Update tools/visit_webpage.py
Browse files- tools/visit_webpage.py +7 -3
tools/visit_webpage.py
CHANGED
|
@@ -11,6 +11,10 @@ class VisitWebpageTool(Tool):
|
|
| 11 |
inputs = {'url': {'type': 'string', 'description': 'The url of the webpage to visit.'}}
|
| 12 |
output_type = "string"
|
| 13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
def forward(self, url: str) -> str:
|
| 15 |
try:
|
| 16 |
import requests
|
|
@@ -23,8 +27,8 @@ class VisitWebpageTool(Tool):
|
|
| 23 |
"You must install packages `markdownify` and `requests` to run this tool: for instance run `pip install markdownify requests`."
|
| 24 |
) from e
|
| 25 |
try:
|
| 26 |
-
# Send a GET request to the URL with a 20-second timeout
|
| 27 |
-
response = requests.get(url, timeout=20)
|
| 28 |
response.raise_for_status() # Raise an exception for bad status codes
|
| 29 |
|
| 30 |
# Convert the HTML content to Markdown
|
|
@@ -43,4 +47,4 @@ class VisitWebpageTool(Tool):
|
|
| 43 |
return f"An unexpected error occurred: {str(e)}"
|
| 44 |
|
| 45 |
def __init__(self, *args, **kwargs):
|
| 46 |
-
self.is_initialized = False
|
|
|
|
| 11 |
inputs = {'url': {'type': 'string', 'description': 'The url of the webpage to visit.'}}
|
| 12 |
output_type = "string"
|
| 13 |
|
| 14 |
+
HEADERS = {
|
| 15 |
+
"User-Agent": "Mozilla/5.0 (compatible; ResearchAgent/1.0; +https://huggingface.co/spaces/Realistvik0/First_agent_template)"
|
| 16 |
+
}
|
| 17 |
+
|
| 18 |
def forward(self, url: str) -> str:
|
| 19 |
try:
|
| 20 |
import requests
|
|
|
|
| 27 |
"You must install packages `markdownify` and `requests` to run this tool: for instance run `pip install markdownify requests`."
|
| 28 |
) from e
|
| 29 |
try:
|
| 30 |
+
# Send a GET request to the URL with a real User-Agent and a 20-second timeout
|
| 31 |
+
response = requests.get(url, headers=self.HEADERS, timeout=20)
|
| 32 |
response.raise_for_status() # Raise an exception for bad status codes
|
| 33 |
|
| 34 |
# Convert the HTML content to Markdown
|
|
|
|
| 47 |
return f"An unexpected error occurred: {str(e)}"
|
| 48 |
|
| 49 |
def __init__(self, *args, **kwargs):
|
| 50 |
+
self.is_initialized = False
|