youtube-content

1#!/usr/bin/env python32"""3Fetch a YouTube video transcript and output it as structured JSON.4 5Usage:6    uv run python3 fetch_transcript.py <url_or_video_id> [--language en,tr] [--timestamps]7 8Output (JSON):9    {10        "video_id": "...",11        "language": "en",12        "segments": [{"text": "...", "start": 0.0, "duration": 2.5}, ...],13        "full_text": "complete transcript as plain text",14        "timestamped_text": "00:00 first line\n00:05 second line\n..."15    }16 17Install dependency:  uv pip install youtube-transcript-api18"""19 20import argparse21import json22import re23import sys24 25 26def extract_video_id(url_or_id: str) -> str:27    """Extract the 11-character video ID from various YouTube URL formats."""28    url_or_id = url_or_id.strip()29    patterns = [30        r'(?:v=|youtu\.be/|shorts/|embed/|live/)([a-zA-Z0-9_-]{11})',31        r'^([a-zA-Z0-9_-]{11})$',32    ]33    for pattern in patterns:34        match = re.search(pattern, url_or_id)35        if match:36            return match.group(1)37    return url_or_id38 39 40def format_timestamp(seconds: float) -> str:41    """Convert seconds to HH:MM:SS or MM:SS format."""42    total = int(seconds)43    h, remainder = divmod(total, 3600)44    m, s = divmod(remainder, 60)45    if h > 0:46        return f"{h}:{m:02d}:{s:02d}"47    return f"{m}:{s:02d}"48 49 50def fetch_transcript(video_id: str, languages: list = None):51    """Fetch transcript segments from YouTube.52 53    Returns a list of dicts with 'text', 'start', and 'duration' keys.54    Compatible with youtube-transcript-api v1.x.55    """56    try:57        from youtube_transcript_api import YouTubeTranscriptApi58    except ImportError:59        print("Error: youtube-transcript-api not installed. Run: uv pip install youtube-transcript-api",60              file=sys.stderr)61        sys.exit(1)62 63    api = YouTubeTranscriptApi()64    if languages:65        result = api.fetch(video_id, languages=languages)66    else:67        result = api.fetch(video_id)68 69    # v1.x returns FetchedTranscriptSnippet objects; normalize to dicts70    return [71        {"text": seg.text, "start": seg.start, "duration": seg.duration}72        for seg in result73    ]74 75 76def main():77    parser = argparse.ArgumentParser(description="Fetch YouTube transcript as JSON")78    parser.add_argument("url", help="YouTube URL or video ID")79    parser.add_argument("--language", "-l", default=None,80                        help="Comma-separated language codes (e.g. en,tr). Default: auto")81    parser.add_argument("--timestamps", "-t", action="store_true",82                        help="Include timestamped text in output")83    parser.add_argument("--text-only", action="store_true",84                        help="Output plain text instead of JSON")85    args = parser.parse_args()86 87    video_id = extract_video_id(args.url)88    languages = [l.strip() for l in args.language.split(",")] if args.language else None89 90    try:91        segments = fetch_transcript(video_id, languages)92    except Exception as e:93        error_msg = str(e)94        if "disabled" in error_msg.lower():95            print(json.dumps({"error": "Transcripts are disabled for this video."}))96        elif "no transcript" in error_msg.lower():97            print(json.dumps({"error": "No transcript found. Try specifying a language with --language."}))98        else:99            print(json.dumps({"error": error_msg}))100        sys.exit(1)101 102    full_text = " ".join(seg["text"] for seg in segments)103    timestamped = "\n".join(104        f"{format_timestamp(seg['start'])} {seg['text']}" for seg in segments105    )106 107    if args.text_only:108        print(timestamped if args.timestamps else full_text)109        return110 111    result = {112        "video_id": video_id,113        "segment_count": len(segments),114        "duration": format_timestamp(segments[-1]["start"] + segments[-1]["duration"]) if segments else "0:00",115        "full_text": full_text,116    }117    if args.timestamps:118        result["timestamped_text"] = timestamped119 120    print(json.dumps(result, ensure_ascii=False, indent=2))121 122 123if __name__ == "__main__":124    main()125