00_readme.md

Get YouTube channel all public playlist items

./get_ch_playlists.py

store json and csv

/csv/
    /channel_id/
               /date/
                    /channel_playlists.csv
                    /playlist_id.csv
/json/
     /channel_id/
                /date/
                     /channel_playlists.json
                     /playlist_id.json

select channel_playlists.csv item for diff.

• publishedAt リスト作成日
• id
• itemCount
• title

select playlist_id.csv item for diff.

• publishedAt アイテム追加日
• id
• videoId
• title

prepare

install googleapiclient

$ pip3 install google-api-client

config : api_key and channel list

$ copy > config.json
{
  "api_key" : "your_api_key",
  "channel" : "channel_id"
}

get_ch_playlist.py

# -*- coding: utf-8 -*-

import os

import googleapiclient.discovery
import googleapiclient.errors
import json
import pandas as pd
import pprint
import datetime

scopes = ["https://www.googleapis.com/auth/youtube.readonly"]

def get_playlists(youtube, ch):

    token=''
    l = []

    while True:

        request = youtube.playlists().list(
            part="snippet,contentDetails",
            channelId=ch,
            maxResults=25,
            pageToken=token
        )
        response = request.execute()

        pprint.pprint(response['pageInfo'])
        d = response['items']
        l.extend(d)

        # if exist next page
        if response.get('nextPageToken'):
            token = response['nextPageToken']
        else:
            break

    return l

def get_playlist_items(youtube, plid):

    token = ''
    l = []
    #df = pd.DataFrame(index=[])

    while True:
        request = youtube.playlistItems().list(
            part="snippet",
            playlistId=plid,
            maxResults=50,
            pageToken=token
        )
        response = request.execute()

        pprint.pprint(response['pageInfo'])
        #pprint.pprint(response)
        d = response['items']

        #pprint.pprint(d)
        #df = df.append(pd.DataFrame(d))
        l.extend(d)

        # if exist next page
        if response.get('nextPageToken'):
            token = response['nextPageToken']
        else:
            break

        #pprint.pprint(l)
        #break # for debug

    return l

def save_channel_playlists(csvp, jsonp, playlists):

    json_path = jsonp + '/channel_playlists.json'
    f = open(json_path,'w')
    json.dump(playlists, f, indent=4, ensure_ascii=False)

    csv_path = csvp + '/channel_playlists.csv'

    l = []
    for i in playlists:
        #pprint.pprint(i)
        d = dict(publishedAt = i['snippet']['publishedAt'],
                 id          = i['id'],
                 itemCount   = i['contentDetails']['itemCount'],
                 title       = i['snippet']['title'])
        l.append(d)

    pd.DataFrame(l).to_csv(csv_path, index=False)

def save_playlist(csvp, jsonp, plid, items):

    json_path = jsonp +'/' + plid + ".json" 
    f = open(json_path,'w')
    json.dump(items, f, indent=4, ensure_ascii=False)

    csv_path = csvp + '/' + plid + ".csv"

    l = []
    for i in items:
        #pprint.pprint(i)
        d = dict(publishedAt = i['snippet']['publishedAt'],
                 id          = i['id'],
                 videoId     = i['snippet']['resourceId']['videoId'],
                 title       = i['snippet']['title'] )
        l.append(d)

    #pprint.pprint(l)
    pd.DataFrame(l).to_csv(csv_path, index=False)

def main():

    api_service_name = "youtube"
    api_version = "v3"

    json_open = open('config.json', 'r')
    json_load = json.load(json_open)
    api_key = json_load['api_key']
    ch = json_load['channel']
    print(ch)

    youtube = googleapiclient.discovery.build(
        api_service_name, api_version, developerKey=api_key)

    date = datetime.datetime.now().strftime("%Y%m%d")

    csv_path = 'csv/' + ch +  '/' + date
    os.makedirs(csv_path, exist_ok=True)
    json_path = 'json/' + ch +  '/' + date
    os.makedirs(json_path, exist_ok=True)

    playlists = get_playlists(youtube, ch)
    #pprint.pprint(playlists)
    save_channel_playlists(csv_path, json_path, playlists)

    #plid = playlists[1]['id']

    for p in playlists:
        plid = p['id']
        print(plid)
        items = get_playlist_items(youtube, plid)
        #pprint.pprint(items)
        save_playlist(csv_path, json_path, plid, items)

        #break # for debug

if __name__ == "__main__":
    main()