Spaces:
Sleeping
Sleeping
Create toolVisitWebpage.py
Browse files- toolVisitWebpage.py +44 -0
toolVisitWebpage.py
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from smolagents import tool
|
| 2 |
+
import requests
|
| 3 |
+
from bs4 import BeautifulSoup
|
| 4 |
+
|
| 5 |
+
MAX_WEBPAGE_SIZE = 3000 # max characters to return from scraped content
|
| 6 |
+
|
| 7 |
+
@tool
|
| 8 |
+
def visit_webpage(url: str) -> dict:
|
| 9 |
+
"""
|
| 10 |
+
Visits a webpage and extracts clean text from it.
|
| 11 |
+
|
| 12 |
+
Args:
|
| 13 |
+
url: The URL of the page to visit (e.g., https://en.wikipedia.org/wiki/OpenAI)
|
| 14 |
+
|
| 15 |
+
Returns:
|
| 16 |
+
dict: A dictionary containing:
|
| 17 |
+
- "text": Truncated page content
|
| 18 |
+
- "url": The original URL
|
| 19 |
+
- "status": HTTP status or error info
|
| 20 |
+
"""
|
| 21 |
+
print(f" Tool:visit_webpage visiting {url}...")
|
| 22 |
+
|
| 23 |
+
try:
|
| 24 |
+
response = requests.get(url, timeout=10)
|
| 25 |
+
response.raise_for_status()
|
| 26 |
+
|
| 27 |
+
soup = BeautifulSoup(response.text, "html.parser")
|
| 28 |
+
text = soup.get_text(separator="\n", strip=True)
|
| 29 |
+
short_text = text[:MAX_WEBPAGE_SIZE]
|
| 30 |
+
|
| 31 |
+
print(f"✅ Extracted {len(short_text)} characters from {url}")
|
| 32 |
+
return {
|
| 33 |
+
"text": short_text,
|
| 34 |
+
"url": url,
|
| 35 |
+
"status": f"Success ({response.status_code})"
|
| 36 |
+
}
|
| 37 |
+
|
| 38 |
+
except Exception as e:
|
| 39 |
+
print(f"🚨 Error in visit_webpage: {e}")
|
| 40 |
+
return {
|
| 41 |
+
"text": "",
|
| 42 |
+
"url": url,
|
| 43 |
+
"status": f"Error: {e}"
|
| 44 |
+
}
|