run.py

"""
Main entry point for YouTube Study Notes AI.
Provides CLI interface and server startup.
"""

import os
import sys
import subprocess
import argparse
from pathlib import Path

from src.utils.logger import setup_logger
from src.utils.config import settings

logger = setup_logger(__name__)


def configure_proxy():
    proxy_url = os.environ.get("PROXY_URL", "").strip()
    if proxy_url:
        os.environ["HTTP_PROXY"] = proxy_url
        os.environ["HTTPS_PROXY"] = proxy_url
        os.environ["ALL_PROXY"] = proxy_url
        os.environ["http_proxy"] = proxy_url
        os.environ["https_proxy"] = proxy_url
        os.environ["all_proxy"] = proxy_url
        logger.info("✅ Global proxy configured from PROXY_URL")


def check_youtube_connectivity():
    import urllib.request
    import urllib.error

    try:
        urllib.request.urlopen("https://www.youtube.com", timeout=5)
        logger.info("✅ YouTube is reachable")
    except (urllib.error.URLError, Exception):
        logger.warning(
            "WARNING: YouTube is not reachable from this host. "
            "Set PROXY_URL or RAPIDAPI_KEY env vars to enable YouTube features."
        )


def check_environment():
    """Log key dependency versions to confirm runtime environment."""
    configure_proxy()
    check_youtube_connectivity()
    # Check ffmpeg (still used by audio processing utilities)
    try:
        ffmpeg_out = subprocess.check_output(
            ["ffmpeg", "-version"], stderr=subprocess.STDOUT
        ).decode().splitlines()[0]
        logger.info(f"✅ ffmpeg available: {ffmpeg_out}")
    except (subprocess.CalledProcessError, FileNotFoundError):
        logger.warning("❌ ffmpeg NOT found — audio extraction will fail")

    # Verify Supadata API key is configured
    supadata_key = os.environ.get("SUPADATA_API_KEY", "").strip()
    if supadata_key:
        logger.info("✅ SUPADATA_API_KEY is set")
    else:
        logger.warning("❌ SUPADATA_API_KEY is NOT set — transcript extraction will fail")


def run_server():
    """Start the FastAPI server with CORS enabled for Flutter Web."""
    import uvicorn
    from fastapi.middleware.cors import CORSMiddleware
    from src.api.main import app

    check_environment()

    logger.info("Configuring CORS for Flutter Web...")

    app.add_middleware(
        CORSMiddleware,
        allow_origins=["*"],
        allow_credentials=True,
        allow_methods=["*"],
        allow_headers=["*"],
    )

    logger.info("Starting YouTube Study Notes AI server...")
    logger.info(f"Server will be available at http://{settings.api_host}:{settings.api_port}")
    logger.info(f"API Documentation: http://{settings.api_host}:{settings.api_port}/docs")

    uvicorn.run(app, host=settings.api_host, port=settings.api_port, log_level="info")


def run_cli(youtube_url: str, output_file: str = None):
    from src.transcription.downloader import YouTubeDownloader

    check_environment()

    downloader = YouTubeDownloader()
    transcript = downloader.get_transcript(youtube_url)
    print(transcript)


def main():
    """Main entry point with argument parsing."""
    parser = argparse.ArgumentParser(
        description="YouTube Study Notes AI - Generate structured notes from educational videos"
    )
    parser.add_argument(
        "mode",
        choices=["server", "cli"],
        help="Run mode: server (API + web UI) or cli (direct processing)",
    )
    parser.add_argument("--url", type=str, help="YouTube video URL (required for cli mode)")
    parser.add_argument("--output", type=str, help="Output file path (optional for cli mode)")

    args = parser.parse_args()

    if args.mode == "server":
        run_server()
    elif args.mode == "cli":
        if not args.url:
            print("Error: --url is required for cli mode")
            sys.exit(1)
        run_cli(args.url, args.output)


if __name__ == "__main__":
    main()