ytstats.py

import json
import requests
from tqdm import tqdm

# Stripped some unneeded code from this regarding individual video statistics to save request time.

class YTstats:

    def __init__(self, api_key, channel_id):
        self.api_key = api_key
        self.channel_id = channel_id
        self.channel_statistics = None
        self.video_data = None

    def extract_all(self):
        self.get_channel_statistics()

    def get_channel_statistics(self):
        """Extract the channel statistics"""
        print('get channel statistics...')
        url = f'https://www.googleapis.com/youtube/v3/channels?part=statistics&id={self.channel_id}&key={self.api_key}'
        pbar = tqdm(total=1)
        
        json_url = requests.get(url)
        data = json.loads(json_url.text)
        try:
            data = data['items'][0]['statistics']
        except KeyError:
            print('Could not get channel statistics')
            data = {}

        self.channel_statistics = data
        pbar.update()
        pbar.close()
        return data

    

    def _get_single_video_data(self, video_id, part):
        """
        Extract further information for a single video
        parts can be: 'snippet', 'statistics', 'contentDetails', 'topicDetails'
        """

        url = f"https://www.googleapis.com/youtube/v3/videos?part={part}&id={video_id}&key={self.api_key}"
        json_url = requests.get(url)
        data = json.loads(json_url.text)
        try:
            data = data['items'][0][part]
        except KeyError as e:
            print(f'Error! Could not get {part} part of data: \n{data}')
            data = dict()
        return data

    def _get_channel_content(self, limit=None, check_all_pages=True):
        """
        Extract all videos and playlists, can check all available search pages
        channel_videos = videoId: title, publishedAt
        channel_playlists = playlistId: title, publishedAt
        return channel_videos, channel_playlists
        """
        url = f"https://www.googleapis.com/youtube/v3/search?key={self.api_key}&channelId={self.channel_id}&part=snippet,id&order=date"
        if limit is not None and isinstance(limit, int):
            url += "&maxResults=" + str(limit)

        vid, pl, npt = self._get_channel_content_per_page(url)
        idx = 0
        while(check_all_pages and npt is not None and idx < 10):
            nexturl = url + "&pageToken=" + npt
            next_vid, next_pl, npt = self._get_channel_content_per_page(nexturl)
            vid.update(next_vid)
            pl.update(next_pl)
            idx += 1

        return vid, pl

    def _get_channel_content_per_page(self, url):
        """
        Extract all videos and playlists per page
        return channel_videos, channel_playlists, nextPageToken
        """
        json_url = requests.get(url)
        data = json.loads(json_url.text)
        channel_videos = dict()
        channel_playlists = dict()
        if 'items' not in data:
            print('Error! Could not get correct channel data!\n', data)
            return channel_videos, channel_videos, None

        nextPageToken = data.get("nextPageToken", None)

        item_data = data['items']
        for item in item_data:
            try:
                kind = item['id']['kind']
                published_at = item['snippet']['publishedAt']
                title = item['snippet']['title']
                if kind == 'youtube#video':
                    video_id = item['id']['videoId']
                    channel_videos[video_id] = {'publishedAt': published_at, 'title': title}
                elif kind == 'youtube#playlist':
                    playlist_id = item['id']['playlistId']
                    channel_playlists[playlist_id] = {'publishedAt': published_at, 'title': title}
            except KeyError as e:
                print('Error! Could not extract data from item:\n', item)

        return channel_videos, channel_playlists, nextPageToken

    def dump(self):
        """Dumps channel statistics and video data in a single json file"""
        if self.channel_statistics is None:
            print('data is missing!\nCall get_channel_statistics() first!')
            return

        fused_data = {self.channel_id: {"channel_statistics": self.channel_statistics}}

        filename = "youtube" + '.json'
        with open(filename, 'w') as f:
            json.dump(fused_data, f, indent=4)
        
        print('file dumped to', filename)