Get YouTube Live chat field in real time with API

Caution

YouTube Terms of Service Use the code below at your own risk.

Get in real time using YouTube API

Overview

Recently, there is a story about spoofing in the YouTube chat column and a leak in one column of the person who spascha. As one of the solutions, I wrote below how to get the chat field using YouTube API.

output

[by username 1 channel URL 1]
Comment 1
[by username 2 channel URL 2]
Comment 2
[by username 3 channel URL 3]
Amount from username 3: "Comment 3"← Spacha looks like this

start :  2020-06-07T05:21:49.871000Z ← The first time of the acquired chat group: +9 hours in Japan time
end   :  2020-06-07T05:22:02.846000Z   ← 　　　　"the last time

--Official API -Obtain according to Get YouTube API key (as of March 25, 2020) --Consume what is called a quota. The upper limit is 10000 quota / day, and increment application is required for more than that. -quota calculator does not have quota consumption of API used in the following code, so it is unknown how much it consumes. --2 quota for chat ID acquisition, 3.5 quota for chat acquisition? --When I tried to get chat every 15 seconds 240 times (variable slp_time = 15, ʻiter_times = 240 in the mainfunction), the consumption was842. --To check quota consumption, Google Cloud Platform> Project created by API key acquisition> Something like" three "in the upper left> IAM and management> Assignment> Service is narrowed down to YouTube Data API v3 --Can be obtained before and after broadcasting ――If the acquisition frequency is too low, is there an omission of acquisition? ―― Below is the code to record all chats including spacha. Rewrite according to the application. --Spacha ismsg ='Amount from Username:" Message "' --EntersupChat for spachas and supSticfor stickers. Just erase the#` and pull out as much as you need. What's inside here

flow

-Get API KEY according to Get YouTube API key (as of March 25, 2020) --Get video ID from YouTube Live URL --https://www.youtube.com/watch?v=***********`` *********** --Get chat ID from video ID -Use this API - {'key': YT_API_KEY, 'id': video_id, 'part': 'liveStreamingDetails'} --Acquire chat fields repeatedly based on chat ID -Use this API - {'key': YT_API_KEY, 'liveChatId': chat_id, 'part': 'id,snippet,authorDetails', 'pageToken': pageToken} --Set pageToken to None for the first time --From the second time onward, specify the previous nextPageToken for pageToken --You can get the difference with this --You can loop by specifying the time, or you can while True until you throw an error (until ctrl + C).

Source code

record_chat.py

import time
import requests
import json

#Pre-acquired YouTube API key
YT_API_KEY = '***************************************'




def get_chat_id(yt_url):
    '''
    https://developers.google.com/youtube/v3/docs/videos/list?hl=ja
    '''
    video_id = yt_url.replace('https://www.youtube.com/watch?v=', '')
    print('video_id : ', video_id)

    url    = 'https://www.googleapis.com/youtube/v3/videos'
    params = {'key': YT_API_KEY, 'id': video_id, 'part': 'liveStreamingDetails'}
    data   = requests.get(url, params=params).json()

    liveStreamingDetails = data['items'][0]['liveStreamingDetails']
    if 'activeLiveChatId' in liveStreamingDetails.keys():
        chat_id = liveStreamingDetails['activeLiveChatId']
        print('get_chat_id done!')
    else:
        chat_id = None
        print('NOT live')

    return chat_id




def get_chat(chat_id, pageToken, log_file):
    '''
    https://developers.google.com/youtube/v3/live/docs/liveChatMessages/list
    '''
    url    = 'https://www.googleapis.com/youtube/v3/liveChat/messages'
    params = {'key': YT_API_KEY, 'liveChatId': chat_id, 'part': 'id,snippet,authorDetails'}
    if type(pageToken) == str:
        params['pageToken'] = pageToken

    data   = requests.get(url, params=params).json()

    try:
        for item in data['items']:
            channelId = item['snippet']['authorChannelId']
            msg       = item['snippet']['displayMessage']
            usr       = item['authorDetails']['displayName']
            #supChat   = item['snippet']['superChatDetails']
            #supStic   = item['snippet']['superStickerDetails']
            log_text  = '[by {}  https://www.youtube.com/channel/{}]\n  {}'.format(usr, channelId, msg)
            with open(log_file, 'a') as f:
                print(log_text, file=f)
                print(log_text)
        print('start : ', data['items'][0]['snippet']['publishedAt'])
        print('end   : ', data['items'][-1]['snippet']['publishedAt'])

    except:
        pass

    return data['nextPageToken']




def main(yt_url):
    slp_time        = 10 #sec
    iter_times      = 90 #Times
    take_time       = slp_time / 60 * iter_times
    print('{}Scheduled to end in minutes'.format(take_time))
    print('work on {}'.format(yt_url))

    log_file = yt_url.replace('https://www.youtube.com/watch?v=', '') + '.txt'
    with open(log_file, 'a') as f:
        print('{}Record the chat field of.'.format(yt_url), file=f)
    chat_id  = get_chat_id(yt_url)

    nextPageToken = None
    for ii in range(iter_times):
        #for jj in [0]:
        try:
            print('\n')
            nextPageToken = get_chat(chat_id, nextPageToken, log_file)
            time.sleep(slp_time)
        except:
            break




if __name__ == '__main__':
    yt_url = input('Input YouTube URL > ')
    main(yt_url)