Files
Scrapling/scrapling/engines/toolbelt/navigation.py
T
Karim shoair 145c03daff Big structure changes (check commit description)
- Moved most of the parser functions/files to the core package.
- Converted tools file to a package and made separate files for similar functions.
- Now all fetcher engines return a Response object
- Instead of selecting an engine to use and passing config to it, we have separate fetcher classes so the user can choose what to use while importing.
- I added a new custom fetcher so the user can create and use an engine.
- More...
2024-11-03 01:04:02 +02:00

44 lines
1.2 KiB
Python

"""
Functions related to files and URLs
"""
import os
from urllib.parse import urlparse, urlencode
def construct_websocket_url(base_url, query_params):
# Validate the base URL structure
try:
parsed = urlparse(base_url)
# Check scheme
if parsed.scheme not in ('ws', 'wss'):
raise ValueError("URL must use 'ws://' or 'wss://' scheme")
# Validate hostname and port
if not parsed.netloc:
raise ValueError("Invalid hostname")
# Ensure path starts with /
path = parsed.path
if not path.startswith('/'):
path = '/' + path
# Reconstruct the base URL with validated parts
validated_base = f"{parsed.scheme}://{parsed.netloc}{path}"
# Add query parameters
if query_params:
query_string = urlencode(query_params)
return f"{validated_base}?{query_string}"
return validated_base
except Exception as e:
raise ValueError(f"Invalid WebSocket URL: {str(e)}")
def js_bypass_path(filename):
current_directory = os.path.dirname(__file__)
return os.path.join(current_directory, 'bypasses', filename)