Add proxy and media blocking configurations

Updated environment variables and application settings to include proxy configurations and media blocking option. The proxy settings allow users to use a proxy service, while the media blocking is an optional feature that can help save bandwidth. Changes have been made in the .env.example, docker-compose.yaml, and main.py files.

apps/api/.env.example

@@ -36,3 +36,10 @@ HYPERDX_API_KEY=
 HDX_NODE_BETA_MODE=1
 
 FIRE_ENGINE_BETA_URL= # set if you'd like to use the fire engine closed beta
+
+# Proxy Settings (Alternative you can can use a proxy service like oxylabs, which rotates IPs for you on every request)
+PROXY_SERVER=
+PROXY_USERNAME=
+PROXY_PASSWORD=
+# set if you'd like to block media requests to save proxy bandwidth
+BLOCK_MEDIA=

apps/playwright-service/main.py

@@ -2,9 +2,16 @@ from fastapi import FastAPI
 from playwright.async_api import async_playwright, Browser
 from fastapi.responses import JSONResponse
 from pydantic import BaseModel
+from os import environ
+
+PROXY_SERVER = environ.get('PROXY_SERVER', None)
+PROXY_USERNAME = environ.get('PROXY_USERNAME', None)
+PROXY_PASSWORD = environ.get('PROXY_PASSWORD', None)
+BLOCK_MEDIA = environ.get('BLOCK_MEDIA', 'False').upper() == 'TRUE'
 
 app = FastAPI()
 
+
 class UrlModel(BaseModel):
     url: str
     wait: int = None
@@ -27,7 +34,18 @@ async def shutdown_event():
 
 @app.post("/html")
 async def root(body: UrlModel):
+    context = None
+    if PROXY_SERVER and PROXY_USERNAME and PROXY_PASSWORD:
+        context = await browser.new_context(proxy={"server": PROXY_SERVER,
+                                                   "username": PROXY_USERNAME,
+                                                   "password": PROXY_PASSWORD})
+    else:
         context = await browser.new_context()
+
+    if BLOCK_MEDIA:
+        await context.route("**/*.{png,jpg,jpeg,gif,svg,mp3,mp4,avi,flac,ogg,wav,webm}",
+                            handler=lambda route, request: route.abort())
+
     page = await context.new_page()
     await page.goto(body.url, timeout=15000)  # Set max timeout to 15s
     if body.wait:  # Check if wait parameter is provided in the request body

docker-compose.yaml

@@ -5,6 +5,10 @@ services:
     build: apps/playwright-service
     environment:
       - PORT=3000
+      - PROXY_SERVER=${PROXY_SERVER}
+      - PROXY_USERNAME=${PROXY_USERNAME}
+      - PROXY_PASSWORD=${PROXY_PASSWORD}
+      - BLOCK_MEDIA=${BLOCK_MEDIA}
     networks:
       - backend