Added headless browser example (#89)

* added headless browser example

* change folder

Author: NilayYadav

web_scraping/headless_chrome/.beamignore

@@ -0,0 +1,28 @@
+# Generated by Beam SDK
+.beamignore
+pyproject.toml
+.git
+.idea
+.python-version
+.vscode
+.venv
+venv
+__pycache__
+.DS_Store
+.config
+drive/MyDrive
+.coverage
+.pytest_cache
+.ipynb
+.ruff_cache
+.dockerignore
+.ipynb_checkpoints
+.env.local
+.envrc
+**/__pycache__/
+**/.pytest_cache/
+**/node_modules/
+**/.venv/
+*.pyc
+.next/
+.circleci

web_scraping/headless_chrome/README.md

@@ -0,0 +1,37 @@
+# Headless Browser API Example
+
+A headless browser API powered by Playwright, capturing full-page website screenshots.
+
+## Deployment
+
+Deploy the app on Beam:
+
+```
+beam deploy app.py:browser
+```
+
+## API Usage
+
+Send a `POST` request to the endpoint with the following JSON body:
+
+```json
+{
+  "url": "https://your-website-url"
+}
+```
+
+### Example Request:
+
+```json
+{
+  "url": "https://example.com"
+}
+```
+
+### Example Response:
+
+```json
+{
+  "output_url": "https://app.beam.cloud/output/id/9dfbb7a1-a3de-489c-a602-423b4c859f84"
+}
+```

web_scraping/headless_chrome/app.py

@@ -0,0 +1,49 @@
+from beam import Output, endpoint, Image, env
+import asyncio
+
+if env.is_remote():
+    from playwright.async_api import async_playwright
+
+image = (
+    Image(python_version="python3.11")
+    .add_python_packages(
+        [
+            "playwright",
+        ]
+    )
+    .add_commands(["playwright install chromium", "playwright install-deps chromium"])
+)
+
+
+@endpoint(name="headless-browser", cpu=2, memory="16Gi", image=image)
+async def browser(url: str = "https://example.com"):
+    print(f"Navigating to: {url}")
+    output_path = "/tmp/screenshot.png"
+
+    async with async_playwright() as playwright:
+        browser = await playwright.chromium.launch(
+            headless=True,
+            args=["--no-sandbox", "--disable-dev-shm-usage"],
+        )
+        print("Browser launched successfully")
+
+        try:
+            page = await browser.new_page()
+            await page.set_viewport_size({"width": 1920, "height": 1080})
+
+            await page.goto(url, wait_until="networkidle")
+            print("Page loaded, waiting 2 seconds for any dynamic content...")
+
+            await asyncio.sleep(2)
+
+            await page.screenshot(path=output_path, full_page=True, type="png")
+            print(f"Screenshot saved as: {output_path}")
+
+        finally:
+            await browser.close()
+
+    output_file = Output(path=output_path)
+    output_file.save()
+    public_url = output_file.public_url(expires=400)
+
+    return {"output_url": public_url}