convert_cookie_for_youtube.py

from urllib.parse import urlparse
import time

from test import get_cookies


def infer_base_domain(original_source_url):
    """
    Infers a correct base domain for YouTube/Google cookies from a
    potentially problematic original_source_url.
    Prioritizes common YouTube/Google domains.
    """
    if "youtube.com" in original_source_url:
        return "youtube.com"
    elif "google.com" in original_source_url:
        return "google.com"
    elif "youtube.com" in original_source_url:  # Catches other YouTube domains if any
        return "youtube.com"
    elif "googleusercontent.com" in original_source_url:  # Generic googleusercontent fallback
        return "googleusercontent.com"

    # Fallback if no specific match, but this might still give bad results
    # if the original_source_url itself is completely unrecognizable.
    parsed_url = urlparse(original_source_url)
    if parsed_url.hostname:
        return parsed_url.hostname
    return "unknown.com"  # Should ideally not be hit for YouTube/Google cookies


def convert_raw_cookie_string_to_netscape_fixed_domain(raw_cookie_str, original_source_url):
    """
    Converts a raw cookie string to Netscape format, inferring a correct
    base domain for YouTube/Google cookies.
    """
    base_domain = infer_base_domain(original_source_url)

    lines = [
        "# Netscape HTTP Cookie File",
        "# http://curl.haxx.se/rfc/cookie_spec.html",
        "# This file was generated by script - Domain inferred for YouTube/Google compatibility."
    ]

    for raw_cookie in raw_cookie_str.split(';'):
        cookie = raw_cookie.strip().split('=', 1)
        if len(cookie) != 2:
            continue

        name, value = cookie[0], cookie[1]

        # Determine domain field. Use a leading dot for general cookies.
        # This handles the . prefix for subdomains correctly.
        domain_field = f".{base_domain}"

        # Heuristics for HttpOnly based on common cookie prefixes
        # Note: True HttpOnly status isn't available from raw string, this is a guess.
        is_httponly = name.startswith("__Secure-") or name.startswith("__Host-") or name.startswith(
            "SID") or name.startswith("SSID") or name.startswith("LSID")

        if is_httponly:
            # Netscape format for HttpOnly is #HttpOnly_.domain
            # Ensure we don't duplicate the leading dot if it's already there from base_domain
            if domain_field.startswith("."):
                domain_field = f"#HttpOnly_{domain_field}"
            else:
                domain_field = f"#HttpOnly_.{domain_field}"

        # The 'flag' column (Include Subdomains) is generally TRUE for Netscape format.
        include_subdomains = "TRUE"

        # The 'path' column is usually '/' for general site cookies.
        path = "/"

        # The 'secure' column. Guess based on __Secure- prefix, but better if known from source.
        secure = "TRUE" if name.startswith("__Secure-") else "FALSE"

        # Expiration: A fixed, far-future date is often sufficient for yt-dlp.
        expires = str(int(time.time()) + 180 * 24 * 3600)  # 180 days from now

        # Construct the line
        line = f"{domain_field}\t{include_subdomains}\t{path}\t{secure}\t{expires}\t{name}\t{value}"
        lines.append(line)

    return lines


# --- Main Execution Block (adapt to your API structure) ---
# Assuming 'cookie_entry' is received at your API endpoint
# For testing purposes, using your provided cookie_entry_data:
# from test import get_cookies # Uncomment if get_cookies is defined elsewhere

cookie_entry_data = get_cookies()

raw_str = cookie_entry_data['youtube']['raw_cookies']
source = cookie_entry_data['youtube']['source_url']

# Convert to Netscape format
output_lines = convert_raw_cookie_string_to_netscape_fixed_domain(raw_str, source)

# Save to file (this would happen on your API server)
with open("cookies.txt", "w", encoding="utf-8") as f:
    f.write("\n".join(output_lines))

print("✅ Saved as cookies.txt in Netscape format with inferred domains.")