memento-flashcards

1#!/usr/bin/env python32"""Fetch YouTube transcripts for Memento quiz generation.3 4Requires: pip install youtube-transcript-api5The quiz question *generation* is done by the agent's LLM — this script only fetches transcripts.6"""7 8import argparse9import json10import re11import sys12 13 14def _out(obj: object) -> None:15    json.dump(obj, sys.stdout, indent=2, ensure_ascii=False)16    sys.stdout.write("\n")17 18 19def _normalize_segments(segments: list) -> str:20    parts = []21    for seg in segments:22        text = str(seg.get("text", "")).strip()23        if text:24            parts.append(text)25    return re.sub(r"\s+", " ", " ".join(parts)).strip()26 27 28def cmd_fetch(args: argparse.Namespace) -> None:29    try:30        import youtube_transcript_api  # noqa: F81131    except ImportError:32        _out({33            "ok": False,34            "error": "missing_dependency",35            "message": "Run: pip install youtube-transcript-api",36        })37        sys.exit(1)38 39    video_id = args.video_id40    languages = ["en", "en-US", "en-GB", "en-CA", "en-AU"]41 42    api = youtube_transcript_api.YouTubeTranscriptApi()43    try:44        raw = api.fetch(video_id, languages=languages)45    except Exception as exc:46        error_type = type(exc).__name__47        _out({48            "ok": False,49            "error": "transcript_unavailable",50            "error_type": error_type,51            "message": f"Could not fetch transcript for {video_id}: {exc}",52        })53        sys.exit(1)54 55    segments = raw56    if hasattr(raw, "to_raw_data"):57        segments = raw.to_raw_data()58 59    text = _normalize_segments(segments)60    if not text:61        _out({62            "ok": False,63            "error": "empty_transcript",64            "message": f"Transcript for {video_id} contained no usable text.",65        })66        sys.exit(1)67 68    _out({69        "ok": True,70        "video_id": video_id,71        "transcript": text,72    })73 74 75def main() -> None:76    parser = argparse.ArgumentParser(description="Memento YouTube transcript fetcher")77    sub = parser.add_subparsers(dest="command", required=True)78 79    p_fetch = sub.add_parser("fetch", help="Fetch transcript for a video")80    p_fetch.add_argument("video_id", help="YouTube video ID")81 82    args = parser.parse_args()83    if args.command == "fetch":84        cmd_fetch(args)85 86 87if __name__ == "__main__":88    main()89