def copyTopicsToPlaylist(self):
topic_list = Topic.get_content_topics()
for topic in topic_list:
playlist = Playlist.all().filter("title =",
topic.standalone_title).get()
if playlist:
logging.info("Copying topic " + topic.standalone_title +
" to playlist.")
child_keys = topic.child_keys
vps = VideoPlaylist.all().filter(
"playlist =",
playlist).order("video_position").fetch(10000)
playlist_keys = []
for vp in vps:
try:
playlist_keys.append(vp.video.key())
except db.ReferencePropertyResolveError:
logging.info(
"Found reference to missing video in VideoPlaylist!"
)
topic_keys = [
key for key in topic.child_keys if key.kind() == "Video"
]
if playlist_keys == topic_keys:
logging.info(
"Child keys identical. No changes will be made.")
else:
# logging.info("PLAYLIST: " + repr([str(key) for key in playlist_keys]))
# logging.info("TOPIC: " + repr([str(key) for key in topic_keys]))
logging.info("Deleting old VideoPlaylists...")
db.delete(vps)
vps = []
for i, child_key in enumerate(topic.child_keys):
if child_key.kind() == "Video":
vps.append(
VideoPlaylist(video=child_key,
playlist=playlist,
video_position=i,
live_association=True))
logging.info("Creating new VideoPlaylists...")
db.put(vps)
else:
logging.info("Playlist matching topic " +
topic.standalone_title + " not found.")
def updateVideoAndPlaylistData(self):
yt_service = gdata.youtube.service.YouTubeService()
# Now that we run these queries from the App Engine servers, we need to
# explicitly specify our developer_key to avoid being lumped together w/ rest of GAE and
# throttled by YouTube's "Too many request" quota
yt_service.developer_key = "AI39si6ctKTnSR_Vx7o7GpkpeSZAKa6xjbZz6WySzTvKVYRDAO7NHBVwofphk82oP-OSUwIZd0pOJyNuWK8bbOlqzJc9OFozrQ"
yt_service.client_id = "n/a"
video_youtube_id_dict = Video.get_dict(Video.all(),
lambda video: video.youtube_id)
video_playlist_key_dict = VideoPlaylist.get_key_dict(
VideoPlaylist.all())
association_generation = int(
Setting.last_youtube_sync_generation_start())
logging.info("Fetching playlists")
playlist_start_index = 1
playlist_feed = yt_service.GetYouTubePlaylistFeed(
uri=
'http://gdata.youtube.com/feeds/api/users/KhanAcademyHebrew/playlists?start-index=%s&max-results=50'
% playlist_start_index)
while len(playlist_feed.entry) > 0:
for playlist in playlist_feed.entry:
logging.info("Playlist: %s", playlist.id.text)
playlist_id = playlist.id.text.replace(
'http://gdata.youtube.com/feeds/api/users/KhanAcademyHebrew/playlists/',
'')
playlist_uri = playlist.id.text.replace(
'users/KhanAcademyHebrew/', '')
query = Playlist.all()
query.filter('youtube_id =', playlist_id)
playlist_data = query.get()
if not playlist_data:
playlist_data = Playlist(youtube_id=playlist_id)
logging.info('Creating Playlist: %s', playlist.title.text)
playlist_data.url = playlist_uri
playlist_data.title = playlist.title.text.decode("utf-8")
playlist_data.description = playlist.description.text.decode(
"utf-8")
playlist_data.tags = []
for category in playlist.category:
if "tags.cat" in category.scheme:
playlist_data.tags.append(category.term)
playlist_data.put()
for i in range(0, 10):
start_index = i * 50 + 1
video_feed = yt_service.GetYouTubePlaylistVideoFeed(
uri=playlist_uri + '?start-index=' + str(start_index) +
'&max-results=50')
video_data_list = []
if len(video_feed.entry) <= 0:
# No more videos in playlist
break
for video in video_feed.entry:
if not video.media.player:
logging.warning(
"Could not parse video - skipping... (%s, %s)",
video, video.media)
continue
video_id = cgi.parse_qs(
urlparse(
video.media.player.url).query)['v'][0].decode(
'utf-8')
video_data = None
if video_youtube_id_dict.has_key(video_id):
video_data = video_youtube_id_dict[video_id]
logging.info(
'Found Video: %s (%s)',
video.media.title.text.decode('utf-8'),
video_id)
if not video_data:
video_data = Video(youtube_id=video_id)
logging.info(
'Creating Video: %s (%s)',
video.media.title.text.decode('utf-8'),
video_id)
video_data.title = video.media.title.text.decode(
'utf-8')
video_data.url = video.media.player.url.decode('utf-8')
video_data.duration = int(video.media.duration.seconds)
if video.statistics:
video_data.views = int(video.statistics.view_count)
if video.media.description.text is not None:
video_data.description = video.media.description.text.decode(
'utf-8')
else:
video_data.decription = ' '
if video.media.keywords.text:
video_data.keywords = video.media.keywords.text.decode(
'utf-8')
else:
video_data.keywords = ''
video_data.position = video.position
video_data_list.append(video_data)
db.put(video_data_list)
playlist_videos = []
for video_data in video_data_list:
playlist_video = None
if video_playlist_key_dict.has_key(
playlist_data.key()):
if video_playlist_key_dict[
playlist_data.key()].has_key(
video_data.key()):
playlist_video = video_playlist_key_dict[
playlist_data.key()][video_data.key()]
if not playlist_video:
playlist_video = VideoPlaylist(
playlist=playlist_data.key(),
video=video_data.key())
logging.info('Creating VideoPlaylist: %s, %s',
playlist_data.title, video_data.title)
else:
logging.info('Updating VideoPlaylist: %s, %s',
playlist_video.playlist.title,
playlist_video.video.title)
playlist_video.last_live_association_generation = association_generation
playlist_video.video_position = int(
video_data.position.text)
playlist_videos.append(playlist_video)
db.put(playlist_videos)
# Check next set of playlists
playlist_start_index += 50
playlist_feed = yt_service.GetYouTubePlaylistFeed(
uri=
'http://gdata.youtube.com/feeds/api/users/KhanAcademyHebrew/playlists?start-index=%s&max-results=50'
% playlist_start_index)
