Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -9,6 +9,11 @@ from bs4 import BeautifulSoup
|
|
| 9 |
from urllib.parse import urljoin, urlparse
|
| 10 |
import urllib.robotparser as robotparser
|
| 11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
from Gradio_UI import GradioUI
|
| 13 |
|
| 14 |
def allowed_by_robots(url: str, user_agent: str = DEFAULT_HEADERS["User-Agent"]) -> bool:
|
|
|
|
| 9 |
from urllib.parse import urljoin, urlparse
|
| 10 |
import urllib.robotparser as robotparser
|
| 11 |
|
| 12 |
+
DEFAULT_HEADERS = {
|
| 13 |
+
"User-Agent": "smolagents-bot/1.0 (+https://example.com/bot) requests"
|
| 14 |
+
}
|
| 15 |
+
|
| 16 |
+
|
| 17 |
from Gradio_UI import GradioUI
|
| 18 |
|
| 19 |
def allowed_by_robots(url: str, user_agent: str = DEFAULT_HEADERS["User-Agent"]) -> bool:
|