Medios-Macina/Provider/youtube.py

from __future__ import annotations

import sys
from typing import Any, Dict, List, Optional

from ProviderCore.base import Provider, SearchResult
from SYS.logger import log


class YouTube(Provider):
    """Search provider for YouTube using the yt_dlp Python package."""

    def search(
        self,
        query: str,
        limit: int = 10,
        filters: Optional[Dict[str, Any]] = None,
        **kwargs: Any,
    ) -> List[SearchResult]:
        # Use the yt_dlp Python module (installed via requirements.txt).
        try:
            import yt_dlp  # type: ignore

            ydl_opts: Dict[str, Any] = {"quiet": True, "skip_download": True, "extract_flat": True}
            with yt_dlp.YoutubeDL(ydl_opts) as ydl:  # type: ignore[arg-type]
                search_query = f"ytsearch{limit}:{query}"
                info = ydl.extract_info(search_query, download=False)
                entries = info.get("entries") or []
                results: List[SearchResult] = []
                for video_data in entries[:limit]:
                    title = video_data.get("title", "Unknown")
                    video_id = video_data.get("id", "")
                    url = video_data.get("url") or f"https://youtube.com/watch?v={video_id}"
                    uploader = video_data.get("uploader", "Unknown")
                    duration = video_data.get("duration", 0)
                    view_count = video_data.get("view_count", 0)

                    duration_str = (
                        f"{int(duration // 60)}:{int(duration % 60):02d}" if duration else ""
                    )
                    views_str = f"{view_count:,}" if view_count else ""

                    results.append(
                        SearchResult(
                            table="youtube",
                            title=title,
                            path=url,
                            detail=f"By: {uploader}",
                            annotations=[duration_str, f"{views_str} views"],
                            media_kind="video",
                            columns=[
                                ("Title", title),
                                ("Uploader", uploader),
                                ("Duration", duration_str),
                                ("Views", views_str),
                            ],
                            full_metadata={
                                "video_id": video_id,
                                "uploader": uploader,
                                "duration": duration,
                                "view_count": view_count,
                            },
                        )
                    )
                return results
        except Exception:
            log("[youtube] yt_dlp import failed", file=sys.stderr)
            return []

    def validate(self) -> bool:
        try:
            import yt_dlp  # type: ignore

            return True
        except Exception:
            return False