Ashish/yt-dlp
1
1
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-23 10:37:18 -05:00
Code Issues 7 Projects Releases Wiki Activity

[youtube] fix: extract artist metadata from ytInitialData (#49)

Browse Source
This commit is contained in:
insaneracist 2020-10-29 14:37:06 -07:00
parent 4932ba4aec
commit 9322f1162d
1 changed files with 46 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

46
youtube_dlc/extractor/youtube.py
View File

@ -1406,6 +1406,44 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
return self._parse_json(
uppercase_escape(config), video_id, fatal=False)
def _get_music_metadata_from_yt_initial(self, yt_initial):
music_metadata = []
key_map = {
'Album': 'album',
'Artist': 'artist',
'Song': 'track'
}
contents = try_get(yt_initial, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'])
if type(contents) is list:
for content in contents:
music_track = {}
if type(content) is not dict:
continue
videoSecondaryInfoRenderer = try_get(content, lambda x: x['videoSecondaryInfoRenderer'])
if type(videoSecondaryInfoRenderer) is not dict:
continue
rows = try_get(videoSecondaryInfoRenderer, lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'])
if type(rows) is not list:
continue
for row in rows:
metadataRowRenderer = try_get(row, lambda x: x['metadataRowRenderer'])
if type(metadataRowRenderer) is not dict:
continue
key = try_get(metadataRowRenderer, lambda x: x['title']['simpleText'])
value = try_get(metadataRowRenderer, lambda x: x['contents'][0]['simpleText']) or \
try_get(metadataRowRenderer, lambda x: x['contents'][0]['runs'][0]['text'])
if type(key) is not str or type(value) is not str:
continue
if key in key_map:
if key_map[key] in music_track:
# we've started on a new track
music_metadata.append(music_track)
music_track = {}
music_track[key_map[key]] = value
if len(music_track.keys()):
music_metadata.append(music_track)
return music_metadata
def _get_automatic_captions(self, video_id, webpage):
"""We need the webpage for getting the captions url, pass it as an
argument to speed up the process."""
@ -2328,6 +2366,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
if release_year:
release_year = int(release_year)
yt_initial = self._get_yt_initial_data(video_id, video_webpage)
if yt_initial:
music_metadata = self._get_music_metadata_from_yt_initial(yt_initial)
if len(music_metadata):
album = music_metadata[0].get('album')
artist = music_metadata[0].get('artist')
track = music_metadata[0].get('track')
m_episode = re.search(
r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
video_webpage)