def __init__(self, api_key, channels):
Session = sessionmaker(bind=engine)
self.logger = getLogger('youtuber.grabber')
self.api = API(api_key=api_key, client_id='', client_secret='')
self.channels = channels
self.session = Session()
def init():
YouTubeController.refresh_token = SecureSettings.get_string(
"youtube_refresh_token")
YouTubeController.api = API(
client_id=SecureSettings.get_string("youtube_client_id"),
client_secret=SecureSettings.get_string("youtube_client_secret"),
api_key=SecureSettings.get_string("youtube_api_key"))
def main():
parser = argparse.ArgumentParser()
parser.add_argument(
'-o',
'--output',
help='Absolute path to output file. Output file can exist.',
required=True)
parser.add_argument('-k',
'--api-key',
help='Youtube API key.',
required=True)
parser.add_argument('-v',
'--visible',
help='Append new videos as visible.',
default=False)
parser.add_argument('-u',
'--update',
help='Update video in output file if it exists.',
default=False)
parser.add_argument(
'--update-attributes',
'--update-attributes',
help=
'Comma separated list of attributes allowed to update. Works only when --update flag is true',
default='description,title,preview')
args = parser.parse_args()
global api
api = API(api_key=args.api_key, client_secret='', client_id='')
original_videos = get_original_videos(args.output)
youtube_videos = fetch_videos()
merged_videos = merge_videos(original_videos, youtube_videos, args.visible,
args.update, args.update_attributes)
save_videos(merged_videos, args.output)
class YoutubeAPI(object):
api_key = None
yt = None
def __init__(self, api_key, *args, **kwargs):
self.api_key = api_key
self.set_up_youtube()
def set_up_youtube(self):
self.yt = API(client_id='', client_secret='', api_key=self.api_key)
def get_youtube_video(self, search_string):
video = self.yt.get('search',
q=search_string,
maxResults=1,
type='video',
order='relevance')
return ("https://www.youtube.com/watch?v=" +
video["items"][0]["id"]["videoId"])
class Grabber:
def __init__(self, api_key, channels):
Session = sessionmaker(bind=engine)
self.logger = getLogger('youtuber.grabber')
self.api = API(api_key=api_key, client_id='', client_secret='')
self.channels = channels
self.session = Session()
def _get(self, resource, part, **kwargs):
try:
result = self.api.get(resource, part=part, **kwargs)
if result['pageInfo']['totalResults'] == 0:
self.logger.error('Resource "{}" is empty for "{}"'.format(
resource, kwargs))
exit(1)
else:
return result['items'], result.get('nextPageToken', None)
except YouTubeException as e:
self.logger.error(e.error_type)
return [], None
def run(self):
for name in self.channels:
channels, _ = self._get('channels', 'id', forUsername=name)
channel_id = channels[0]['id']
playlists, _ = self._get('playlists',
part='id',
channelId=channel_id,
maxResults=50)
playlist_ids = [item['id'] for item in playlists]
for pl_id in playlist_ids:
page_token = None
while True:
pl_items, page_token = self._get(
'playlistItems',
part='snippet,contentDetails',
playlistId=pl_id,
maxResults=50,
pageToken=page_token)
for item in pl_items:
video_id = item['contentDetails']['videoId']
video_published_at = item['contentDetails'].get(
'videoPublishedAt', None)
if not video_published_at:
continue # deleted video
published_at = datetime.strptime(
video_published_at, '%Y-%m-%dT%H:%M:%S.%fZ')
title = item['snippet']['title']
video = self.session.query(Video).get(video_id)
if not video:
self.session.add(
Video(video_id=video_id,
channel=name,
title=title,
published_at=published_at))
self.session.commit()
stats, _ = self._get('videos',
part='statistics',
id=video_id)
try:
video_stat = stats[0]
except IndexError:
continue
count = int(video_stat['statistics']['viewCount'])
likes = int(video_stat['statistics']['likeCount'])
dislikes = int(
video_stat['statistics']['dislikeCount'])
views = Views(video_id, count, likes, dislikes)
self.session.add(views)
self.session.commit()
if not page_token:
break
def add_song_info_from_spotify_url(url: str, username):
details = dict(sp.track(url))
name = details['name']
duration = round(int(details['duration_ms']) / 60000, 2)
cover_url = details['album']['images'][0]['url']
add_song_if_not_exists(name, duration, username, cover_url)
# Youtube
## login
g_client_id = "724786455286-h797c1jpr7nq4ivsqca73gplkbol5u1c.apps.googleusercontent.com"
g_client_secret = "AxRk4DHa9IfLHd4kDMUXuilQ"
g_key = "AIzaSyDvyloHqIVcizjDdwu0CvTwVvZTGDmrrbE"
g = API(client_id=g_client_id, client_secret=g_client_secret, api_key=g_key)
## Functions
def get_id_from_url(url: str):
if "youtu.be" in url:
result = urlparse(url)
v_id = result.path.replace('/', '')
return v_id
else:
result = urlparse(url)
v_id = result.query.replace('v=', '')
return v_id
def set_up_youtube(self):
self.yt = API(client_id='', client_secret='', api_key=self.api_key)
import json
from yaml import load, FullLoader
from youtube import API
from youtube_transcript_api import YouTubeTranscriptApi
with open("keys.yaml", "r") as keys:
config = load(keys, Loader=FullLoader)
api = API(client_id=config['client_id'], client_secret=config["client_secret"], api_key=config["api_key"])
terms = [
"linear algebra",
"calculus",
"derivative",
"shakespear",
"python coding tutorial"
]
def get_transcript(video_id):
return YouTubeTranscriptApi.get_transcript(video_id, languages=["en"])
def write_transcript_to_file(name, transcript):
with open("transcripts/" + name + '.json', 'w') as writefile:
json.dump(transcript, writefile)
def get_search_ids(term, max_ids=10000):
ids = list()
token = None
while len(ids) < max_ids:
response = api.get(
'search',
q=term,