Create visit_webpage_tool.py
Browse files- visit_webpage_tool.py +35 -0
visit_webpage_tool.py
ADDED
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from transformers import Tool
|
2 |
+
import requests
|
3 |
+
from markdownify import markdownify as md
|
4 |
+
from requests.exceptions import RequestException
|
5 |
+
import re
|
6 |
+
|
7 |
+
class VisitWebpageTool(Tool):
|
8 |
+
name = "visit_webpage"
|
9 |
+
description = "Visits a webpage at the given URL and returns its content as a markdown string."
|
10 |
+
inputs = {
|
11 |
+
"url": {
|
12 |
+
"type": "text",
|
13 |
+
"description": "The URL of the webpage to visit.",
|
14 |
+
}
|
15 |
+
}
|
16 |
+
output_type = "text"
|
17 |
+
|
18 |
+
def forward(self, url: str) -> str:
|
19 |
+
try:
|
20 |
+
# Send a GET request to the URL
|
21 |
+
response = requests.get(url)
|
22 |
+
response.raise_for_status() # Raise an exception for bad status codes
|
23 |
+
|
24 |
+
# Convert the HTML content to Markdown
|
25 |
+
markdown_content = md(response.text).strip()
|
26 |
+
|
27 |
+
# Remove multiple line breaks
|
28 |
+
markdown_content = re.sub(r"\n{3,}", "\n\n", markdown_content)
|
29 |
+
|
30 |
+
return markdown_content
|
31 |
+
|
32 |
+
except RequestException as e:
|
33 |
+
return f"Error fetching the webpage: {str(e)}"
|
34 |
+
except Exception as e:
|
35 |
+
return f"An unexpected error occurred: {str(e)}"
|