();
+app.use(express.json());
+
+app.post('/search', async (req, res) => {
+  const { query } = req.body;
+  if (!query) return res.status(400).json({ error: 'Missing query' });
+  let browser;
+  try {
+    browser = await chromium.launch({ headless: true });
+    const context = await browser.newContext({
+      userAgent: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
+      viewport: { width: 1280, height: 800 },
+      locale: 'en-US',
+    });
+    const page = await context.newPage();
+    // Use Bing's HTML search endpoint
+    const searchUrl = 'https://www.bing.com/search?q=' + encodeURIComponent(query);
+    await page.goto(searchUrl, { waitUntil: 'domcontentloaded' });
+    await page.waitForTimeout(1200 + Math.floor(Math.random() * 800)); // Random delay
+    // Get visible text from the results page
+    const visibleText = await page.evaluate(() => document.body.innerText);
+    res.json({ result: visibleText });
+  } catch (e) {
+    console.error('Playwright error:', e);
+    res.status(500).json({ error: e.message });
+  } finally {
+    if (browser) await browser.close();
+  }
+});
+
+const PORT = process.env.PLAYWRIGHT_PORT || 5000;
+app.listen(PORT, () => {
+  console.log(`Playwright service running on port ${PORT}`);
+});

run.sh

@@ -3,10 +3,14 @@
 # start nginx
 service nginx start
 
+# start the Playwright service
+node playwright_service.js > /dev/stdout 2>&1 & echo $! > playwright.pid
+
 # start the processes
 python tools.py > /dev/stdout 2>&1 & echo $! > tools.pid
 python app.py > /dev/stdout 2>&1 # blocking
 
 # when unblocked, kill other processes and clean up
 pkill -F tools.pid
-rm tools.pid
+pkill -F playwright.pid
+rm tools.pid playwright.pid

tools.py

@@ -51,38 +51,17 @@ def calculate() -> Response:
 @tools.route("/websearch", methods=["POST"])
 def google_search() -> Response:
   try:
-    data = json.loads(request.data) 
-    global driver
-    driver = webdriver.Chrome(options=chrome_options)
-    driver.get("https://www.google.com/")
-    search_bar = driver.find_element(By.NAME, "q")
-    search_bar.send_keys(data["query"])
-    search_bar.send_keys(Keys.RETURN)
-    time.sleep(1)
-
-    search_results = driver.find_elements(By.CSS_SELECTOR, "div.kno-rdesc span span")
-    if len(search_results) > 0:  # check if google quick answer box exists
-      return jsonify(result=search_results[0].text)
-    else:  # otherwise, find list of search results
-      search_results = driver.find_elements(By.CSS_SELECTOR, "div.g")
-      for result in search_results:
-        try:  # first sentence of last span element in result
-          first_result = result.find_elements(By.CSS_SELECTOR, "div.VwiC3b span")
-          return jsonify(result=first_result[-1].text.split("...")[0])
-        except IndexError:  # if no span element in result, go to next result
-          pass
-      return jsonify(result="No search results found")
+    data = json.loads(request.data)
+    import requests
+    # Call the Playwright service
+    resp = requests.post("http://localhost:5000/search", json={"query": data["query"]}, timeout=30)
+    resp.raise_for_status()
+    result = resp.json().get("result", "No result returned")
+    return jsonify(result=result)
   except KeyError as e:
     abort(400, description="Missing value in request body: " + str(e))
   except Exception as e:
     abort(400, description="Error: " + str(e))
-  finally:
-    try:
-      # quit selenium and kill the vnc
-      driver.quit()
-      os.system("pkill -1 Xvnc")
-    except Exception:
-      pass
 
 @tools.route("/", methods=["GET"])
 def home() -> str: