Spaces:
Sleeping
Sleeping
Add Bright Data Scraper Tool
Browse files- app.py +5 -0
- requirements.txt +2 -0
- tool.py +56 -0
app.py
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from smolagents import launch_gradio_demo
|
| 2 |
+
from tool import BrightDataScraperTool
|
| 3 |
+
|
| 4 |
+
tool = BrightDataScraperTool()
|
| 5 |
+
launch_gradio_demo(tool)
|
requirements.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
requests
|
| 2 |
+
smolagents
|
tool.py
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from typing import Any, Optional
|
| 2 |
+
from smolagents.tools import Tool
|
| 3 |
+
import requests
|
| 4 |
+
import os
|
| 5 |
+
|
| 6 |
+
class BrightDataScraperTool(Tool):
|
| 7 |
+
name = "brightdata_web_scraper"
|
| 8 |
+
description = """
|
| 9 |
+
Scrape any webpage and return content in Markdown format.
|
| 10 |
+
This tool can bypass bot detection and CAPTCHAs.
|
| 11 |
+
Use this when you need to extract content from websites.
|
| 12 |
+
"""
|
| 13 |
+
inputs = {'url': {'type': 'string', 'description': 'The URL of the webpage to scrape'}}
|
| 14 |
+
output_type = "string"
|
| 15 |
+
|
| 16 |
+
def forward(self, url: str) -> str:
|
| 17 |
+
"""
|
| 18 |
+
Scrape a webpage using Bright Data's API.
|
| 19 |
+
|
| 20 |
+
Args:
|
| 21 |
+
url: The URL to scrape
|
| 22 |
+
|
| 23 |
+
Returns:
|
| 24 |
+
The scraped content in Markdown format
|
| 25 |
+
"""
|
| 26 |
+
import os
|
| 27 |
+
import requests
|
| 28 |
+
|
| 29 |
+
api_token = os.getenv("BRIGHT_DATA_API_TOKEN")
|
| 30 |
+
unlocker_zone = os.getenv("BRIGHT_DATA_UNLOCKER_ZONE", "web_unlocker_1")
|
| 31 |
+
|
| 32 |
+
if not api_token:
|
| 33 |
+
raise ValueError("BRIGHT_DATA_API_TOKEN not found in environment variables")
|
| 34 |
+
|
| 35 |
+
api_url = "https://api.brightdata.com/request"
|
| 36 |
+
headers = {
|
| 37 |
+
"Authorization": f"Bearer {api_token}",
|
| 38 |
+
"Content-Type": "application/json",
|
| 39 |
+
}
|
| 40 |
+
|
| 41 |
+
payload = {
|
| 42 |
+
"url": url,
|
| 43 |
+
"zone": unlocker_zone,
|
| 44 |
+
"format": "raw",
|
| 45 |
+
"data_format": "markdown",
|
| 46 |
+
}
|
| 47 |
+
|
| 48 |
+
try:
|
| 49 |
+
response = requests.post(api_url, json=payload, headers=headers)
|
| 50 |
+
response.raise_for_status()
|
| 51 |
+
return response.text
|
| 52 |
+
except requests.exceptions.RequestException as e:
|
| 53 |
+
return f"Error scraping URL: {str(e)}"
|
| 54 |
+
|
| 55 |
+
def __init__(self, *args, **kwargs):
|
| 56 |
+
self.is_initialized = False
|