hf_scrapper.py

import os
import requests
import json
import urllib.request
from requests.exceptions import RequestException

def get_system_proxies():
    try:
        proxies = urllib.request.getproxies()
        print("System proxies:", proxies)
        return {
            "http": proxies.get("http"),
            "https": proxies.get("http")
        }
    except Exception as e:
        print(f"Error getting system proxies: {e}")
        return {}

def stream_file(file_url, token, proxies):
    print(f"Streaming file from URL: {file_url} with proxies: {proxies}")
    try:
        response = requests.get(file_url, headers={'Authorization': f'Bearer {token}'}, proxies=proxies, stream=True)
        response.raise_for_status()
        for chunk in response.iter_content(chunk_size=8192):
            if chunk:
                yield chunk
    except RequestException as e:
        print(f"Error streaming file: {e}")
        yield b''  # Return empty bytes to indicate an error

def download_file(file_url, token, output_path, proxies):
    print(f"Downloading file from URL: {file_url} with proxies: {proxies}")
    try:
        response = requests.get(file_url, headers={'Authorization': f'Bearer {token}'}, proxies=proxies, stream=True)
        response.raise_for_status()
        with open(output_path, 'wb') as f:
            for chunk in response.iter_content(chunk_size=8192):
                if chunk:
                    f.write(chunk)
        print(f'File {output_path} downloaded successfully.')
    except RequestException as e:
        print(f"Error downloading file: {e}")
    except IOError as e:
        print(f"Error writing file {output_path}: {e}")

def get_file_structure(repo, token, path="", proxies=None):
    api_url = f"https://huggingface.co/api/models/{repo}/tree/main/{path}"
    headers = {'Authorization': f'Bearer {token}'}
    print(f"Fetching file structure from URL: {api_url} with proxies: {proxies}")
    try:
        response = requests.get(api_url, headers=headers, proxies=proxies)
        response.raise_for_status()
        return response.json()
    except RequestException as e:
        print(f"Error fetching file structure: {e}")
        return []

def write_file_structure_to_json(file_structure, file_path):
    try:
        with open(file_path, 'w') as json_file:
            json.dump(file_structure, json_file, indent=2)
        print(f'File structure written to {file_path}')
    except IOError as e:
        print(f"Error writing file structure to JSON: {e}")