Python to extract a YouTube video's chapters

gistfile1.txt

import requests
import json
from typing import Optional

class YoutubeChaptersGetter:
    def __init__(self, youtube_url: str = "https://youtube.com"):
        self.youtube = youtube_url

    def _get_video(self, video_id: str) -> Optional[str]:
        try:
            response = requests.get(f"{self.youtube}/watch?v={video_id}")
            response.raise_for_status()
            return response.text
        except requests.RequestException as error:
            print(f"getVideo err: {error}")
            return None

    def _get_script(self, html: str) -> Optional[str]:
        try:
            script_tag_start = "var ytInitialData = "
            script_tag_end = "</script>"
            start_index = html.find(script_tag_start)
            if start_index == -1:
                return None
            end_index = html.find(script_tag_end, start_index)
            if end_index == -1:
                return None
            yt_initial_data = html[start_index + len(script_tag_start):end_index].strip()
            if yt_initial_data.endswith(";"):
                return yt_initial_data[:-1]
            return yt_initial_data
        except Exception as error:
            print(f"getScript err: {error}")
            return None

    def _millis_to_time(self, millis: int) -> str:
        """Convert milliseconds to a time string like '5:22'."""
        seconds = millis // 1000
        minutes = seconds // 60
        seconds = seconds % 60
        return f"{minutes}:{seconds:02d}"

    def get_chapters(self, video_id: str) -> str:
        """
        Get chapters for a YouTube video ID and return as JSON string.
        Returns chapter data with title and time on success, empty JSON array on failure.
        """
        try:
            html = self._get_video(video_id)
            if not html:
                print("No HTML content retrieved")
                return "[]"

            script_data = self._get_script(html)
            if not script_data:
                print("No script data extracted")
                return "[]"

            yt_initial_data = json.loads(script_data)
            
            # Navigate to the chapters location
            try:
                player_overlays = yt_initial_data["playerOverlays"]
                overlay_renderer = player_overlays["playerOverlayRenderer"]
                decorated_bar = overlay_renderer["decoratedPlayerBarRenderer"]["decoratedPlayerBarRenderer"]
                player_bar = decorated_bar["playerBar"]
                multi_markers = player_bar["multiMarkersPlayerBarRenderer"]
                markers_map = multi_markers["markersMap"]
                
                if not markers_map or len(markers_map) == 0:
                    print("No markersMap data found")
                    return "[]"
                
                chapters_data = markers_map[0]["value"]["chapters"]
                
                chapters = []
                for chapter_item in chapters_data:
                    chapter = chapter_item["chapterRenderer"]
                    time_millis = chapter["timeRangeStartMillis"]
                    time_str = self._millis_to_time(time_millis)
                    
                    chapters.append({
                        "title": chapter["title"]["simpleText"],
                        "time": time_str
                    })
                
                # Return as JSON string
                return json.dumps(chapters)
            
            except KeyError as key_err:
                print(f"Key error in path: Missing key {key_err}")
                return "[]"

        except json.JSONDecodeError as json_err:
            print(f"JSON parsing error: {json_err}")
            return "[]"
        except Exception as error:
            print(f"Unexpected error in get_chapters: {error}")
            return "[]"

# Singleton instance for use in other applications
youtube_chapters_getter = YoutubeChaptersGetter()

def get_youtube_chapters(video_id: str) -> str:
    """
    Public function to get YouTube chapters as JSON string.
    Takes a video ID and returns chapters or empty JSON array.
    """
    return youtube_chapters_getter.get_chapters(video_id)

# Example usage when run directly
if __name__ == "__main__":
    result = get_youtube_chapters("YOUTUBE_VIDEO_ID_COMES_HERE")
    print(result)