Adds gen har, network requests, meta tags

api/generate-har.js

@@ -0,0 +1,54 @@
+const puppeteer = require('puppeteer-core');
+const chromium = require('chrome-aws-lambda');
+
+exports.handler = async (event, context) => {
+  let browser = null;
+  let result = null;
+  let code = 200;
+  
+  try {
+    const url = event.queryStringParameters.url;
+
+    browser = await chromium.puppeteer.launch({
+      args: chromium.args,
+      defaultViewport: chromium.defaultViewport,
+      executablePath: await chromium.executablePath,
+      headless: chromium.headless,
+    });
+    
+    const page = await browser.newPage();
+
+    const requests = [];
+    
+    // Capture requests
+    page.on('request', request => {
+      requests.push({
+        url: request.url(),
+        method: request.method(),
+        headers: request.headers(),
+      });
+    });
+
+    await page.goto(url, {
+      waitUntil: 'networkidle0',  // wait until all requests are finished
+    });
+
+    result = requests;
+    
+  } catch (error) {
+    code = 500;
+    result = {
+      error: 'Failed to create HAR file',
+      details: error.toString(),
+    };
+  } finally {
+    if (browser !== null) {
+      await browser.close();
+    }
+  }
+
+  return {
+    statusCode: code,
+    body: JSON.stringify(result),
+  };
+};

api/meta-tags.js

@@ -0,0 +1,58 @@
+const axios = require('axios');
+const cheerio = require('cheerio');
+
+exports.handler = async (event) => {
+  let url;
+  try {
+    // Add https:// prefix if not present
+    url = new URL(event.queryStringParameters.url);
+    if (!url.protocol) {
+      url = new URL('https://' + event.queryStringParameters.url);
+    }
+  } catch (error) {
+    // Return error if URL is not valid
+    return {
+      statusCode: 400,
+      body: JSON.stringify({
+        error: 'Invalid URL provided.',
+      }),
+    };
+  }
+  
+  try {
+    // Fetch the page
+    const response = await axios.get(url.toString());
+  
+    // Parse the page body with cheerio
+    const $ = cheerio.load(response.data);
+  
+    // Extract meta tags
+    const metaTags = {};
+  
+    $('head meta').each((index, element) => {
+      const name = $(element).attr('name');
+      const property = $(element).attr('property');
+      const content = $(element).attr('content');
+      
+      if (name) {
+        metaTags[name] = content;
+      } else if (property) {
+        metaTags[property] = content;
+      }
+    });
+  
+    // Return meta tags
+    return {
+      statusCode: 200,
+      body: JSON.stringify(metaTags),
+    };
+  } catch (error) {
+    // Return error if there's a problem fetching or parsing the page
+    return {
+      statusCode: 500,
+      body: JSON.stringify({
+        error: error.message,
+      }),
+    };
+  }
+};

api/network-requests.js

@@ -0,0 +1,58 @@
+const puppeteer = require('puppeteer-core');
+const chromium = require('chrome-aws-lambda');
+
+exports.handler = async (event, context) => {
+    const urlParam = event.queryStringParameters.url;
+    if (!urlParam) {
+      return { 
+        statusCode: 400, 
+        body: JSON.stringify({ error: 'Missing url parameter' }) 
+      };
+    }
+  
+    let url;
+    try {
+      url = new URL(urlParam.includes('://') ? urlParam : 'https://' + urlParam);
+    } catch (error) {
+      return {
+        statusCode: 500,
+        body: JSON.stringify({ error: 'Invalid URL format' }),
+      };
+    }
+
+    // Launch the browser and open a new page
+    const browser = await puppeteer.launch({
+      args: chromium.args,
+      defaultViewport: { width: 800, height: 600 },
+      executablePath: process.env.CHROME_PATH || await chromium.executablePath,
+      headless: chromium.headless,
+      ignoreHTTPSErrors: true,
+      ignoreDefaultArgs: ['--disable-extensions'],
+    });
+    const page = await browser.newPage();
+
+    // To store network activity
+    let networkActivity = [];
+
+    // Register an event listener for network requests
+    page.on('request', (request) => {
+        networkActivity.push({
+            url: request.url(),
+            method: request.method(),
+            headers: request.headers(),
+            postData: request.postData(),
+        });
+    });
+
+    // Navigate to the page and wait for it to load
+    await page.goto(url, { waitUntil: 'networkidle2' });
+
+    // Close the browser
+    await browser.close();
+
+    // Return network activity
+    return {
+        statusCode: 200,
+        body: JSON.stringify(networkActivity),
+    };
+};