#!/usr/bin/env python3
"""Auto-rotate Google Scholar proxy for paper-search-mcp.

Fetches fresh free proxies, tests them against scholar.google.com,
and updates ~/.config/paper-search-mcp/.env with the first working one.

Usage:
    python3 rotate_gs_proxy.py          # one-shot test & update
    python3 rotate_gs_proxy.py --force  # always replace (even if current works)
"""

import os
import sys
import time
import requests

ENV_PATH = os.path.expanduser("~/.config/paper-search-mcp/.env")
SCHOLAR_URL = "https://scholar.google.com"
TEST_TIMEOUT = 5
PROXY_LISTS = [
    "https://raw.githubusercontent.com/TheSpeedX/SOCKS-List/master/http.txt",
    "https://raw.githubusercontent.com/ShiftyTR/Proxy-List/master/http.txt",
    "https://raw.githubusercontent.com/clarketm/proxy-list/master/proxy-list-raw.txt",
]
MAX_TEST = 150


def read_env() -> dict:
    env = {}
    if os.path.exists(ENV_PATH):
        with open(ENV_PATH) as f:
            for line in f:
                line = line.strip()
                if "=" in line:
                    k, v = line.split("=", 1)
                    env[k.strip()] = v.strip()
    return env


def write_env(updates: dict):
    lines = []
    if os.path.exists(ENV_PATH):
        with open(ENV_PATH) as f:
            lines = f.readlines()
    # Update in-place
    new_lines = []
    for line in lines:
        stripped = line.strip()
        for k, v in updates.items():
            if stripped.startswith(f"{k}="):
                new_lines.append(f"{k}={v}\n")
                break
        else:
            new_lines.append(line)
    # Append missing keys
    for k, v in updates.items():
        if not any(ln.strip().startswith(f"{k}=") for ln in new_lines):
            new_lines.append(f"{k}={v}\n")
    with open(ENV_PATH, "w") as f:
        f.writelines(new_lines)
    return len(updates)


def test_proxy(proxy_url: str) -> bool:
    proxies = {"http": proxy_url, "https": proxy_url}
    try:
        r = requests.get(SCHOLAR_URL, proxies=proxies, timeout=TEST_TIMEOUT,
                         headers={"User-Agent": "Mozilla/5.0"})
        return r.status_code == 200
    except Exception:
        return False


def find_working_proxy() -> str | None:
    seen = set()
    for list_url in PROXY_LISTS:
        try:
            r = requests.get(list_url, timeout=10)
            raw = r.text
        except Exception:
            continue
        lines = [l.strip() for l in raw.split("\n") if l.strip() and ":" in l]
        count = 0
        for entry in lines:
            if entry in seen:
                continue
            seen.add(entry)
            proxy_url = f"http://{entry}"
            if test_proxy(proxy_url):
                print(f"✓ Working proxy: {proxy_url}")
                return proxy_url
            count += 1
            if count >= MAX_TEST:
                break
    return None


def main():
    force = "--force" in sys.argv
    env = read_env()

    current = env.get("PAPER_SEARCH_MCP_GOOGLE_SCHOLAR_PROXY_URL", "")

    # If current proxy works and not forcing, keep it
    if not force and current and test_proxy(current):
        print(f"✓ Current proxy works: {current}")
        return

    if current:
        print(f"✗ Current proxy failed: {current}")

    print("🔍 Searching for a working free proxy...")
    proxy = find_working_proxy()
    if proxy:
        write_env({"PAPER_SEARCH_MCP_GOOGLE_SCHOLAR_PROXY_URL": proxy})
        print(f"✅ Updated to: {proxy}")
    else:
        print("❌ No working proxy found.")
        sys.exit(1)


if __name__ == "__main__":
    main()