feat: add infinte scrolling

scrapegraph-js/README.md

@@ -107,6 +107,30 @@ const schema = z.object({
 })();
 ```
 
+#### Scraping with Infinite Scrolling
+
+For websites that load content dynamically through infinite scrolling (like social media feeds), you can use the `numberOfScrolls` parameter:
+
+```javascript
+import { smartScraper } from 'scrapegraph-js';
+
+const apiKey = 'your-api-key';
+const url = 'https://example.com/infinite-scroll-page';
+const prompt = 'Extract all the posts from the feed';
+const numberOfScrolls = 10; // Will scroll 10 times to load more content
+
+(async () => {
+  try {
+    const response = await smartScraper(apiKey, url, prompt, null, numberOfScrolls);
+    console.log('Extracted data from scrolled page:', response);
+  } catch (error) {
+    console.error('Error:', error);
+  }
+})();
+```
+
+The `numberOfScrolls` parameter accepts values between 0 and 100, allowing you to control how many times the page should be scrolled before extraction.
+
 ### Search Scraping
 
 Search and extract information from multiple web sources using AI.

scrapegraph-js/examples/smartScraper_infinite_scroll_example.js

@@ -0,0 +1,15 @@
+import { smartScraper } from 'scrapegraph-js';
+import 'dotenv/config';
+
+const apiKey = process.env.SGAI_APIKEY;
+// Example URL that requires scrolling (e.g., a social media feed or infinite scroll page)
+const url = 'https://example.com/infinite-scroll-page';
+const prompt = 'Extract all the posts from the feed';
+const numberOfScrolls = 10; // Will scroll 10 times to load more content
+
+try {
+  const response = await smartScraper(apiKey, url, prompt, null, numberOfScrolls);
+  console.log('Extracted data from scrolled page:', response);
+} catch (error) {
+  console.error('Error:', error);
+} 

scrapegraph-js/src/smartScraper.js

@@ -10,10 +10,11 @@ import { zodToJsonSchema } from 'zod-to-json-schema';
  * @param {string} url - The URL of the webpage to scrape
  * @param {string} prompt - Natural language prompt describing what data to extract
  * @param {Object} [schema] - Optional schema object defining the output structure
+ * @param {number} [numberOfScrolls] - Optional number of times to scroll the page (0-100). If not provided, no scrolling will be performed.
  * @returns {Promise<string>} Extracted data in JSON format matching the provided schema
  * @throws - Will throw an error in case of an HTTP failure.
  */
-export async function smartScraper(apiKey, url, prompt, schema = null) {
+export async function smartScraper(apiKey, url, prompt, schema = null, numberOfScrolls = null) {
   const endpoint = 'https://api.scrapegraphai.com/v1/smartscraper';
   const headers = {
     'accept': 'application/json',
@@ -34,6 +35,13 @@ export async function smartScraper(apiKey, url, prompt, schema = null) {
     }
   }
 
+  if (numberOfScrolls !== null) {
+    if (!Number.isInteger(numberOfScrolls) || numberOfScrolls < 0 || numberOfScrolls > 100) {
+      throw new Error('numberOfScrolls must be an integer between 0 and 100');
+    }
+    payload.number_of_scrolls = numberOfScrolls;
+  }
+
   try {
     const response = await axios.post(endpoint, payload, { headers });
     return response.data;